Regulating The Broadband Internet: Towards An Interdisciplinary Analysis

Internet Regulation 01/02/2006
Regulating the Broadband Internet:

Inter--Disciplinary Analysis
Towards an Inter
Christopher T. Marsden LL.B. (Hons), LL.M. (Hons)
Researcher, Centre for Socio-Legal Studies, Oxford U niversity;
Teaching fellow, U niversity of Essex School of Law
Senior Analyst, RAN D Europe, Cam bridge
1st February 2006
1
Table of Contents
Chapter 1: Broadband Internet: The Regulatory Challenge__________________________________ 4
1.1. Regulating the European Broadband Internet_________________________________________ 4
1.2. Organisation of the Argum ent................................................................................................. 10
1.3. Libertarian N on-regulation ..................................................................................................... 11
1.3. How to regulate the Internet_____________________________________________________ 12
1.4. Code as Law and Technological Determ inism : Lex Inform atica............................................... 14
1.5. The Internet as a Regulatory Object _______________________________________________ 18
1.6 Methodological N ote ............................................................................................................... 20
Chapter 2: Broadband Internet Structure and Regulatory Challenges _________________________ 24
2.1. Internet Regulation: Back to the Future? ___________________________________________ 24
2.2. W hom to Regulate on the Internet? _______________________________________________ 26
2.3. Jurisdictional Response to the Internet .................................................................................... 27
2.2. Internet Industry and the Role of ISPs..................................................................................... 30
2.4.1. V ertical Integration of ISPs with ICPs .................................................................................. 31
2.4.2. Liability for Harm ful and Potentially Illegal Content on the Internet ................................... 33
2.5. Regulated Self-Regulation and European Concepts of Self-Regulation ____________________ 36
2.6 From European to G lobal Internet Regulation Challenges........................................................ 42
Chapter 3: The Institutional Perspective: Beyond N ational Regulation________________________ 43
3.1. W hy Doesn’t the Internet Change Everything? ______________________________________ 43
3.2. G lobal Regulation and the Internet _______________________________________________ 46
3.3. Socio-Econom ic Analysis of Inform ation N etworks as G lobal Public G oods ________________ 49
3.4. Public goods _________________________________________________________________ 50
3.4.1. Public G oods and Private Actors .......................................................................................... 52
3.4.2. Public Choice and ‘Races to the Bottom ’.............................................................................. 54
3.4.3. G lobal Public G oods ............................................................................................................ 55
3.5. N ew Institutional Econom ics, Public G oods and the Com m ons _________________________ 62
Chapter 4: Telecom m unications Regulation – Towards Behavioural Regulation ________________ 65
4.1. Introduction: Institutional Failures in Com petition Policy______________________________ 65
4.2. Broadband Carriage: Infrastructure Deploym ent ..................................................................... 70
4.3. Com parative Broadband Penetration: Q uantitative Evidence of Com petition _______________ 72
4.4. European Behavioural Com petition Policy__________________________________________ 78
4.5. The N ew Experim ent: Telecom Regulation and Co-Regulation _________________________ 81
Chapter 5: Renewed Inter-G overnm ental Regulation of the Internet _________________________ 84
5.1. Introduction: Is Internet regulation broken?............................................................................ 84
5.2. Five W aves of Internet Regulation, and U nited States Leadership ________________________ 86
5.2.1. ICTs, Regulation, and G lobalization .................................................................................... 89
2
5.2.2. Researching ICTs and W hy History Matters: Path Dependency........................................... 92

5.3. Internet governance at an Inflexion Point ................................................................................ 93
5.3.1. Innovation, Politics and G overnance .................................................................................... 94
5.3.2. Three prelim inary conclusions:............................................................................................. 98
5.4. A Suggested Research Agenda ____________________________________________________ 99
5.5. Proposal to Create Internet Developm ent Scenarios and Futures ........................................... 100
5.5.1. Objective of an Internet Futures Study............................................................................... 103
5.6. Conclusion: Internet G overnance After Tunis ....................................................................... 106
Chapter 6: Co-Regulation of Internet Content _________________________________________ 111
6.1. W hat is co-regulation?........................................................................................................... 111
6.1.1. W hat can be learnt from existing studies in analysing co-regulation?................................... 112
6.2. Self-regulation of the electronic gam e industry______________________________________ 116
6.3. Mobile Internet Services and Protection of Minors from Adult Content __________________ 123
6.3.2. Self-Regulation in Practice in Mobile Content ................................................................... 126
6.4. Analysis of Adult and Illegal Content Regulation................................................................... 127
6.5. The U K Case Study: Mobile Content Code of Conduct ....................................................... 129
6.5.1. Six Differences Between Fixed and W ireless Internet Regulation ........................................ 133
6.5.2. Free Speech Only W ithin a W alled Proprietary G arden?.................................................... 136
6.6. Conclusions: A Long Future for Co-Regulation of Internet Content ..................................... 137
6.6.1. Analysis of Successful Co-Regulation in the Media............................................................. 138
Chapter 7: Standards and the Internet ________________________________________________ 141
7.1. Is the Internet broken? ________________________________________________________ 141
7.2. Applying Standards Theory to Internet Regulation __________________________________ 144
7.3. Standards Organisations and Their Analysis ________________________________________ 145
7.4. From G overnm ent-Diktat to Corporate-Led: The N ew Approach to Standards ____________ 147
7.5. ICT Interoperability and Open Standards ............................................................................. 150
7.5.1. Interoperability: V ideo Exam ple................................................................................ 152
7.5.2. The Role of Consortia and Fora _______________________________________________ 153
7.5.3. Open standards ......................................................................................................... 155
7.5.4. ‘Multistakeholderism ’: U ser Focus and Participation in Standards Consortia...................... 155
7.5.5. IPR Policies: Consensus related to processes and deliverables..................................... 156
7.6. G lobal standards _____________________________________________________________ 158
Chapter 8; Conclusion: Towards an Interdisciplinary Analysis of the Internet?_________________ 159
8.1. Regulating the Internet: Medium Law?____________________________________________ 159
8.2. ISP Actors: Self-Regulatory Resources and Internet Policy..................................................... 165
8.3. Medium Law and Its Regulatory Cost-Benefits...................................................................... 168
3
Chapter 1: Broadband Internet: The

T he Regulatory Challenge
“I have heard and read here and there, that Brussels intends to regulate the Internet, to
introduce new red tape. Frankly, this is nonsense! N ever ever has the Com m ission had
such a foolish idea! But let m e ask you som e questions: who in this room is in favour of
child-pornography on the new m edia? W ho stands for the freedom to spread incitem ent
to racial hatred on the new m edia? … It is the duty of the Com m ission to propose a
fram ework under which these shared European values are protected. But I have no
intention to ‘regulate the Internet’!”
European Com m issioner V ivian Reding, 2005 (em phasis in original)1
1.1. Regulating the European Broadband Internet
Of course the Internet has always been regulated2. As Reidenberg3 and later Lessig4 stated, the
environm ent of the Internet is itself a determ inant of its physical and virtual boundaries. The
software that m akes the Internet work is a pre-existing ‘law’ of the Internet, just as gravity and
other Laws of Motion regulate the hum ans and their artefacts who interact via the Internet. But
the use of legislative and jurist’s tools to regulate the Internet has largely been by m eans of
applying existing offline laws to the online environm ent: for jurisdiction, for crim inal obscenity,
for libel, and m ore prevalent, for copyright infringem ent5. January 2006 saw the announcem ent
of several publications and policies that reflect this broadband Internet crisis. First, academ ically,
the announcem ent of new books from a cyber-realist perspective on the extent to which national
1
Reding, V. (2005) Better regulation for Europe’s media industry: the Commission’s approach, Speech
05/532.
2
My earlier works have dealt extensively with these themes .See especially [2000](ed) Regulating the
Global Information Society Routledge; [2001] Cyberlaw and International Political Economy: Towards
Regulation of the Global Information Society L. Rev. M.S.U.-D.C.L. 355; [2001] The Start of End-to-
End: Internet Protocol Television 29 Intermedia 3; [2003] Video over IP, the Challenges of
Standardisation: Towards the Next-Generation Internet Chapter 8 in Eli Noam et al (eds) ‘Internet
Television’, Lawrence Erlbaum, Mahwah, New Jersey; [2004] Hyperglobalized Individuals: the
Internet, globalization, freedom and terrorism 6 Foresight 3 at 128-140; [2005] Co-Regulation in
European Media and Internet Sectors Multimedia und Recht 1 at 3-7; Cowie, C and Marsden, C. (1999)
Convergence: Navigating Through Digital Pay-TV Bottlenecks 1 Info 1 at 53-66; Marsden, C. & S.
Verhulst [1999](eds) Convergence in European Digital TV Regulation Blackstone: OUP
Marsden, Tambini and Leonardi [2006, April] Codifying Cyberspace: Self Regulation of Converging
Media, Routledge
3 See further Redeinberg, J. (2005) Technology and Internet Jurisdiction, 153 UNIV. OF PENN. L.
REV. 1951 http://ssrn.com/abstract=691501
4
Lessig, L. (1996) Reading the Constitution in Cyberspace 45 Emory Law Journal 869
5
See Reidenberg, Joel (2004) States And Internet Enforcement, 1 UNIV. OTTAWA L. & TECH. J. 1
Goldsmith, Jack (1998) Against Cyberanarchy, 65 U. CHI. L. REV. 1199
Bick, Jonathan D. (1998) Why Should The Internet Be Any Different?, 19 PACE L. REV. 41, 63
Goldsmith, Jack (1998) What Internet Gambling Legislation Teaches About Internet Regulation, 32
INT’L LAWY. 1115
4
legislation and im plem entation can affect the Internet, reflecting the six years of the Yahoo!cases,
China's censorship m odel and eight years of Microsoft's case with the European Com m ission6.
Second, both G oogle's decision to reject U S governm ent data requests yet launch a google.cn
Chinese search engine which blocks access to anti-governm ent sites, and the reflection on the
fracturing of Internet standards because of those censorship and security issues raised by Yahoo!,
G oogle, Microsoft and others7.
The Internet has proved a regular target for legislation and regulation8. Students of m edia history
are well aware of the m oral panic which new m edia induce, and pornography is a regular target,
from printed press and colour photography to cinem a to video recording to satellite television9.
The difference with the Internet is that governm ent regulation has only taken place in special
circum stances, with ‘regulated self-regulation’ – or co-regulation as it is increasingly known – the
norm . Price and V erhulst assert the lim its of both governm ent and private action in this sphere,
and assert the interdependence of both – there is little purity in self-regulation without at least a
lurking governm ent threat to intervene where m arket actors prove unable to agree. They draw on
regulatory theory and em pirical studies of advertising and newspaper regulation, dem onstrating
that in areas of speech, the Internet included, governm ent preference in liberal dem ocracies is for
self-regulation10. G overnm ents’ belief that their regulatory goals could be m et by ISP activities in
6
See Geist, M. (2006) New Yahoo Decision Raises Old Questions
http://news.bbc.co.uk/2/hi/technology/4641244.stm and
http://www.michaelgeist.ca/index.php?option=com_content&task=view&id=1083
Goldsmith Jack and Timothy Wu (2006) Digital Borders: National boundaries have survived in the
virtual world-and allowed national laws to exert control over the Internet at
http://www.legalaffairs.org/issues/January-February-2006/feature_goldsmith_janfeb06.msp
Murray, A. (2006 forthcoming) Regulatory Webs and Webs of Regulation: Regulating the Digital
Environment, Glasshouse/UCL Press.
Zittrain, J. (2006 forthcoming) The Future of the Internet-And How to Stop It
Zittrain, J. (2006) Without a Net - http://www.legalaffairs.org/issues/January-February-
2006/feature_zittrain_janfeb06.msp
7
See Wakefield, Jane (2006) Google faces China challenges, 25 January BBC News
http://news.bbc.co.uk/1/hi/technology/4647468.stm and Rhoads, Christopher (2006) Should the World
Make Room for Another Wide Web? Wall Street Journal 22 January page 1, at
http://www.courant.com/business/hc-wsjinternet.artjan22,0,4344057.story?coll=hc-headlines-business
8
See Reed, J. and Angel, I. (2003) Computer Law, 5th Ed, which is the best of the several textbooks in
the UK market.
9
See for example Hoffman-Reim (1996)
10
Price, Monroe and Verhulst, Stefaan (2000) ‘In search of the self: charting the course of self-
regulation on the Internet in a global environment’, Chapter 3 in Marsden, C. (ed). One must
acknowledge the strength of the democratic principle enunciated: see Viktor Mayer-Schonberger and
Teree E. Foster (1997) A Regulatory Web: Free Speech and the Global Information Infrastructure in
Kahin and Nesson (eds) Borders in Cyberspace; Ingrid Volkmer (1997) Universalism and
Particularism: The Problem of Cultural Sovereignty and Global Information Flow in Kahin and
Nesson (eds); Price, M. (1995) Television, The Public Sphere and National Identity Oxford: Oxford
University Press
5
self-regulation were exposed in a series of court cases in the period from 2001, especially the
Yahoo! Vs. France saga, which are discussed in m ore depth in Chapter 2.
This book does not exam ine crim inally illegal content, covered by the Council of Europe
Cybercrim e Convention which cam e into force on March 18 200411. A Protocol to the
Convention m ainly intended to cover hate speech (European law but vehem ently opposed by the
U S negotiators in the Convention) was signed in 2004 but has not yet gained the ratifications to
bring it into force – and will not in the U S. There is here a change of role, in that generally the
Council of Europe12 – a hum an rights body form ed in 1949 with the signature of the European
Convention on H um an Rights (ECHR)13 in 1950 before the European Econom ic Com m unity was
form ed in 1955 – issues ‘soft law’ instrum ents such as Recom m endations, or Conventions that
are generally supplanted by European Com m unity law. However, and despite the huge body of
‘soft law’ em anating from the European Com m ission, it is the hard law of the Convention which
has led the way.
‘Soft law’ has specific m eaning in European law as a non-binding legal instrum ent that is
followed as a m atter of inform al practice by Mem ber States, such as a Recom m endation. The
usage here and throughout the book is less specific, referring to any non-binding regulatory
direction issued by inter-governm ental organisation (IG O), governm ent agency or m inistry14. In
the field of Internet regulation, there is a m uch m ore relevant and broad field of legal declarations
and recom m endations than useful case law (with the exceptions of IPR and crim inal law cases,
not covered in this book) 15. The non-binding policy instrum ent is – as will be seen - a staple of
11
Council of Europe Convention on Cybercrime, CETS No.185, opened for signature Budapest 23
November 2001, see http://press.coe.int/cp/2004/135a(2004).htm
12
See http://conventions.coe.int/Treaty/Commun/ListeTraites.asp?CM=8&CL=ENG
13
See Convention for the Protection of Human Rights and Fundamental Freedoms CETS No.: 005
entered into force 3 September 1953, at
http://conventions.coe.int/Treaty/Commun/QueVoulezVous.asp?NT=005&CM=8&DF=04/10/2005&C
L=ENG
14
For the increasing use of soft law as a regulatory instrument, see Morth, U. ed (2004) Soft Law in
Governance and Regulation: An Interdisciplinary Analysis, Edward Elgar.
See further the Commission’s ‘soft law’ website:
http://europa.eu.int/comm/development/body/legislation/softlaw_en.htm
15
For a practitioner viewpoint see National Consumer Council (2001) Soft Law in the European Union,
at http://www.ncc.org.uk/europe/softlaw.pdf
The NCC references several European legal instruments that give rise to self-regulatory soft law
obligations:
Directive 97/7 on Distance Selling especially article 11.4
Directive 95/46/EC on Data Protection
Recommendation 98/257/CE on out-of-court settlements
Directive 92/59/EEC on General Product Safety
6
self-regulation backed by the threat – but not im plem entation – of specific legal authority to
im pose direct state regulation16.
I consider copyright and intellectual property only in the context of the possibilities that dem and-
led growth of broadband can be accelerated by reform of copyright law. I acknowledge that very
sim ilar institutional constraints block copyright reform , as those that have prevented broadband
access reform . The property rights regim es in both cases are very m uch those created in the
analogue past, adaptation to the digital networks and services taking place increm entally and
according to the path dependence of that past. These types of problem s with actors following
path dependent evolution in the face of apparently transform ative technology are by no m eans
unusual, being fam iliar to environm ental policy analysts, where sustainable growth has replaced
unfettered growth as the new policy paradigm . In the case of digital inform ation networks, the
near-zero cost of distribution and copying on broadband networks – overlaid on the existing
analogue infrastructure - raises a different paradigm than the (adm ittedly very low) costs of
distribution of analogue networks and services. Evidence from prior industrial transform ations
suggests change will be increm ental17.
My thesis is: that interdisciplinary approaches provide the necessary insight to analyze the
developm ent of Internet m arkets, policies and consum er dem and. It is not possible to analyze
from a narrowly legalistic basis, nor from an econom ic analysis of the law18. Addition of
technological insight is of course necessary but not a sufficient addition to the disciplinary and
theoretical skillset19. Though several authors achieve this m uch of a skillset, they do not typically
take the necessary additional contextual leap to consideration of the socio-political analysis of the
field. Only in adapting to, and taking cognisance of, all these disciplinary elem ents can one
16
For a treatment attempting to exhaustively deal with European and German case law, see Scheuer,
Sections 5-6 in Held and Scheuer (2006) Draft Report: Study on Co-Regulation Measures in the Media
Sector; Study for the European Commission, Directorate Information Society and Media Unit A1
Audiovisual and Media Policies, Hans Bredow Institute.
17
See Hirschman (1970) and more specifically but polemically Lessig (2004).
18
Neo-classical economics is a useful but partial tool of analysis of the problems of market structure
and access in this new environment. Cost modelling has proved useful in analyzing the various costs of
the new digital economy. However, attempts to use predominantly econometric tools in analyzing anti-
competitive conduct meet very substantial empirical and theoretical difficulties. Approaches from
game theory provide deeper insight into the behaviour of dominant and other market actors. Recent
applications include those of the European Commission in the WorldCom/Sprint merger veto of 2002.
See generally G. Madden (ed.), International Handbook of Telecommunications Economics V. III, E.
Elgar; Einhorn, Michael (2004) Media, Technology, and Copyright: Integrating Law and Economics,
Edward Elgar; Laffont J.J. and Tirole J. (2000), Competition in Telecommunications, MIT Press
Buiges, P. and P. Rey eds (2004) The Economics of Antitrust and Regulation in Telecommunications,
Edward Elgar.
19
On interdisciplinary analysis see Snyder, F. (1995) “Out on the Weekend”: Reflections on European
Union Law in Context in Wilson, G.P. Frontiers of Legal Scholarship, Wiley & Sons, at 128.
7
analyze the field of inquiry with any hope of successfully capturing the m eta-analysis or even
m acro-analysis. I do not claim a return to the nineteenth century tradition of political econom y,
even were that possible within the specialisations and quantitative m odelling techniques of the
various elem ents of the social sciences. I do claim that the insights from the various developm ents
in quantitative m odelling from each discipline can be brought to bear, for instance the m odelling
of individual choice and collective action from sociology and political science, that of
econom etrics from econom ics, the social psychology of consum ers, and the socio-econom ic
analysis of the law and the legislative process20.
It is perhaps obvious from the form er argum ent that a com bination of theoretical and em pirical
elem ents, qualitative and quantitative m ethods is necessary. In the field of research chosen, the
developm ent of a global network for inform ation sharing, there are several other fields necessary
to consider. One rather obvious case is international political econom y (IPE), a field of
international relations m uch neglected until the end of the Cold W ar and increasingly
influential21. Another is inform ation m anagem ent and m anagem ent science, though here I claim
no particular insight beyond that of a practitioner in the largest Internet com pany (2001-2).
The um brella for all these fields that is of interest in particular for the public lawyer is that of
regulation22. The interdisciplinary field of regulation encom passes two broad approaches:
quantitative and qualitative, or social and econom ic, in crude and generalised term s. The
attem pts to specialise in either field are extrem ely successful so long as one approach does not
im pinge on the other. W ere it does – as for instance with U niversal (access or) Service
Obligations in telecom s or broadcasting – the need for inter-m ethodological research is pressing.
These fields are brought together in the study of the Internet for m any reasons, not the least
being the seam less nature of the m edium . It is at once a public m eans of com m unication with
m any public goods qualities that require regulatory analysis, and a private m eans of inform ation
sharing that has trem endous im plications as a foundational technology for all sophisticated and
distributed econom ic activities in both private and public m arkets. It is a local m eans of sharing
20
See advances such as Stiglitz, Joseph E. (1985) Information and economic analysis: A perspective,
Economic Journal 95 (supplement), , pp21-41; Richardson, M. and Hadfield, G. (1999) The Second
Wave of Law and Economics, Federation Press, Sydney; Hayek, F. (1979) Law, Legislation and
Liberty. A New Statement of the Principles of Justice and Political Economy. London, Routledge.
21
See Strange (1998) Who are EU? Ambiguities in the Concept of Competitiveness, 36 Journal of
Common Market Studies 1 at 101-114, at 105 ; Strange, S. (1970) International Relations and
International Economics: A Case of Mutual Neglect, 46 International Affairs 2 at 304
22
See , Bishop, Kay and Mayer (1995) The Regulatory Challenge Oxford University Press; Baldwin,
R. and McCrudden, C. (1987) Regulation and Public Law; Baldwin, R. et al (1995)Regulation in
Question: The Growing Agenda Merck, Sharp and Dohme; Baldwin, R. et al (1998) Socio-Legal
Reader on Regulation, Oxford University Press.
8
of goods, services and narratives, and a global network for provision of internationally shared
goods. These com binations, of public and private, of local and global, m ake the Internet a
particularly hybridised form of regulatory object. In different fields, its problem atique is entirely
different23. In econom ics, typically it is exam ined as an enabler of econom ic change and growth.
In m anagem ent science, it is exam ined as an enabler of m arket displacem ent. In network
engineering, it is exam ined as a problem set for m ore efficient yet distributed inform ation
distribution. In law, it is exam ined as a sovereignty and jurisdictional problem , as well as
representative of a legitim acy problem shared with politics and international relations. In
sociology, it is regarded as form ing and accelerating the developm ent of non-territorial and non-
traditional social relations that contribute towards globalisation. In accounting and financial
econom ics, it is viewed in som e quarters as the im possible m echanism to which to assign
com ponent value and cost. W hat all these, and other, disciplinary perspectives share is their
assessm ent of the Internet as dynam ically changing in itself, and changing the relations of other
factors dependent upon, or enabled by, it.
Dynam ic hybrid factors of change are unwelcom e to m ost Cartesian worldviews and m odels. A
law based on nation states and jurisdictional com petence, a social m arket based on indirect
dem ocracy, ideological politics and industrial m arket structure, with dem ography based on the
heterosexual fam ily unit based close to the physical workplace and educational establishm ent
within that nation-state, is the basis for m ost of the social scientific consensus of the Cold W ar
period. U nfortunately that does not fit the globalised environm ent which feeds, and is fed by, the
Internet. That is not to say that the findings of social science have been overtaken: the Internet
does not ‘change everything’. But it does produce a greater need for m ore dynam ic, inter- and
trans-disciplinary understand of how m odels need to respond to the changes taking place.
Henderson explains:
Forecasters from m any disciplines: econom ics, technology assessm ent, gam e theory,
ecology, or chaos and com plex adaptive system m odels, now agree that equilibrium
m odels drawn from Cartesian-N ewtonian worldviews of a determ inistic, "clockwork"
universe no longer fit.24
23
See generally Carnoy, Martin, Manuel Castells, Stephen S. Cohen and Fernando Henrique Cardoso
(1993) The N ew G lobal Econom y in the Inform ation Age; Reflections on Our Changing W orld Macm illan
24
Henderson, Hazel (1998) Viewing "The New Economy" From Diverse Forecasting Perspectives, 30
Futures 4, PP. 267-275
9
1.2. Organisation of the Argum ent

The book is organised as follows: the first three chapters outline the challenges of the Internet as a
new inform ation m edium , the m odels adopted for Internet regulation, especially self-regulation,
and the challenges of the regulatory form s adopted for freedom of expression. The following four
chapters exam ine in depth case studies, explaining the m ethodological approach adopted. The
final chapter explain the significance of the results for further research. I begin by outlining the
basis for Internet self-regulation.
This and the second chapter explains the origins and developm ent of U nited States Internet
regulation, the industry structures and liabilities that have been enshrined in law, and the m anner
in which private enforcem ent of liability has developed in the governm ental vacuum that
developed in the late 1990s. It then analyses European U nion response to the prevailing Internet
self-regulatory paradigm of the U nited States, and the developm ent of what becam e known as co-
regulation: self-regulation with a legislative backstop or ‘lurking threat’. This lays the foundation
for consideration of a m ore nuanced approach in Chapter 3 to analysis of the Internet, drawing
on econom ics and other disciplines to provide analytical tools to aid the lawyer.
This treatm ent is prim arily European, but with substantial interest to other jurisdictions, notably
the U S. There have been no m onographs of which I am aware which exam ine the law and
regulation of the Internet, from a European regulatory perspective. There have been several
collections of essays, and various considerations from a m edia/content perspective, but no
m onographs25. In the U nited States, there has fam ously been Lessig. There are also a large
num ber of works that begin from public policy perspectives, again m assively dom inated by U S
treatm ents which are predom inantly political science based. The treatm ent of the response of
regulation and law to dynam ic technological change has, in the U K , focussed on the
environm ent, utilities, financial services and biotechnology, as well as m ore general concerns
under the broad survey of law, regulation and globalisation . W hereas a volum e of work has built
up around Internet regulation, these works tend to speak specialised technical regulatory
25
See variously Thierer, A. (2004) ed. Who Rules the Net? Cato Institute, Washington DC;; Noam, E.
ed, (2003) Video over the Internet Lawrence Erlbaum, New Jersey; Brown Allan ed (2004) Digital
Terrestrial Television in Europe Lawrence Erlbaum, New Jersey; Garzaniti, L. J. H. F. (2000)
Telecommunications, broadcasting, and the Internet: E.U. competition law and regulation. Sweet and
Maxwell; Reed and Angel (2003) Computer Law, 5th Ed, OUP; Marsden, C. (2000). Regulating the
Global Information Society. New York: Routledge; Neuman, R., McKnight, L., & Solomon, R. (1997).
The Gordian Knot: Political gridlock in the information highway. Cambridge, MA: MIT Press; Owen,
B. (1999). The internet challenge to TV. Cambridge, MA, Harvard University Press; Mueller, Milton
(2002) Ruling the Root: Internet Governance and the Taming of Cyberspace, MIT Press.
10
languages from a predom inantly U S perspective. It is som ewhat technical and does not address
user groups or social scientists (other than econom ists).
In the following section I consider the events that led the U S Internet com m unity to continue its
technologically-led libertarian path.
on--regulation
1.3. Libertarian N on
It is claim ed by som e social pioneers such as John Perry Barlow that the Internet is a global
phenom enon beyond nation-state control, in which unregulated any-to-any com m unications is
possible. Com paring ‘cyberspace’ to Outer Space (i.e. extra-terrestrial) activity when the m ajority
of EU citizens have accessed the Internet negates the dem ocratic im portance of preventing harm
occurring on the Internet. The preface to the ‘Internet Com m ons Treaty’ 2004 recognises this,
even while continuing to proclaim the non-governm ental libertarian ideal26:
“The Internet seem s to have lost the special status that led m ost governm ents (sic) to
‘hands off’ policies during the 90s”.
Barlow’s declaration in 1996 in Davos and his institutionalisation of a libertarian perspective on

Internet content by the foundation of the Electronic Freedom Foundation in 199027, were part of
the 1990s zeitgeist of ‘globaloney’ declarations of the end of the nation-state’s role. It is shared by
Ohm ae28 and Internet studies and globalisation studies had a shared im m aturity and totalizing
m entality which were both burst by the shocking return to nation state values after the
retrenchm ents resulting from recession in 2000-2 and the attacks on N ew York and W ashington
on 9/11/200129. N evertheless, the libertarian perspective continues in EFF contributions, and
m ore profoundly in Internet Society (ISOC) contributions. The latter is m ore striking: ISOC is a
global m em bership organisation with country branches in m any countries. It is form ally the host
of the various Internet standard setting for a, yet m akes policy pronouncem ents on behalf of the
26 See Treaty on Principles Governing the Activities of States in the Exploration, Development and
Use of The Internet Commons at http://www.internationalunity.org/ This is the 2004 version of John
Perry Barlow and the Electronic Freedom Foundation’s 1996 Declaration of the Independence of
Cyberspace, http://www.eff.org/~barlow/Declaration-Final.html
27
See http://www.eff.org/about/history.php The co-founder was John Gilmore. The EFF is less
libertarian than in its early days, with moderating board members including Lawrence Lessig, Pamela
Samuelson and David Farber; http://www.eff.org/about/board/
28
See Ohm ae, K enichi(1990) The Borderless W orld: Power and Strategy in the Interlinked Econom y,
Collins, London
29
The backlash began earlier, see Dorem us, P. N ., W illiam W . K eller, Louis W . Pauly, and Sim on Reich
(1999) The Myth Of The G lobal Corporation Princeton U niversity Press; W eiss, L. (1999) The Myth Of The
Powerless State: G overning The Econom y In A G lobal Era, Polity Press
11
global technical m em bership30. A fierce internal debate rages in ISOC, with country m em bers
from outside the U S challenging the legitim acy of policy pronouncem ents m ade by these U S-
based and libertarian-leaning spokespeople.
K evin W erbach m aintains that the Internet has clearly been regulated since its origins, and that a
‘founding m yth’ of non-regulation has no basis in fact31. He establishes that the three Com puter
Inquiries which created federal regulatory policy towards data networks are the basis for Internet
interconnection: the decision not to regulate is a regulatory decision! W erbach’s insight is
irrefutable in the regulation of Internet interconnection. The ‘founding m yth’ m ust therefore be
viewed from the unique perspective of First Am endm ent law, and the free speech jurisprudence
of the U nited States. According to this, and its difference from the free speech pertaining in the
rest of the world, is found the uniqueness of Internet regulation.
A them e running throughout this book is the fundam ental question: does technical architecture
prevent public policy goals in regulation in a technologically determ inist or philosophically
deliberate fashion? Can control be re-exerted over Internet content, without causing m ore harm
in dam age to innovation than public good in reasserting dem ocratic control and m ajoritarian
values?
The law does play a role in the global Internet.
1.3. How to regulate the Internet

The censorship of the Internet for spam and anti-virus purposes can be undertaken for political
or com m ercial purposes. The idea of the Internet as anti-censorship is grounded in libertarian
m yth, not techno-political reality. Just as the G utenberg press, radio transm itters, the facsim ile
m achine and the photocopier could be used to extend both propaganda and free speech, so the
Internet does the sam e. The types of censorship undertaken overtly by dictatorial regim es,
fam ously China, and by dem ocratic regim es, m ost extensively via the Echelon em ail and Internet
detection system , are extensions of national security concerns that predate the Internet and are
ubiquitous in telephone, broadcast, postal and other com m unications system s32.
W hat is apparent is that these public policy and public law realm s are m ore system ic regulators of
the Internet than private law actions. To take exam ples, the placing of viruses on others’
30
Its three key actors are all closely associated with US policy: Fred Baker of Cisco and formerly chair
of the IETF; Mike Nelson of IBM and formerly advisor to Al Gore when Vice-President of the United
States 1992-2000; Vint Cerf of Google and formerly of corrupt bankrupt largest ISP WorldCom-MCI.
31
Werbach (2005)Federal Computer Commission North Carolina Law Review, at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=763905
32
See Spar (2001) and Standage (1999).
12
com puters has been illegal for m any years in m ost national laws33. The sending of spam is
contrary to European law in the 2002 Privacy in Electronic Com m unications Directive34. The
successful claim in civil law by N igel Roberts on 28 Decem ber 2005 in connection with this law
will not end spam 35. Other exam ples include the CAN -Spam Act in the U S, passed in 2003 and
claim ed by the Federal Trade Com m ission as effective in reducing unsolicited electronic m ail36.
W hile law enforcem ent and civil actions by individuals and corporations (notably Microsoft
Corporation) will have som e deterrent effect, spam will continue to exist so long as any one
jurisdiction perm its it and a critical m ass of Internet users do not filter its arrival. In the sam e
way, copying of m edia files will continue between peers on the Internet until a com pletely
effective technical solution is deployed across the network. Individual actions are in these cases
not directly curtailed by the threat of undirected legal action.
Its crisis has been analyzed in two dim ensions, and these dim ensions are opposed. From the
libertarian techno-econom ic perspective, the Internet is still too inefficient. It is too slow and too
one-speed, too ‘dem ocratic’ and too obsessed with collective action problem s. From the
paternalist socio-political-dem ocratic perspective, it is too anarchic and too open, too
unaccountable to nation states and too open to unchecked behaviours in speech and sexuality. So
is it too slow changing or too fast?
The answer is a m ixture of both, which raises a series of increasingly interesting questions for
Internet lawyers. Those concerned with the regulation of the Internet and its place in the law
have a new set of problem s to confront. The concern regarding the place of the Internet and
national sovereignty in the past 3 years has focussed on the W orld Sum m it on the Inform ation
Society (W SIS) and the reform ation of the Internet Corporation for Assigned N am es and
N um bers (ICAN N ). The latter37 has retreated from earlier com m itm ents to: a dem ocratic control
via direct elections by registered Internet users after the election of 2002; and the rem oval of
supervision by the U S Departm ent of Com m erce via its N ational Telecom s and Inform ation
33
For instance the 1990 Computer Misuse Act in the UK
34
Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the
processing of personal data and the protection of privacy in the electronic communications OJ L 201,
31.7.2002.
35
A small claims court uncontested case resulted in payment by direct marketing company Media
Logistics UK of £300. See Lewis, Paul (2005) Court verdict hailed as spam stopper, 28 December, The
Guardian, at http://business.guardian.co.uk/story/0,16781,1674429,00.html
36
Federal Trade Commission (2005) Effectiveness and Enforcement of the CAN-SPAM Act: A
Federal Trade Commission Report to Congress , December at
http://www.ftc.gov/reports/canspam05/051220canspamrpt.pdf
37
ICANN runs the ‘telephone numbers’ for the Internet and oversees the 246 ccTLDs Country Code
Top level Dom ains such as .de and .uk
13
Adm inistration (N TIA). In 2005, ICAN N rem oved four ccTLDs from their owners and
transferred them without appeal to governm ent control, with obvious political ram ifications in
K azakhstan and Iraq38. Further the N TIA declared its four principles of Internet control in June
2005. In response, the U nited N ations sum m it process retreated from attem pts to rem ove U S
control, instead instituting a continued ‘travelling circus’ of Internet ‘governance’ sum m its under
the aegis of the U N Secretary G eneral’s e-Envoy Markus K um m er. The next such is in Athens,
G reece, in autum n 200639. This future developm ent is discussed in m ore detail in later chapters,
as is co-regulation and the standards ‘revolution’, before which this chapter sketches out the
current argum ent.
1.4. Code as Law and Technological Determ inism : Lex Inform atica 40
The Internet is built on software code and hardware engineering. The Internet has of course gone
through m any m aturing phases since its anarcho-libertarian and vehem ently public sphere (or
‘Com m ons’) origins up until the early 1990s. A new type of radical free m arket libertarianism
grew up from then, consistent in its espousal of free speech but m ore concerned with corporate
and com m ercial freedom : avoiding governm ent econom ic and access, as opposed to social,
regulation. Lessig views software as the ‘code of cyberspace’, not in a narrow engineering sense,
but m ore widely, in the civil law sense as the constitution of cyberspace41. He sees regulation
developing from law, econom ics, norm ative values and architecture. The last such is – for
Internet regulation - the software code of the Internet Protocol and other software.
38
See coverage: The Register (2006) 2 January.
39
Author’s interview with Kummer at Department of Trade seminar and afterwards, London 17
January 2006
40
Several legal authors have discussed the technical engineering of the Internet and its effect on law
and policy, notably Reidenberg, J. (1993) Rules of the Road for Global Electronic Highways: Merging
the Trade and Technical Paradigms, 6 Harvard Journal of Law and Technology 287
41
Lessig (1999) at Chapter 7: What Things Regulate
14
Figure: The Intersecting Elem ents in Lessig’s Model of Regulation
Law
Social Norms
Architecture and Values
including Software
Code
Market Incentives and

Economics
Lessig’s form ula, which calls for m ultidisciplinary exam ination of the Internet from social sciences
and engineering, is from an analysis shared with m any Internet experts from engineering (e.g.
Abbate42, Bar43 and Cleevely44), law (e.g. Reidenberg and Benkler45), econom ics (e.g. David46 and
N oam ) and sociology and politics (e.g. Dutton47 and Drake48). These com m entators draw
different conclusions, but all exam ine the Internet from a range of interdisciplinary perspectives.
N orm s are here defined as a rule supported by a pattern of inform al sanctions, following
Ellickson49, who observes three faults in individual actor analysis: neglects of socialization, of
socially enforced norm s, and of the hum an quest for status. Lessig aim s to explicitly m odel the
42
Kahin, Brian and Abbate, Janet eds. (1995) Standards Policy for Information Infrastructure,
Cambridge MA: MIT Press
43
Bar, Francois, Cohen S., Cowhey P., De Long B., Kleeman M., Zysman J. (2000) Access and
Innovation Policy for the Third-Generation Internet, 24 Telecoms Policy at 489-518
44
David Cleevely (2000) Standards: Laying the Foundations,London: UK Department of Culture
Media and Sport
45
Benkler, Yochai, (1998a) ‘Communications infrastructure regulation and the distribution of control
over content’ Telecommunications Policy, vol.22, no. 3, pp. 183-196, at:
http://www.benkler.org/PolTech.pdf (accessed 16.03.05)
Benkler, Yochai (1998b) ‘Overcoming agoraphobia: building the commons of the digitally networked
environment’, Harvard Journal of Law and Technology, vol. 11, pp. 287-400, at:
http://www.law.nyu.edu/benklery/agoraphobia.pdf (accessed 16.03.05)
46
David, P. and Shurmer, M. (1996) Formal standards setting for global communications and
information services: towards an institutional regime transformation? 20 Telecommunications Policy
10 789-816
47
See Dutton W.H. ed (1997) Information and Communication Technologies: Visions and Realities,
Oxford University Press; Dutton W.H. ed (1998) Society on the Line: The Information and
Communication Revolution, Oxford University Press
48
See Drake, William J. (2000) ‘The rise and decline of the international telecommunications regime’,
Chapter 4 in Marsden C. (ed) Regulating the Global Information Society, Routledge.
49
See Ellickson, R.C. (1998) Law and Economics Discovers Social Norms, XXVII Journal of Legal
Studies Part II at 537-552
15
three Ellickson constraints, in addition to m arket constraints. He term s socialization

“architecture”, viewing society in these term s. His version of regulation is broader than
intentional policy action50, including ‘architectural’ constraints, for instance geographical
distance51. Lessig aim s not to dim inish law in relation to the three exogenous constraints, but to
exam ine its effects on those restraints. That law is a clum sy blunt instrum ent is itself relatively
well founded, and an attem pt to reassert the role of law by Lessig is a statem ent of the
interdependence between the four constraints described. Legal effects can thus be both direct and
indirect, in that it can regulate the regulatee (e.g. ISP), or regulate the other three constraints –
while those constraints act upon each other and the ISP 52. He adm its that the theoretical
application shares the universalist dilem m a: “The regulation of this school is totalizing. It is the
effort to m ake culture serve power.”53
V erhulst considers Lessig’s four forces of regulation in term s of their overall im pact on
‘regulability’ of the Internet. It is particularly interesting to note that the prior design of the
Internet regulates behaviours between users without needing any intervention by the
conventional apparatuses of social control – the m arket, social forces, or political and legal
activity. This engineering solution, which applies to m uch of the basic protocols of the Internet,
raises challenges for policym akers’ presum ption that legitim acy stem s from the social and
political. If engineering can either protect users in a space, or otherwise prevent norm al social
controls from applying, is the presum ption of social control reversed, replaced by a
technologically determ ined environm ent in which policy m akers m ust establish harm before
intervening? The potency of this argum ent has lost m uch of its force in the ‘dem ocratising’
period of m ass Internet adoption, for at least three reasons:
• harm is seen to have occurred as online behaviour prom pted off-line crim e;
• technologically determ inist legitim acy claim s for such a pervasive m edium of
com m unication are increasingly difficult to m aintain in the face of public scepticism in
the role of experts in nanotechnology, biotechnology, environm ental science, and now
the Internet;
50
See Ogus, A. I. (1994) Regulation: Legal Form and Economic Theory,
51 Bentham, J. (1995) The Panopticon Writings,; Foucault, M. (1979) Discipline and Punishment: The Birth of the Prison.
52 Supra n.192 at 667.
53 Ibid at 691.
16
• governm ents have increasingly intervened in the Internet’s architecture to secure

surveillance and other tools to invade users’ anonym ity, puncturing the m yth of the
‘unregulable’ Internet, particularly in surveillance post-9/11.
Table: Adapting V erhulst’s Characterization of Lessig ‘Realm s of Social Order’
Architecture – Econom ics Social N orm s Law

‘Code’
Enforcem ent Designers Market Peer G roup Police courts
Agent
Prior or Post Prior – design During - process Prior/Post - Post - sanction
Facto reputational
Constraint
Constraint Physical Econom ic Opprobrium Sanction
Type
Basis of Structural design Production and Social/group Power
Interaction exchange
Basis of U ser Buyer/Seller G roup m em ber Citizen/Subject
Participation
Prim ary Protocols Corporation Fam ily State Public
Institutions Engineering Market Com m unity Inform ation
The incursion of the new m eans of distribution and transaction over the Internet via electronic
com m erce into the previously open public space of the Internet is the ‘com m erce’ which, to those
sym pathetic to free expression but sceptical of m onopolistic corporate control of the underlying
architecture of the Internet, such as Lessig, were taking over the ‘code’ in the late 1990s. The
architectures of cyberspace are causing such re-exam ination of regulation and legitim acy. If the
Internet is to create an inform ation society, from the early hacker and libertarian com m unities,
that will prove a global challenge. The alternative is to perm it techno-econom ic factors,
com m erce and architecture as code developers, to construct an inform ation econom y with
increasing dissonance from nation-state societies and polities.
There is thus a three-way conflict of values, between cyber-libertarians epitom ised by John Perry
Barlow, corporate free m arketers, and those espousing social responsibility on the part of Internet
social and com m ercial entrepreneurs, to m aintain the Internet’s dynam ism , freedom and vitality,
but control its wilder anarchic excesses. Often the debate is highlighted by sensational excesses,
such as the webcasts of celebrity births, or the tragic crim inal cases of crim es inspired by the
Internet, including recent convictions of killers in G erm any and England, whose fantasies for
17
cannibalism and necrophilia were expressed on the Internet but then acted out with tragic off-line
physical results54.
1.5. The Internet as a Regulatory Object
The Internet is a ‘networks of networks’ that connects users by sending packets of bits (digital
data) from any point on that network of networks to any other point. K ahn and Serf adopt the
broad Federal N etworking Council definition of the Internet55:
“Of particular note is that it defines the Internet as a global inform ation system , and
included in the definition, is not only the underlying com m unications technology, but
also higher-level protocols and end-user applications, the associated data structures and
the m eans by which the inform ation m ay be processed, m anifested, or otherwise used”.
"The Federal N etworking Council (FN C) agrees that the following language reflects our
definition of the term "Internet".
"Internet" refers to the globalinform ation system that --

(i) is logically linked together by a globally unique address space based on the Internet Protocol
(IP) or its subsequentextensions/follow -ons;
(ii) is able to support com m unications using the Transm ission ControlProtocol/Internet
Protocol(TCP/IP) suite or its subsequentextensions/follow -ons,and/or other IP-com patible
protocols;and
(iii) provides,uses or m akes accessible,either publicly or privately,high levelservices layered
on the com m unications and related infrastructure described herein."56
For the ‘official’ history of the Internet, see Leiner et al57. For m ost dom estic, consum er users, it is
a com bination of the W orld W ide W eb, a graphical interface that perm its hyperlink surfing
(‘clicking’ from one webpage to another) and electronic m ail. However, it is actually the m ost
prevalent and fastest growing inform ation sharing network ever devised. It grew from a sm all
scientific network into a network accessed by alm ost a billion people in less than 10 years (1994-
2003), only exceeded in rate of diffusion by m obile telephones.58
54 See Wearden, Graham (2004) UK Police Chief: Shut Down Abhorrent Websites, at http://news.zdnet.co.uk/internet/0,39020369,39147312,00.htm
55 Robert E. Kahn and Vinton G. Cerf (December, 1999) What Is The Internet (And What Makes It Work)
Internet Policy Institute, Washington D.C. at http://www.internetpolicy.org/briefing/12_99.html
56 October 24, 1995, Resolution of the U.S. Federal Networking Council in Kahn and Cerf, op cit.
57 Barry M. Leiner, Vinton G. Cerf, David D. Clark, Robert E. Kahn, Leonard Kleinrock, Daniel C. Lynch, Jon Postel, Larry G. Roberts, Stephen Wolff, "A
Brief History of the Internet," www.isoc.org/internet/history/brief.html
58
Numbers of Internet users have consistently been under-estimated in official statistics, and there is no
accurate estimate for the global total. See Minges, M. (2003) at http://www.itu.int/ITU-
D/ict/publications/wtdr_03/material/WTDR03.pdf explaining the World Telecommunications
18
The dynam ic developm ent of Inform ation and Com m unications Technologies (ICTs) has the
potential to lower inform ation, and hence transaction, costs. The key to the transform ative effect
of these productivity gains is that networks increase the productivity effect with each new
addition to the network (this is known as Metcalfe’s Law), thus creating an exponential
‘bandwaggoning’ growth in the adoption of ICTs.
Econom ic Process Com ponent Cost-Efficiency Effect

Moore’s Law Microprocessor Doubles every 18 m onths e.g. from 2G Hz to 4G Hz
Metcalfe’s Law N etwork Increases potential value of network by square of
num ber of nodes – any new user is both receiver
and sender of inform ation e.g. e-m ail
Disc Law59 Storage – hard disk Doubles storage cost-efficiency each year
Data Packet Data Com pression Increases: boosted by im proved codecs e.g. DivX,
Transfer RealPlayer, W indows Media
G ilder’s Law60 Transm ission Potential bandwidth increases three tim es faster
Equipm ent than m icroprocessor power – Moore’s Law x3 –
every 6 m onths
Fibre Law Transm ission Capacity doubles every 9 m onths
N etwork
All these laws of the network and device have ‘network effects’ on the others – high processing
speed (Moore’s Law) and storage (Disc Law) are needed in PCs to process and store the highly
com pressed (Com pression) data files sent via switchers (G ilder’s Law) and optical fibre (Fibre
Law) over the Internet (Metcalfe’s Law). Two ‘laws’, of term inals and networks, create the
dynam ism of the Internet. Together, they create the law of the m icrocosm ’61. Moore’s Law,
coined by G ordon Moore of Intel in the 1960s, estim ated future m icroprocessor efficiency
doubling every year, and then every two years in the following decade (som e aggregated this as
Development Report. There are no definitive totals of Internet users worldwide, unsurprising given the
numbers of email accounts, use of cybercafes, numbers online through a third party subscription (e.g.
work, school, library). Global Internet Statistics estimates 940 million online by end-2004:
http://www.glreach.com/globstats/index.php3 As GSM Association statistics cover subscribers, it is not
an exact comparison – prepay phones and the fact that Internet users are not measured by subscribers to
ISPs distorts the comparison. GSMA statistics claim 987.2million subscribers at end-January 2004,
with 100 million or more CDMA subscribers to wireless systems in the Americas:
http://www.gsmworld.com/news/statistics/index.shtml accessed 28 February 2004
59
See further, Seeley Brown, John (2002) The Social Life of Innovation in the Digital Age, 15 July
presentation at
http://www.ruschlikon.net/INTERNET/rschwebp.nsf/(ID)/6C5A73B4FEBA95A9C1256C13002820A2
/$FILE/JSB-speech-.pdf
60
See Gilder, G. (1993) Metcalfe’s Law and Legacy, Forbes ASAP 13 September at
http://www.seas.upenn.edu/~gaj1/metgg.html
61
An excellent source of information on the pioneers is the ‘Nobel prize of communications’, the
Marconi Foundation’s Marconi Prize: see
http://www.marconifoundation.org/pages/news_room/index.htm
19
every eighteen m onths). Metcalfe’s Law applies to networks, especially real networks of users, as
in telephony, railways, telegraphs, highways, telephones. It holds that an additional user on the
network exponentially increases its usage, as both new user’s outward com m unications with the
rest of the network, and those existing users with the new user, are increased. This com bination
of ultra-powerful ubiquitous com puting and even stronger network effects creates the dynam ic
for an extraordinary growth in ‘bandwidth’: the capacity of the Internet com m unity to
com m unicate.
1.6 Methodological N ote

The interdisciplinary approach is essentially qualitative, though it is inform ed by the quantitative
studies of others, in for instance m arket developm ent from a com petition policy perspective. The
literature sources are broadly drawn, both geographically – with m ore references to U S sources
than typically in English law treatm ents of the topic – and in disciplinary term s, with a bias
towards political econom y and com parative studies. The references to technical literature are kept
to a m inim um for a broad audience. Both the politics and the econom ics of com m unications are
m essy hybrids of state and m arket processes62. The regulation of this space m ust necessarily
therefore be contingent, dynam ic, flexible and trans-disciplinary63.
The m ain source for m uch of the new evidence and m aterial uncovered is elite interviews with
academ ic analysts and the prim ary actors involved in the processes of Internet regulation.
Forem ost am ongst these are three sets:
1. Market actors:
2. Regulatory actors:
3. Academ ic experts:
These groups cooperated particularly in the studies of Internet governance (Chapter 6), standards
and security (Chapter 7), telecom s (Chapter 4) and broadcasting (Chapter 5). For all these areas,
I am grateful to participants and particularly board m em bers of the Telecom s Policy Research
Conference (www.tprc.org) especially in the years 1997-2000, 2004-5. For Internet governance,
I am grateful to participants at the European Consortium on Political Research (www.ecpr.org)
2005, the U nited N ations (U N ) Inform ation and Com m unications Technologies (ICT) Berlin
m eeting of N ovem ber 2005, the W orld Sum m it on the Inform ation Society Decem ber 2003
G eneva, and various Oxford Internet Institute (www.oii.ox.ac.uk) events in 2004.
62 An excellent thorough examination of these issues is Braithwaite, J. and Drahos, P. (2000) Global Business Regulation, Cambridge University Press
63 Marsden, C. (2000) MM-S-PL 1999-12 Final: Pluralism in the Multi-channel Market: Suggestions for Regulatory Scrutiny at section 5, at
http://www.ijclp.org/4_2000/ijclp_webdoc_5_4_2000.html
20
For telecom s and broadcasting, the list is necessarily m uch longer and includes regulatory, m arket
and academ ic analysts in the period 1995-2005, including those at various events and locations
hosted by: the European Com m ission; Council of Europe; Organisation for Security and
Cooperation in Europe; British governm ent; CEPT (Com m ission European des Postes et
Telegraphes); Federal Com m unications Com m ission (FCC); N ational Telecom and Inform ation
Adm inistration (N TIA); International Telecom s U nion (ITU ); International Telecom s Society
(1998, 2003, 2004); governm ent of France; European Internet Coregulation N etwork (EICN );
Institute for Prospective Technology Studies, Seville (1999), and several com m ercial conference
organizations, notably Inform a IIR. In this tim e, I consulted for several organizations in the fields
of telecom s and broadcasting, notably the Council of Europe (1999); Shortm edia Ltd (2000-1);
the Independent Television Com m ission (2000-1); MCI W orldCom U K Ltd (20001-2); Re-
think! Consultants (2002-3); OSCE (2004); K internet (K orean ISP Association) 2004; Chinese
state telecom s regulators under the Ministry of Inform ation Industries (2004); RAN D Europe
(2005-date). I was Research Fellow or V isitor at several academ ic institutions: Harvard
Inform ation Infrastructure Project (1999-2000); ESRC Centre for the Study of G lobalisation
and Regionalisation, W arwick U niversity (2000-1); Centre for the Study of Managem ent under
Regulation, W arwick Business School (2001-2); U niversity of Southern California (2003);
Centre for Socio-Legal Studies, Oxford U niversity (2004-6); Oxford Internet Institute (2004);
U niversity of Essex School of Law (2004-6); Cam bridge-MIT Institute Com m unications
Research N etwork (www.com m unicationsresearch.net). To all m y colleagues and interviewees in
this period, I offer m y sincere thanks. The holistic range of experiences thus offered is what
confirm ed and broadened m y understanding of institutional barriers to radical technological
change, and the regulatory devices thus used, for public good and public choice purposes.
Finally, I m ust acknowledge two extraordinary groups of experts whom I have worked with in the
period since 2001. The Harvard-Swiss Re Rueschlikon Conference held each June since 2001 is a
four-day m eeting of experts chosen each year to discuss for four days an inform ation policy issue.
The Trans-Atlantic Dialogue on Broadcasting and the Inform ation Society (TADOBATIS) is
sim ilarly held each year in June, with a three-day agenda to discuss these issues from the
broadcasting perspective. The participants at these events have generously given of their tim e to a
m uch m ore junior and callow colleague, and I have found these discussions invaluable in testing
m y hypotheses with som e of the founders and m ost influential policym akers and m arket actors in
the Inform ation Society.
21
The com bination of a rigorous academ ic analysis with this inform ed expert interview style
am ounts to a m odified Delphi exercise (see Chapter 6), in which experts’ opinions are sought,
m odified and consensus reached on their interpretation of path dependence and future possible
scenarios for developm ent. The Delphi technique of long-range forecasting separately elicits and
refines the opinions of a group of advisers without contact am ong them , and calculating a
statistical "group response." The original D elphi project was pioneered by RAN D in the U S in the
1950s to enable long-range forecasting,64 and continues to be widely used.
The procedure was designed to overcom e the disadvantages com m on to com m ittees and sm all
groups. The experts reply to written questionnaires, elite interviews or an online com puter,
receive statistical feedback through form al lines of com m unication, and resubm it their estim ates.
During the process, opinions do converge; where answers can be checked against reality, it is
found that the m edian response tends to m ove in the direction of the true answer. Delphi
participation has im pacts on the opinions of the experts them selves. In addition to their
consensus view, they often change or deepen their thinking through the experience of testing
their ideas and fram eworks in this kind of 'neutral' (because it is focussed on a 'crosscutting'
question) environm ent. Thus, Delphi can produce com plem entary as well as convergent views.
This is im portant to a futures analysis that seeks to be robust, resilient and engaging to a range of
stakeholders – it also helps in the design of sensible policies capable of decentralised (or
'stovepiped'im plem entation.
This approach of course has resource im plications: it relies on very high levels of participation by
experts and is extrem ely tim e-consum ing. To conduct this exercise am ongst such a broad group
of participants in so m any fields has been an exercise that required both a decade of research and
a great deal of good fortune in gaining access not only to the experts, but to environm ents such as
Rueschlikon, TADOBATIS and TPRC where participants are genuinely interested in applying
their expertise towards finding new analyses of em erging problem s in inform ation policy. It
would be difficult to form alise such a process for even one of the substantive chapters in this
study. Students looking for suitable research design processes are therefore not encouraged to
attem pt to replicate this exercise on a large scale unless both tim e and financial resources are in
practical term s unlim ited. N evertheless it has produced a focus and insight that I subm it is of
unparalleled utility in analysing these policy problem s.
In the following chapter, I outline som e initial thoughts on how the Internet m ay be
characterized as a regulatory subject, exam ining institutional analysis in Chapter 3, before 4
64
Dalkey (1967) Delphi RAND P-3704, RAND Corporation, Santa Monica, CA.
22
analysing the current ‘state of the art’ in Internet content self-regulation in Chapters 5-7,
exam ining recent developm ents in broadband co-regulation in Europe, from which I draw
conclusions in Chapter 8. In Chapter 2, we consider the range of self-regulatory types and
techniques available, before going on in later chapters to assess their successes and failings. In
concluding chapters, we m ust return to the developm ent of the Internet, in order to draw som e
conclusions for self-regulation from its further developm ent. First, we explore in policy how
Internet regulation has developed at U S and European level.
23
Chapter 2: Broadband Internet Structure and Regulatory Challenges
“The Internet seem s to have lost the special status that led m ost governm ents (sic) to
‘hands off’ policies during the 90s”
Preface to International Cyberspace ‘Treaty’ 200465
2.1. Internet Regulation: Back to the Future?
W e saw in Chapter 1 that Reidenberg and Lessig have popularised the view that two types of
code regulate behaviour on the Internet: legal code and software code. There is a third type which
receives less attention - codes of conduct for interm ediaries, including ISPs, and the term s of use
for their end users, which deal with inappropriate behaviour on the Internet. It is this third type
am ongst 'code, code, code'that this chapter exam ines. In particular, it asks whether harm ful and
illegal content types can be regulated effectively by legal code or ISP conduct codes. If such codes
are ineffective, or require such radical intervention into freedom of expression that the end-to-end
principle enshrines, software code's role m ay facilitate regulation. I exam ine the third of three
separate content types that users rule inappropriate and that have attracted public policy
attention:
1. unsolicited com m ercial m essages (spam ),
2. m alicious code (spyware and viruses particularly), and
3. harm ful or inappropriate content (typically unsolicited adult m aterial, particularly for
children).
(The first two are illegal and therefore beyond the scope of this book). A regulatory response that
institutes rules at either the network or user level to prevent these content types is increasingly
urgent in broadband networks. Chapter 8 draws conclusions for regulatory policy based on a
study of European self-regulatory types. W ithout a unified constitutional and com petition law
approach, a regulatory m odel that is less suitable for both network openness and freedom of
expression will em erge.
65 See Treaty on Principles Governing the Activities of States in the Exploration, Development and
Use of The Internet Commons at http://www.internationalunity.org/ This is the 2004 version of John
Perry Barlow and the Electronic Freedom Foundation’s 1996 Declaration of the Independence of
Cyberspace, http://www.eff.org/~barlow/Declaration-Final.html
24
The creation of the Mosaic browser in April 1993 m ade consum er hom e use of the Internet
feasible for non-specialists66. In the following eleven years, m ore than a billion people have used
the Internet. Alm ost 200 m illion households have broadband access to the Internet in 2006, of
som e form , with m any m ore accessing broadband at school, college or work. These two step
changes to the Internet, from a research and business tool to a consum er activity, and from
narrowband to broadband67, have transform ed the Internet.
As Crandall and Singer state:
“narrowband ISPs were decim ated by the m igration to higher-speed broadband services
… this shift to broadband has forced dial-up service providers to consolidate and exit the
m arket unless they can find som e way of providing broadband services over som eone
else’s network.”68
By March 2004, the announcem ents that BSPs such as Tiscali were increasingly challenging
incum bents’ voice revenues by selling V OIP substantially increased the entry costs for ISPs
wishing to offer equivalent sophisticated BSP service. AOL, form erly the world’s largest ISP,
announced its withdrawal from the U .S. broadband m arket. The m arketing costs and potential
revenue gains of entering a com bined data-voice-content m arket im m ediately, perm anently and
significantly raised the costs of providing Internet connectivity. From a universe of thousands of
European ISPs in 2000, no m ore than 20 large BSPs can be expected to dom inate the European
m arket by end-2006. The regulatory im plications are profound69.
G iven that alm ost all the m ajor developm ent innovations, both scientific and com m ercial, on the
Internet have been N orth Am erican, unsurprisingly the initial governance m odel has m irrored
that developm ent. The Internet since that early beginning has becom e a global phenom enon.
There are now m ore Europeans than N orth Am ericans on the Internet since 2002, and that
difference will grow by 2007, though m ore Asians than Europe-U .S. com bined will be
connected.
66
Mosaic was launched in April 1993. See
http://www.ncsa.uiuc.edu/Divisions/PublicAffairs/MosaicHistory/impact.html
67
From 150Kb/s in some Western European entry levels to 100Mb/s where Fibre To The Home
(FTTH) has been deployed, in Japan, South Korea, Sweden and Italy especially. See Chapter 4.
68
See Crandall, Robert W. and H.J. Singer (2005) Are Vertically Integrated DSL Providers Squeezing
Unaffiliated ISPs (and Should We Care)? at http://ssrn.com/abstract=710601at 2
69
See for instance these contributions: Wu, Tim (2003) Network Neutrality & Broadband
Discrimination, 2 J. Telecom. & High Tech 141; Yu, Christopher (2004) Would Mandating Network
Broadband Neutrality Help Or Hurt Competition? A Comment On The End-To-End Debate, J.
Telecom. & High Tech; Farrell Joseph & Philip J. Wesier (2003) Modularity, Vertical Integration, and
Open Access Policies: Toward a Convergence of Antitrust and Regulation in the Internet Age,
Department Of Economics UCB Paper E02-325; Woroch, Glen A. (2002) Open Access Rules and The
Broadband Race, 3 L. REV. M.S.U.–D.C.L
25
2.2. W hom to Regulate on the Internet?
Concerns regarding inappropriate and potentially harm ful content on the Internet are as old as
the public Internet itself, but began to surface in public policy debate in about 1994, when V int
Cerf classified three types of regulation: technical constraints, legal constraints and m oral
constraints70. He stated that: ‘In reality, all of these tools are com m only applied to channel
behavioural choices.’ He explains that it was university and research institute conditions of use,
including Codes of Conduct, that regulated online behaviour from the Internet’s invention. After
the privatisation of the Internet in 1989-94, Codes of Conduct, inherited from the public service
past, continued to be the default approach. Legacy inheritance thus conditioned private Internet
use. Cerf em phasised the need for m otivated self-regulation: ‘guidelines for conduct have to be
constructed and m otivated in part on the basis of self-interest’.
In a 1995 response to the threat of U S legislation against illegal and harm ful m aterial on the
Internet71, the W orld W ide W eb Consortium began to develop the Platform for Internet
Content Selection (PICS)72, the basis of filtering that was incorporated into browser software and
used to classify web pages by the m ajor ISP-portals in the U nited States – and by default in
70
Cerf, Vint (14 Aug 94) Guidelines For Conduct On And Use Of Internet Draft v0.1 at
http://www.isoc.org/internet/conduct/cerf-Aug-draft.shtml Though an incomplete draft it is fascinating
as an example of very early public policy making on Internet content
71
The Communications Decency Act, Title 47 U.S.C.A., 223(a) and (d), 1996 was introduced on 30
January 1995, passed by Congress in December 1995 and signed into law by President Clinton in
January 1996, before being substantially but not wholly declared unconstitutional by the Supreme
Court in ACLU v. Reno Supreme Court Case No. 96-511, 1997
72
Paul Resnick and James Miller (1996) PICS: Internet Access Controls Without Censorship,
Association for Computing Machinery vol. 39(10), pp. 87-93 at http://www.w3.org.PICS.iacwcv2.htm
26
Europe. The idea was sim ple: to engineer websites and user software to enable control of content
at the device – the end of the network – rather than by ISP or another interm ediary.
In addition to the adm inistrative sanctions, generally inform al, triggered by the original Codes of
Conduct of Internet users described by Cerf, the Internet also developed its own form of vigilante
to enforce self-generated norm s of behaviour. Sanctions included the spread of viruses which
incapacitate recipients’ PCs via e-m ail, hacking into governm ent or corporate sites, or
reputational dam age (consider eBay auction site user ratings of sellers). Clearly these lack
legitim acy. Form al dem ocratic decision-m aking for the global issues which Internet governance
raises are extrem ely im m ature, as Froom kin has dem onstrated73. Citizen dem ands for protection
and security creates a classic global public goods issue, which governm ents are now addressing (as
we will see in Chapter 3)74. Environm ental, labour and financial m arket analysts will find these
reflections unsurprising exam ples of both the lim itations of global governance and the rapid
m aturing and thus increasing com plexity of ‘civil society’.
2.3. Jurisdictional Response to the Internet
The Internet creates m ultiple com peting regulatory jurisdictions, both in European law75 and in
broader international com parisons, not least in broadcasting and e-com m erce regulation76. This
search for a m ore holistic explanation is approached from law by W eiler, in his study of the
European U nion constitution77 “with particular regard to its living political m atrix: the
interaction between norm s and norm -m aking, constitution and institutions, principles and
73
Unsurprisingly, ICANN falls far short of the inclusive participatory standard required. Froomkin,
Michael A. (2000) ‘Semi-private international rule-making: lessons learned from the WIPO domain
name process’, Chapter 11 in Marsden, C. (ed).
74
See Kaul, Inge, Isabelle Grunberg and Marc Stern (eds) (1999) Global Public Goods : International
Cooperation in the 21st Century, Oxford University Press, NY, NY.
75
For a theoretical discussion of policy convergence, accompanied by communications case studies,
see Levy, D.A.L. (1997) Regulating Digital Broadcasting in Europe: The Limits of Policy Convergence
20 West European Politics 4 at 24-42. Holznagel, Bernd (1998) European Audiovisual Conference -
Results from Working Group III in 1 International Journal of Communications Law and Policy 1 at
http://www.digital-law.net/IJCLP/final/current/ijclp_webdoc_9_1_1998.html;.
The more general role of the ECJ is examined by Weiler, J. (1991) The Transformation of Europe 100
Yale Law Journal 2405, (1993) Journey to an Unknown Destination: A Retrospective and Prospective
of the European Court of Justice in the Arena of Political Integration 31 Journal of Common Market
Studies 417 and A Quiet Revolution: The ECJ and its Interlocutors (1994) 17 Comparative Political
Studies 510
76
See Spar D. L. (1996) Ruling Commerce in the Networld, 2 J. Computer-Mediated Communication 1
77
Weiler, J.H. H. (1999) The Constitution of Europe, Cambridge University Press, at 15. His chapter 2,
from which this methodological note is taken, is updated from his classic 1991 essay The
Transformation of Europe 100 Yale Law Journal 2403.
27
practice”. The decision whether, when, how, why and whom to regulate is political, however
bounded by technological factors.78 Ayres and Braithwaite state79:
Practical people who are concerned with outcom es seek to understand the intricacies of
interplays between state regulation and private orderings...
Internet users are increasingly using the Internet for electronic com m erce, inform ation and
entertainm ent, education, form ing friendships and virtual com m unities. However, trust in the
m edium is affected negatively by software viruses, unsolicited electronic m ail (spam ), real and
perceived inappropriate or harm ful contact and content from a social and cultural context, threat
of prosecution for copyright and even crim inal activity online. Invasion of privacy – the alternate
to freedom of expression – is a constant source of discom fort to Internet users. The European
Com m ission expresses som e of the pitfalls of Internet surfing com pared with television or radio:
W hereas in traditional broadcasting (analog or digital) the individual broadcaster is easily

identifiable, it is difficult and som etim es im possible to identify the source of content on
the Internet. Access to harm ful and illegal content is easy and can even occur without
intent. In addition, the volum e of inform ation in the Internet is m assive in com parison
to broadcasting.80
It has long been m aintained that Internet service is not analogous to broadcasting, but print
publishing. In this view, the lack of control of the network by any one com pany, relatively low
entry and distribution costs to publishers and authors, and pluralism of supply m eans that there is
com petition for users. In this case, the only ‘control’ is on the distribution of the m aterial to the
final user, as in a newsagent’s kiosk, and the editorial choice of the end-user81. In the tangible case
of print, newsagents m ight refuse to sell pornography to m inors, but in the case of the Internet,
that choice is m ore difficult due to the technology. Below, I show the controls used for analogue
free-to-air and digital pay-TV , the printed press and basic voice telecom m unications.
78
Florini, A.M. (2000) Who does what? Collective action and the changing nature of authority,
Chapter 1, pp15-31 in Higgott et al (2000), , especially at 20-27.
79
Ayres, Ian and John Braithwaite (1992) Responsive Regulation: Transcending the Deregulation
Debate O.U.P.at 3.
80 See Second Evaluation Report From The Commission To The Council And The European Parliament on the application of Council Recommendation of 24
September 1998 concerning the protection of minors and human dignity COM(2003) 776 final of 12 December at
http://europa.eu.int/comm/avpolicy/legis/reports/com2003_776final_en.pdf at p6.
81 As indicated in the case of previous use of ‘gatekeepers by Mann and Belzley:
Mann Ronald J. & Seth R. Belzley (2005) The Promise Of Internet Intermediary Liability
Kraakman, Reinier H. (1986) Gatekeepers: The Anatomy Of A Third-Party Enforcement Strategy, 2 J.L. Econ. & Org. 53
Kraakman, Reinier H. (1984) Corporate Liability Strategies And The Costs Of Legal Controls, 93 Yale L.J. 857
Zittrain, Jonathan (2003) Internet Points Of Control, 44 B.C. L. Rev. 653
28
Table: V ertical layer control of the m edia industries82
Control Analogue ‘free-to-air’ Printed Press Telecom s Digital pay-TV

Stage? TV
Content Monopsony of terrestrial N o – low
Peer to Peer Yes – prem ium
creation and broadcasters – self- entry barriers
except prem ium sports rights and
production regulation supervised by
content self- m ovies
state agency regulated by
IARN
Bundling Com m and and control N o – low N ot applicable Yes – scale
content into – licensed by state entry barriers econom ies &
‘portals’ or agency long term
channels exclusivity create
dom inance
Packaging Com m and and control N o – low N ot applicable Yes – control
m ultiple – broadcasters allocated entry barriers over interactive
software into channel by agency services by
m ultim edia platform owner
Transm ission Com m and and control N o – Com m on Yes – access
and delivery83 – transm ission m ultiple carriage regulated but
controlled by agency distributors, only light touch
post com m on regim e
carriage
N avigational N o – TV rem ote control N o –1000s of N ot applicable – Yes – 500
control of only m edia; open interface channels and
user subscription Electronic
Program m e
G uide
U ser N o – Tax on TV N o N ot applicable – Yes – PIN
Controls: Households; content open interface num bers for
Decoder or im posed by evening prem ium and
software ‘watershed’ adult content
browser
In addition to the regulatory m odels seen above, there are other purer self-regulatory m odels in
broadcast advertising, film and video classification, and com puter gam es rating, that rely on
rating the content and using inform ation regulation to inform the end-user that content m ay be
unsuitable for m inors. However, trust in the m edium of the Internet appears to be weak, and
knowledge of ‘trust m arks’ – logos that indicate that a site has been certified by a self-regulatory
82Adapted from Cowie, C. and Marsden (1999) Convergence: Navigating Through Digital Pay-TV Bottlenecks 1 Info 1 at 55. Flynn offers SIXTEEN separate
bottlenecks in the transmission chain: Flynn, B. (1999) Opening the Box: Issues in Digital Gatekeeping, in Montreux Symposium ’99 of 10-15 June,
Symposium Records pp698-706 at p699.
83 Note the environmental restraints which apply to all distributional technologies – both terrestrial and satellite receiving apparatus is unsightly and may be
banned in historically preserved areas, for instance parts of London and Amsterdam. Cable installation creates different and temporary environmental
disturbance. Local environmental policy may ban one or more of these technologies, substituting a monopoly, which is generally in SMATV – multichannel
provision to a single apartment building via cable. This situation pertains in many cities, for instance Stockholm.
29
body such as ICRA – is very poor in Europe84. EU citizens’ ignorance about to whom to report
potentially harm ful content is particularly profound: 57% do not know whom to contact, and
only 13% are aware of a self-regulatory solution of ISP or hotline (with N etherlands citizens
highest at 26% ).
Figure: Eurobarom eter figures for 2003 from sam ple of 17,000 citizens
2.2. Internet Industry and the Role of ISPs
N oam has shown that consolidation in the Internet industry increased in the U .S. from about
1996, though m ost sectors rem ain com petitive85. He exam ines eight sub-sectors: Internet
backbones; Internet service providers (ISPs); broadband providers; portals; browser software;
search engines; m edia-player software; and Internet Protocol (IP) telephony. He explains that:
com m on elem ents are high econom ies of scale (scalability) based on the high fixed costs
and low m arginal costs, and the way they are often com plem ented on the dem and side
by network effects (which econom ists call ‘positive externalities’).
Representing the value chain diagram m atically:
• ICP (Internet Content Producer) encom passes portal (though often integrated into ISP
functions), search engine, and IP telephony;
• N etwork which would be located prior to ICP, encom passes broadband providers and
backbone providers, such as U U N et;
84 Only 10% of EU citizens in 2003 were aware of trustmarks, including only 8% of otherwise ‘cyber-savvy’ Swedes and 15% of risk-averse Germans Source:
Eurobarometer (2004).
85
Noam, Eli (2003) Oxford Internet Institute Issue Brief No.1 The Internet: Still Wide Open and
Competitive? (August 2003). He defines the Internet sector as: ‘the core industries that provide
instrumentalities and infrastructure components underlying the Internet’s basic functioning’ (at 2).
30
• ISPs provide the actual connectivity to the end-user, and the two m ost com m on
browsers, N etscape and Internet Explorer, are owned by two large ISPs, AOL and MSN
respectively86;
• The two largest m edia player com panies are integrated into conglom erates with an ISP
(W indows Media) and an ICP (Real N etworks).
ICP ISP ISP U ser Public access Device Control

Code of Term s of (schools,
Program m e codes Conduct service libraries) filters,
Editorial & Ethical Acceptable Com puter awareness,
Privacy
G uidelines U se Misuse codes zoning
Policy
Obviously public access, through work, governm ent institution, cybercafe or school, and the
device itself, are not included in N oam ’s list; but the filtering software that end-users and these
interm ediaries rely on is integrated into such software as search engines, m edia players, portals
and especially browser software. Filtering software is now com pulsory in libraries in the U S87 and
schools in France88 – where the state can control public access to harm ful content, it does so.
2.4.1. V ertical Integration of ISP s with ICPs
ISPs have integrated with content services and access suppliers. Most large ISPs provide a default
hom e page ‘portal’, with news, features, search facility. The largest ISPs are subsidiaries of either
access providers or software com panies such as Microsoft.
Access Content Access Peer to Peer Browser

Com m on carrier’ M SN BC M icrosoft
Instant Software
Broadband local AOL Tim e N etw ork
Messenger Explorer
network W arner V oice AOL Firefox
Backbone peering Music over IPN etscape N etscape
Movies Com puserve
Search Engines N avigator
Often, ISPs are joint venture partners with content or access providers, such as BT Yahoo!(U K )
or Yahoo! Softbank (Japan), or AOL Deutschland (form erly a Bertelsm ann joint venture). Both
Microsoft and AOL are also content providers, own search engines, have Instant Messenger
services. All European access providers and U S cable com panies provide proprietary Internet
86 The terms Internet Access Provider and ISP (Internet Service Provider) are often used interchangeably, though some people consider IAPs to be a subset of
ISPs. Whereas IAPs offer only Internet access, ISPs may provide additional services, such as leased lines and Web development. In contrast to both IAPs and
ISPs, Internet Content Providers provide their own proprietary content, often in addition to Internet access. See Yen. A. (2000) Internet Service Provider
Liability for Subscriber Copyright Infringement, Enterprise Liability and the First Amendment 88 Georgetown L.J. At
http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID236478_code000726304.pdf?abstractid=236478
87 Child’s Internet Protection Act 2003, building on the Communications Decency Act 1996.
88 Reuters (March 18 2004) Central Filter Against Web Hate for French Schools at http://www.reuters.com/newsArticle.jhtml;?storyID=4599516
31
services for their custom ers, m aking AOL and Microsoft unusual ISPs in that their content-
software focus has prevented their leveraging their ISP dom inance into access. In broadband
m arkets, those ISPs who also control access include T-Online in G erm any, W anadoo in France,
Telefonica in Spain, BT Yahoo!in U K . Though other ISPs can access the local loop at wholesale
prices, com petitors fear that the regulated access price leaves them disadvantaged.
The Table below illustrates the stages in the value chain of the Internet com pared to its telecom s
infrastructure supplier, and the printed press, to give a sense of the hybrid nature of its business
m odel. N ote that the areas controlled in the telecom s area – notably com m on carriage and
interconnection – are not regulated in either press or Internet. However, without that enforced
access onto telecom s networks, the ISPs could not serve end-custom ers. They do therefore rely on
regulation for their distribution.
Table: V ertical layer control potential of the Press, Telecom and Internet industries89
Control Stage? Printed Press Telecom s Internet

Content creation N o – low entry Peer to Peer N o – 100 billion web pages.
and production barriers except prem ium N ote im portance of peer-to-
content peer sharing of m usic and
video
Bundling N o – low entry N ot applicable Yes – econom ies of scale and
content into barriers long-term exclusivity result in
‘portals’ or dom inance
channels
Packaging N o – low entry N ot applicable Control over interactive
m ultiple software barriers services e.g. m edia players,
into m ultim edia voice over Internet Protocol
Transm ission N o – m ultiple Com m on Access regulated at wholesale
and delivery90 distributors, Post carriage level
com m on carriage
N avigational N o – newsagent N ot applicable Yes – virus scanning, spyware
control of user supplies thousands – open interface denial and search engines
of m edia;
89
For the layers model in connection with the Internet, see Lemley, Mark A. & R. Anthony Reese
(2004) Reducing Digital Copyright Infringement Without Restricting Innovation, 56 Stan. L. Rev.
1345 ; Solum, Lawrence B. And Minn Chung (2004) The Layers Principle: Internet Architecture And
The Law, 79 Notre Dame L. Rev. 815, 821; Lemley Mark A. &Lawrence Lessig (2001) The End Of
End-To-End: Preserving The Architecture Of The Internet In The Broadband Era, 48 UCLA L. Rev.
925
Werbach, Kevin (2002) A Layered Model For Internet Policy, 1 J. Telecomm. & High Tech. L. 37, 59
90
Note the environmental restraints which apply to all distributional technologies – both terrestrial and
satellite receiving apparatus is unsightly and may be banned in historically preserved areas, for instance
parts of London and Amsterdam. Cable installation creates different and temporary environmental
disturbance. Local environmental policy may ban one or more of these technologies, substituting a
monopoly, which is generally in SMATV – multichannel provision to a single apartment building via
cable. This situation pertains in many cities, for instance Stockholm.
32
subscription
available
Controls on N o N ot applicable Possible – PIN num bers for
V iewing: – open interface prem ium and adult content;
Decoder box or filtering; supervision by
software browser N etN anny, SurfPatrol
2.4.2. Liability for Harm ful and Potentially Illegal Content

Content on the Internet
Internet operation requires the passive reproduction and distribution of m aterial. ISPs
autom atically reproduce and distribute m aterial to subscriber requests. Content creators upload
to web pages by instructing the ISP’s com puter to store a copy of the uploaded m aterial. The
ISP’s com puter also m akes copies of the m aterial every tim e a com puter asks to view the
subscriber’s web page and sends those copies through the Internet. That file does not travel
directly to the user. Instead, it generally goes through other com puters hooked up to the Internet.
Each of these com puters m akes at least a partial copy of the relevant file. As Yen has described, “a
practically unlim ited scope of liability soon follows.”91 President Clinton’s 1995 Copyright
Taskforce supported such liability92. In order that these nodes on the network between content
provider and end-user are not all held strictly liable93 for the billions of web files they continually
copy in the act of transm ission, legislators in the U S and European U nion have held that only a
lim ited liability holds for these interm ediaries, typically ISPs94. In the U S, liability regim es have
differed according to speech-based and copyright-based liabilities95. The Com m unications
Decency Act of 199696 provides that “N o provider or user of an interactive com puter service shall
be treated as the publisher or speaker of any inform ation provided by another inform ation
91
Yen. A. (2000) Internet Service Provider Liability for Subscriber Copyright Infringement, Enterprise
Liability and the First Amendment 88 Georgetown L.J. At
http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID236478_code000726304.pdf?abstractid=236478
92
Working Group On Intellectual Property Rights, Information Infrastructure Task Force, Intellectual
Property And The National Information Infrastructure 1-6, 114-24 (1995)
93
Some legal commentators forcefully argued that strict liability should apply. See Trotter Hardy, The
Proper Legal Regime for “Cyberspace”, 55 U. PITT. L. REV. 993, 1042-46 (1994) (advocating strict
ISP liability); Kelly Tickle, Comment, The Vicarious Liability of Electronic Bulletin Board Operators
for the Copyright Infringement Occurring on Their Bulletin Boards, 80 IOWA L. REV. 391, 416
(1995) (favoring limited ISP liability).
94
See, for example, Elkin-Koren, Niva (1995) Copyright Law and Social Dialogue on the Information
Superhighway: The Case Against Copyright Liability of Bulletin Board Operators, 13 CARDOZO
ARTS & ENT. L.J. 345, 399-410, who argues opposing liability.
95
See Lichtman Doug & E. Posner (2004) Holding Internet Service Providers Accountable 3 U. Chi. L.
& Econ., Olin Working Paper No. 217, July 2004, Available At
http://Www.Ssrn.Com/Abstract=573502
Landes William & Douglas Lichtman (2003) Indirect Liability For Copyright Infringement: An
Economic Perspective, 16 Harv. J.L. & Tech. 395
96
Section 30, 47 U.S.C. § 230(c)(1) (Supp. II 1996).
33
content provider97. This language m ight shield ISPs from liability for subscriber copyright
infringem ent as well. However, Section 230(e)(2) specifically states, “N othing in this section shall
be construed to lim it or expand any law pertaining to intellectual property.” As Yen states:
the general philosophy m otivating these decisions—nam ely, that the liability against ISPs
for subscriber libel would result in undesirable censorship on the Internet—rem ains
vitally im portant in assessing the desirability of ISP liability.
Holznagel has indicated that U S courts have applied these ‘safe harbour’ provisions to widely
protect ISPs, even where [a] it was aware of unlawful hosted content; [b] if it had been notified of
this by a third party; [c] if it had paid for the data98. Frydm an and Rorive see courts as “in line
with the legislative intent…applied the im m unity provision in an extensive m anner”99.
In Europe, ‘safe harbour’ protection of ISPs from liability was only im plem ented on 17 January
2002, when the E-Com m erce Directive cam e into force. Article 12 protects the ISP where it
provides ‘m ere conduit’ with no knowledge of, nor editorial control over, content or receiver
(“does not initiate [or] select the receiver”). Benoit and Frydm an establish that it was based on
the 1997 G erm an Teleservices Act, though with “slightly m ore burden on the ISPs in com parison
with the form er G erm an statute”100. W here ISPs provide hosting services, under Article 14 they
are protected from liability, in two ways:
[a] the provider does not have actual knowledge of illegal activity or inform ation and, as
regards claim s for dam ages, is not aware of facts or circum stances from which the illegal
activity is apparent; or
[b] the provider, upon obtaining such knowledge or awareness, acts expeditiously to
rem ove or to disrupt access of the inform ation.
Like the proverbial three blind m onkeys, ISPs, IAPs, and web hosting services should ‘hear no
evil, see no evil, speak no evil’. As m ere ciphers for content, they are protected; should they
engage in any filtering of content they becom e liable. Thus m asterly inactivity except when
prom pted by law enforcem ent is the only – and econom ically m ost advantageous - policy open to
them . Frydm an and Rorive state “undoubtedly the Directive seeks to stim ulate coregulation”. It
97 Section 30, 47 U.S.C. § 230(c)(1) (Supp. II 1996).

98
Holznagel, B. (2000) Responsibility for Harmful and Illegal Content as Well as Free Speech on the
Internet in the United States of America and Germany, in C.Engel and H. Keller (eds) Governance of
Global Networks in Light of Differing Local Values, Nomos, Baden Baden
99
Frydman, B. and Rorive, I. (2002) Regulating Internet Content Through Intermediaries in Europe
and the USA, Zeitschrift fur Rechtssoziologie Bd.23/H1, July 2002, Lucius et Lucius
100 Ibid at 54.
34
does this by form ally perm itting national courts to over-ride the safe harbour in the case of actual
or suspected breach, of national law, including copyright law.
W hereas in the U S, the absolute speech protection of the First Am endm ent and procedural
concerns m ean that N otice and Take Down is counter-balanced by ‘put back’ procedures, in
Europe no such protection of free speech exists, where speech freedom is qualified by state rights.
In both jurisdictions, N otice and Take Down regim es cause Frydm an and Rorive to state that
“this m ay lead to politically correct or even econom ically correct unofficial standards that m ay
constitute an inform al but quite efficient m echanism for content-based private censorship”101. It
is clear that the econom ic incentive for ISPs is sim ply to rem ove any content notified, otherwise
do nothing to m onitor content, and let end-users, the police and courts, and ultim ately the ethics
of the content providers decide what is stored and sent over their access networks102. Frydm an
and Rorive state that:
Business operators should never be entrusted with … guidelines defining the lim its of the
right to free speech and offering procedural guarantees against censorship… which
belong to the very core of the hum an rights of a dem ocratic people103.
That is nevertheless the situation which ISP Codes of Conduct seek to self-regulate.
Could a stronger case be m ade to m ake ISPs responsible for a class of their content, where it
serves their com m ercial benefit? V icarious liability tests the ability to benefit and control:
[i] the right and ability to supervise and
[ii] a financial direct interest.
This tends to m ake ISPs choose not to m onitor even for law enforcem ent. The financial direct
benefit is interesting in view of the ‘killer application’ for broadband: does this include Peer-to-
Peer if the access charges received by the ISP is based on traffic i.e. adverts on portal or
bandwidth usage? ISPs arguably benefit from the existence of copyright infringem ent on the
Internet. Thousands of users desire Internet service precisely because it offers free access to
copyrighted m aterials. As Yen argues based on the Polygram case104, an ISP could m ake copyright
com pliance part of its system rules and then m onitor for violations. Thus, do ISPs have the sam e
degree of financial interest in infringem ent on the Internet as does a trade show in copyright
101 Ibid at 56.

102
Especially in the case of peer-to-peer networks, where blanket immunity is held by ISPs: Shapiro,
Andrew L. (1998) Digital Middlemen And The Architecture Of Electronic Commerce, 24 Ohio N.U. L.
Rev. 795; Netanel, Neil (2003) Impose A Noncommercial Use Levy To Allow Free Peer-To-Peer File
Sharing, 17 Harv. J.L. & Tech. 1, 4–5 ; Wu, Tim (2003) When Code Isn’t Law, 89 Va. L. Rev. 679
103 Ibid at 59.
104
Like the Polygram trade show operator Polygram Int’l Publ’g v. Nevada/TIG, Inc., 855 F. Supp.
1314, 1317-18 (D. Mass. 1994).Ibid at 19.
35
infringem ent? How about m obile networks and spam rem oval – do ISPs there exercise the right
and ability to m onitor content? Can they control inform ation on their network? V icarious
liability could therefore follow.
ISPs, evolving into m uch m ore com plex content, service and access providers, are the key link in
control of the end-user’s Internet experience, and the focus of this chapter. The final section of
the chapter considers their current developm ent in European law and policy.
2.5. Regulated Self

Self--Regulation and European Concepts of Self-
Self-Regulation
The European Com m ission recognizes that co-regulation can be used as a m eans to im plem ent
objectives set by Directives and has outlined in the W hite Paper on European G overnance108 a set
of conditions under which it will consider the use of co-regulation. Co-regulation is a pragm atic
response to the com m on perception that regulatory fram eworks m ust quickly adapt and
continually be optim ized to m aintain relevance and effectiveness to rapidly evolving m arkets.
Throughout the 1990s self-regulation was heavily advocated by the European Com m ission, m ost
notably under the Safer Internet Action Plan, but this approach seem s to have fallen out of favour
with recent m easures prom oting a m ore “co-regulatory” approach. W hereas “self”-regulation
im plies a degree of independence from direct state regulation, “co”-regulation im plies that the
state is involved in jointly developing rules and regulations. This m ight be thought m ore desirable
to the extent that research on self-regulation has shown that m any self-regulatory m odels in the
Internet industry lack proper procedures for oversight and enforcem ent, and am ount to little
m ore than declarations of good will. W e exam ine this further in Chapter 6.
The Council of Europe and the European Com m ission issued a series of reports and
recom m endations prom oting Internet self-regulation during the 1990s109. The intervening years
105 See Decision No 276/1999/EC OJ L 33, 6.2.1999, p.1 as amended by Decision No 1151/2003/ 16 June OJ L 162
106 See Council Directive 89/552/EEC of 3 October 1989 OJ [1989] L 298/23, as amended by Directive 97/36/EC of 30 June OJ [1997] L 202/60
107 Source: European Audiovisual Observatory, Yearbook 2003, volume 1
108 Com (2001) 428 Final, European Governance – A White Paper, at p21, see http://europa.eu.int/eurlex/en/com/cnc/2001/com2001_0428en01.pdf
109
A non-exhaustive list of EU reports and recommendations in the ‘early’ period of Internet policy
discussion would include:
White Paper on growth, competitiveness, and employment - The challenges and ways forward into the
21st century, COM(93) 700, Brussels, 5 December 1993;
Europe and the global information society, Recommendations of the Bangemann Group to the
European Council, 26 May 1994;
COM(94) 96 Green Paper on Strategy Options to Strengthen the European Programme Industries in the
context of the Audiovisual Policy of the European Union of 6 April 1994;
COM (94) 347, Europe' s way to the information society: An Action Plan 19 July 1994;
COM(96)395, Communication of the Commission to the Council, the European Parliament, the
Economic and Social Committee and the Committee of the Regions. on The Information Society: From
Corfu to Dublin -The new emerging priorities, 24 July 1996.
36
have seen the em ergence of a fertile ecology of rulem aking, regulatory com petition, alternative
dispute resolution and a com plex interaction between state, co- and self-regulatory practices in
the Internet. Im agery of state ‘invasion’ of a self-governing Internet is m isleading. Internet
developm ent does entail som e public policy issues that engage the institutions of dem ocratic
governance, and it is only within the legal fram ework for Internet liability that Internet users and
service providers can enjoy the relative freedom s to self-regulate that Barlow invokes. As Tam bini
et al state:
“em erging structures of m edia regulation develop through com petition in response to a
dem and for regulation which is in a constant negotiation between private and public
institutions … Self-regulation is the laboratory of law and regulation for the Internet..”110
As indicated by the positions adopted by U .S. policym akers, the choices of regulatory venue
adopted have been chosen in public-private partnership between the econom ic super-powers, the
U .S. and EU , and their high-technology private partners. The choices m ade have been largely
private, from the ICAN N dom ain nam es procedures to the self-regulatory standards setting
procedures which have overtaken the traditional U nited N ations agency, the ITU 111. The U K E-
Com m erce G reen Paper of 1999 states: “E-com m erce is essentially a global, rather than a
national, issue”112. It further states that initiatives have been taken by the OECD (Organization
for Econom ic Cooperation and Developm ent) in a cryptography conference of October 1998113,
by the U nited N ations Com m ission on International Trade Law (U N CITRAL) in a Model Law
Commission Communication on Europe at the Forefront of the Global Information Society: Rolling
Action Plan, COM(96)607 final, 27.11.1996;
Council Resolution on New political priorities regarding the information society, of 21 Nov. 96 OJ
C386, 12.12.96, p.1;
COM(97)157 A European Initiative in Electronic Commerce, April 1997;
Soete, L. et al (1997) Building the European Information society for us all, Final report of the High
Level Experts Group, April 1997.
110
Tambini, Leonardi and Marsden (2006) Codifying Cyberspace, Routledge, forthcoming August.
111
Formal legitimacy is offered by ITU, or WIPO in the case of ICANN, but there is no doubt that
private actors such as Motorola, IBM, Nokia, Ericsson, and others, play a critical role as non-state
actors. Drake describes the formation of the ITU as a classic government-sponsored functionalist
problem-solving activity, and its latter history as hijacked by the U.S. government and its clients,
whether carriers or manufacturers, leaving no doubt of the fact that non-state actors are intimately
involved in the formerly state-controlled process of rule and standard-setting for the Internet and
telephony. Drake, William J. (2000) ‘The rise and decline of the international telecommunications
regime’, Chapter 4 in Marsden, C. (ed).
112
Department of Trade and Industry (5 March 1999) Building Confidence in Electronic Commerce: A
Consultation Document at 6, ‘International Context’.
113
See Ypsilanti, D. (1999) A borderless world: The OECD Ottawa Ministerial Conference and
initiatives in electronic commerce in 1 Info 1 at pp23-34. For a critical perspective, see Love, J. (1999)
Democracy, privatization and the governance of cyberspace: An alternative view of the OECD meeting
on electronic commerce in 1 Info 1 at pp15-22.
37
on Electronic Com m erce114, and in other EU initiatives115. The U nited States governm ent took
the view that the challenges of global e-com m erce over the Internet should be m et through
private, and avowedly econom ically rational, m echanism s of self-regulation. The sustainability of
this m odel in view of its lack of consum er/citizen representation is a central dilem m a in the
future developm ent of the Internet, e-com m erce and the Inform ation Society. Functionalist and
technologist concerns regarding security, encryption and dom ain nam e allocation becom e
increasingly difficult to separate from individual rights concerns regarding privacy, freedom of
expression and public governance of the com m ons. The tension between new dynam ic processes
of self-regulation and existing institutions of m ore form alized and accountable governm ent-
m ediated regulation is central to an understanding of 21st century com m unications regulation.
As in m ost European responses to technologically led change in the socio-econom ic landscape

since 1945, the policy and m arket lead in technological convergence has been taken by U nited
States corporations, academ ics and governm ent. There is a defi am erican116 in European response
to convergence. Language also separates the European and U S debates on the m acro questions of
industrial policy: the future of both European delivery networks, and the content which they will
carry. W here the U S says ‘video’, Europe says ‘l’audiovisuel’, reflecting the traditional policy
distinction between the socio-cultural aspirations of European com m unications policy and the
caricatured U nited States profit m axim ization m arket-led philosophy. (Such distinctions are m ore
distinct in theory than in fact.) The m arkets for cinem a exhibition of film , pay-TV subscription
of them atic channels, and the m odel of regulation chosen for Internet and video-on-dem and, are
links in a value chain which m ust be exam ined in the round, rather than separately. This reflects a
different regulatory legacy. European telecom m unications m arkets were liberalized on 1 January
1998, belatedly following the 1984 exam ple set by the U nited K ingdom 117 and U nited States.
The Bangem ann Report of 1994118 shared som e free m arket characteristics with the U .S.
‘Inform ation Superhighway’ which Al G ore had claim ed prior to the 1992 Presidential cam paign,
114
See United Nations General Assembly Resolution 51/162 Of 16 December 1996 at
http://www.un.or.at/uncitral/en-index.htm
115
Additionally, work in this field is conducted through various self-regulatory bodies. See
Bangemann, M. (1999).
116
Servan-Schreiber, Jean (1967) Le Defi American translated in Penguin (1968) (The American
Challenge).
117
Telecommunications Act 1984 c.12 as amended by the Competition Act 1998 c.41.
118
Bangemann, M. et al (1994) Europe and the Global Information Society, The Report of the High
Level Group http://www.ispo.cec.be/infosoc/backg/bangeman.html; see further Bangemann, Martin
(1997) A New World Order for Global Telecommunications – The Need for an International Charter
Telecom Inter@ctive 97, International Telecommunications Union, Geneva, 8 September 1997
(http://www.ispo.cec/be/infosoc/promo/speech/geneva.html); Bangemann, Martin (1997) Europe and
38
but Europeanised the G ore/Clinton ‘Inform ation Superhighway’119 of early 1993 to the
‘Inform ation Society’ analogy. DG Inform ation Society120 was established to consider regulation
and prom otion of the Inform ation Society. Bangem ann121 appeared less convinced of the need for
com m and-and-control regulation of content: “a new approach is required as content becom es
network-independent and as control of content (and responsibility for its use) shifts from
governm ent to the individual.”
In European debate, the overall regulatory response was considered in a 1996 ‘convergence’
report122 com m issioned by the European Com m ission. This report form ed the backdrop for the
debates123 which led to the 1999 Com m ission proposals for a new fram ework for
com m unications regulation124, com ing into force in August 2003. The July 1997 Declaration at
the Bonn Ministerial Conference m ade plain the Council of Ministers’ desire to see end-user
filtering rather than interm ediary liability125. A further allusion to the international challenge of
Internet regulation, even in the case of crim inal law, is m ade:
65. Ministers recognise the specific challenges posed by the m isuse of G lobal Inform ation
N etworks. They consider, therefore, that international co-operation is essential in this
area. Ministers will actively encourage the reinforcem ent of police and judicial co-
operation, particularly in the area of technology training and m utual assistance, to
prevent and com bat illegal content and high technology crim e. They support the
establishm ent of international networks of hot lines.
66. Ministers welcom e the recent initiative of the OECD aim ing at a com parative study of
national legislations and an exchange of experiences on the issue of illegal content on the
Internet. Supporting a m ultilateral as well as a European approach, they consider that the
the Information Society: The Policy Response to Globalization and Convergence speech presented in
Venice, 18 September available at <http://www.ispo.cec.be/infosoc/promo/speech/venice.html;
Bangemann, M. (1999) Which Rules for the Online World? The European Union Contribution 1 Info 1
at 11-15
119
http://www.ntia.doc.gov/
120
http://www.ispo.cec.be/infosoc/
121
http://www.ispo.cec.be/infosoc/promo/speech/venice.html
122
KPMG (1996) Public Policy Issues Arising From Telecommunications and Audiovisual
Convergence, at http://www.ispo.cec.be/infosoc/promo/pubs/exesum.html
123
COM (1997) 623, Green Paper on the regulatory implications of convergence between the
telecommunications, media and information technology sectors: towards and Information Society
approach, Brussels, December, at: http://europa.eu.int/ISPO/convergencegp/97623en.doc
124
See COM (1999) 539 Review of the telecommunications regulatory framework– a new framework
for electronic communications infrastructures and associated services, Brussels, December, at:
http://europa.eu.int/ISPO/infosoc/telecompolicy/review99/review99en.pdf
125
Bonn Ministerial Declaration 8 July 1997 at
http://europa.eu.int/ISPO/bonn/Min_declaration/i_finalen.html
39
international dim ension is crucial in the building of trust and confidence in the G lobal
Inform ation N etworks.
Discussion during the renegotiation126 of the ‘Television W ithout Frontiers’ Directive127 led to a
Recom m endation in 1998 that continues to serve as the Com m ission’s policy towards content
regulation. Further Com m ission legal instrum ents including the 2000 E-Com m erce Directive128
has m aintained the co-regulatory approach to Internet regulation laid out in the 1998
Recom m endation129.130
The im portance of the Internet as the prim ary inform ation network for the creation of the
Inform ation Society has been recognized by the European U nion, from the Bangem ann Report
of 1994 to the eEurope+ Progress Report of 2004131. The com plete new fram ework for
com m unications regulation132 finally cam e into force in August 2003. The European
Com m ission has declared in its Com m unication of 25 February 2004133 that the pillars of
inform ation society developm ent are the regulatory fram ework, governm ent investm ent in
attractive content and support for research and innovation. In its eEurope Action Plan Review of
18 February 2004, the Com m ission134 further identified elim inating legal barriers to e-com m erce,
the copyright bottleneck for broadband content, and the increasing nuisance of spam as
126
See Whitehead, Phillip (1997) Draft Report on the Commission Green Paper on The Protection of
Minors and Human Dignity in Audiovisual and Information Services (COM [96] 0483 - C4-0621/96)
PE 221.804 of 24 April 1997, which formed the basis of European Parliament debate.
127
Directive 97/36/EC of the European Parliament and of the Council of 30 June 1997 amending
Council Directive 89/552/EEC on the coordination of certain provisions laid down by law, regulation
or administrative action in Member States concerning the pursuit of television broadcasting activities
OJ L 202, 30. 7. 1997.
128
Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal
aspects of information society services, in particular electronic commerce, in the Internal Market OJ L
178, 17.7.2000
129
See further Decision No 276/1999/EC of the European Parliament and of the Council of 25 January
1999 adopting a Multiannual Community Action Plan on promoting safer use of the Internet and new
online technologies by combating illegal and harmful content primarily in the area of the protection of
children and minors OJ L 33, 6.2.1999, p.1 as amended by Decision No 1151/2003/EC of the European
Parliament and of the Council of 16 June 2003 OJ L 162, 1.7.2003.
130 Communication COM(97) 487, 16.10.97; Council Recommendation 98/560/EC OJ L 270, 7.10.1998.
131
eEurope Progress Report, February 2004, analysing accession countries’ progress to eEurope
targets, at: http://www.emcis2004.hu/dokk/binary/30/17/3/eEurope__Final_Progress_Report.pdf
132 See COM(2002) 695 Eighth report on the implementation of the telecommunications regulatory
package Brussels, December, at:
http://europa.eu.int/information_society/topics/telecoms/implementation/annual_report/8threport/finalr
eport/annex2.pdf
133
IP/04/261 (2004 Towards a global partnership in the information society: way forward for the EU at
http://europa.eu.int/rapid/start/cgi/guesten.ksh?p_action.gettxt=gt&doc=IP/04/261|0|RAPID&lg=EN&
display=
134
COM(2004) 108 final ‘eEurope 2005 Mid-term Review’ at
http://europa.eu.int/information_society/eeurope/2005/doc/all_about/acte_en_version_finale.pdf of 18
February 2004.
40
priorities135. Regulation is declared to be a key elem ent in the developm ent of an inclusive and
safe Inform ation Society in which users m ake full use of the potential of the Internet.
Further Com m ission legal instrum ents including the E-Com m erce Directive of 2000 have
m aintained the co-regulatory approach to new m edia regulation laid out in the 1998
Recom m endation136. The European Com m ission states: “W hereas in traditional broadcasting
(analog or digital) the individual broadcaster is easily identifiable, it is difficult and som etim es
im possible to identify the source of content on the Internet. Access to harm ful and illegal content
is easy and can even occur without intent. In addition, the volum e of inform ation in the Internet
is m assive in com parison to broadcasting.”137 The European Com m ission has readdressed co-
regulation of the m edia in 2004:
The Recom m endation on the protection of m inors has a cross-
cross-m edia approach
and em phasizes the cross-
cross -border exchange of best practices and the
developm ent of coregulatory and self-
self -regulatory m echanism ss. (em phasis in
original).
It explains how best to achieve the regulatory goals:
A co-regulatory approach m ay be m ore flexible, adaptable and effective than
straight forward regulation and legislation… Co-regulation im plies however,
from the Com m ission’s point of view, an appropriate level of involvem ent by
the public authorities.
End-user tools such as filtering or the fam ous ‘V -chip’, im posing rules on children’s use of
com puter gam es and the W orld W ide W eb, and reporting inappropriate or illegal content to
hotlines established by Internet com panies have had only lim ited success.
There are m arkets for regional and/or national television, radio, newspapers, telecom s, satellite
and cable pay-TV , all recognized in case law138. The use of data-com pression and increases in
cost-effective bandwidth such as Digital Subscriber Lines (DSL) allows m ore and better point-to-
point delivery139. In this environm ent flexibility of regulatory fram eworks will be of param ount
135
COM(2004) 28 Unsolicited Commercial Communications or ‘spam’. More at:
http://europa.eu.int/information_society/topics/ecomm/highlights/current_spotlights/spam/index_en.ht
m
136 See further Directive 2002/58/EC OJ L 201, 31.7.2002; Directive 2000/31/EC OJ L 178, 17.7.2000.
137 See Second Evaluation Report COM (2003) 776 final of 12 December at
http://europa.eu.int/comm/avpolicy/legis/reports/com2003_776final_en.pdf at p6.
138 See Harcourt, A. (1998) Regulation of European Media Markets: Approaches of the European Court of Justice and the Commission’s Merger
Task Force, 9 Utilities Law Review 6 at 276-291; Larouche, P. (1998) EC Competition law and the convergence of the telecommunications and broadcasting
sectors 22 Telecommunications Policy 3
139 See Marsden, C. Video over IP: the challenges of standardization – towards the next generation Internet [2003] Chapter 8 in Noam, et al (eds)
Internet Television.
41
im portance to ensure that regulators m eet the current and future needs of the m arket place and
m aintain the confidence of consum ers through the protection of public interests. The dynam ic
developm ent of the sector and its regulatory landscape, lacks clarity as to the nature of the co-
regulatory/self-regulatory approaches taken, the areas where they are applied, consistency with
public interest objectives, im pact on the single m arket and ultim ately, effectiveness in achieving
regulatory objectives. W e consider co-regulation in depth in the case study on content co-
regulation in Chapter 6.
2.6 From European to G lobal Internet Regulation Challenges

In this chapter, we have exam ined the liability regim e for ISPs, the structure of the industry and
the theory of self and co-regulation on the Internet. W e have considered the intersection between
U S and European regulation. W e have not yet considered the over-arching problem in full: the
Internet is neither U S nor European but global. This introduces a broader conception of the
regulation of the m edium , because the public interest goal in such regulation is neither national,
nor regional, but global. The following chapter considers this global challenge to the political
econom y of Internet regulation.
42
Chapter 3: The Institutional Perspective: Beyond N ational Regulation
“It is no accident that econom ic m odels of the polity developed in the public choice literature
m ake the state into som ething like the m afia… Inform al constraints m atter. W e need to know
m uch m ore about culturally derived norm s of behaviour and how they interact with form al
rules to get better answers to such issues.”
Douglass N orth (1990)140
3.1. W hy Doesn’t the Internet Change Everything?

The claim s I m ade in the opening two chapters are well-known and broadly uncontroversial
am ongst specialists in the field. That regulation needs to adopt an increasingly trans-disciplinary
and non-state-centric approach to problem s that are global and dynam ic, as I stated in the first
chapter, is broadly agreed, whatever the m ethodological and cultural problem s it m ay present.
Succeeding chapters will consider telecom s infrastructure (Chapter 4), co-regulation of m obile
and gam es industries (Chapter 5), Internet governance debates at international levels (Chapter 6),
and industry standard setting (Chapter 7) to illustrate these claim s. It is at this point that I want
to go further, to introduce new perspectives and to deal with the thornier problem that arises: if
the Internet can be regulated, as we have shown, then how can governm ent best encourage that
regulation to produce innovation, without stifling the very dynam ism it is trying to nurture? The
problem can be stated thus: how can static state-centric institutions be adapted to regulate
without stym ieing innovation?
This chapter sets out the argum ent that any attem pt to assess the regulation of the Internet m ust
m ove beyond a ‘law and econom ics’ approach, however nuanced141. I state explicitly that it is the
institutional inertia of incum bent actors – both governm ents and m arkets - that is the greatest
barrier to adoption of the broadband Internet paradigm . That paradigm is m uch greater
inform ational efficiency and transparency, accom panied by greater ability to m anipulate
inform ation. N ote that this includes both greater freedom to speak and greater surveillance of
140
North, D. (1990) Institutions, Institutional Change and Economic Performance, Cambridge
University Press at 140.
141
See Trebilock, M. (1999) The Value and Limits of Law and Economics, pp12-29 at 29, in
Richardson supra; Lessig, L. (1998) The New Chicago School XXVII Journal of Legal Studies Part II
661-691; Posner, R.A. (1998) Social Norms, Social Meaning and Economic Analysis of Law: A
Comment XXVII Journal of Legal Studies Part II at 553-565; Deakin, S. (1999) Law versus
economics? Reflections on the Normative Foundations of Economic Activity, pp30-49 in Richardson,
ed (1999) at 46.
43
speech, and I am in this respect not m aking norm ative claim s for the benefits or otherwise of
broadband. My concern is change and the barriers in its way.
The institutions of which I speak are not sim ply ‘organizations’, whether political, m arket or
social, though all three are encom passed. The concept of institution goes further to consider
environm ent: the regulatory environm ent of the Internet encom passes m any institutions, som e of
great longevity, others m ore recent; som e ideological, others driven by financial or power
m otives142. An obvious exam ples is the broadcasting ‘establishm ent’, the collection of interlocking
regulatory stakeholders in the production, distribution, public taxation, advertising and
consum ption of television and radio broadcasts. Their interest in extending their analogue system
into the digital age is driven by consideration both of public goods and of public choice143. It is
too sim plistic to claim either a selfless m otive on behalf of the com m unity at large, or solely a
selfish m otive driven by inertia, sunk costs (both financial and intellectual) in the current system .
Douglass N orth’s com m ent that pure public choice theory m akes the State appear a m afia m akes
that point well. Equally the selflessness of the public goods argum ent too easily falls into a self-
serving prophecy that change in the broadcasting or telecom s ‘ecology’ top-slices revenues and
im poverishes the public’s access to universal reception or high quality output144. Coase goes
further, stating an institutional view of econom ics in forthright term s:
I find it difficult to ignore the role of stupidity in hum an affairs … I have often
wondered why econom ists, with these absurdities all around them , so easily adopt the
view that m en act rationally145.
W hile obviously agreeing with Coase, I therefore do not argue that a paradigm of com petition is
the panacea to overcom ing barriers to change in the Internet’s adoption. That is a statem ent
which requires further clarification:
1. I posit that com petition cannot of itself solve all Internet problem s, especially universal
access to high quality content, acknowledging that non-m arket form s of production,
distribution and consum ption are vigorous com plem ents to the m arket m echanism 146;
142
See generally Blom-Hansen, J. (1997) A ‘New Institutionalist’ Perspective on Policy Networks, 75
Public Administration 669-693, at 674-5; Hirschman A. (1970) Exit, Voice and Loyalty – Responses to
Decline in Firms, Organizations and States, Harvard University Press; Lindblom, C. (1977) Politics and
Markets: The World’s Political-Economic Systems Basic Books: NY
143
See Marsden and Verhulst (1999) supra, Hoffman-Reim (1996), Goldberg and Verhulst (1997) and
Levy (1999).
144
See Ungerer 2005.
145
Coase had been informed that the FCC (whose Commissioner asked him whether his proposed
auction was a joke) would not impose licence auctions because all nations monopolised spectrum, and
that only a Latin American dictator would permit the sale. Coase, R. (1998) Comment on Thomas W.
Hazlett, J. Law and Economics XLI pp 577-580 at 580
146
See for instance Benkler 1998a supra.
44
2. I argue that the ‘com petition law’ argum ent driven by neo-liberal interpretation of the
evidential burden in com petition law is in fact a ‘toothless tiger’ and that structural
change is needed in m arkets before any behavioural ‘tinkering at the m argins’ can achieve
m eaningful outcom es that accom plish consum er benefits in the long-term (see Chapter
4);
3. I argue that Internet policy is in fact driven by a greater or lesser recognition of the
essential problem , that of path dependence and incum bent dom ination of the essential
facilities and political decision-m aking that is necessary to change;
4. I acknowledge – though others have far greater com petence than I to discuss – the role of
state-sanctioned m onopoly in distorting m arkets over long periods, a problem that is
being slowly and partially eroded in telecom s but which inevitably results from grants of
patents, tradem arks and copyrights. In the latter area, the extension of copyright
accom panied by clum sy attem pts to technologically bind consum ers and innovators via
Digital Rights Managem ent system s has created a peculiar set of anti-com petitive
constraints on the Internet’s developm ent147;
5. I acknowledge the role of com m unity-led standards, which can have extrem ely powerful
effects on adoption of new developm ents. This is true in part for consum er adoption of
technologies and business m odels, though com panies with m arket power can often
persuade consum ers to adopt less effective technologies via m arketing and other tools. In
the supply rather than dem and side, standards are all-pervasive, as should be expected in
network system s. These standards can raise efficiencies via network effects, lock
technologies into place, or prevent adoption of new paradigm s (see Chapter 7). A
particular type of standard is that adopted in regulated self-regulation, or co-regulation,
in which industry form ally or inform ally adopts a standard for conduct which substitutes
for governm ent action. This is pervasive in content regulation on the Internet, and I
consider two lesser-known recent exam ples, com puter gam ing and m obile telephony co-
regulation, to assess its prospects (see Chapter 5).
Finally, I acknowledge that security and other technological developm ents can overtake as
distributed and networked a system as that of the Internet. In conclusion I consider the
technologist argum ent that the current Internet is insufficiently robust against anonym ous
threats, such as piracy, spam , Denial of Service attack and other problem s, that the collective
147
See Lessig 2001 supra.
45
action problem is presently insurm ountable. If the Internet itself can m o longer be com petitively
evolved into a m ore efficient network, then ossification beckons (concluding Chapter 8).
In analyzing all these factors, I of course repeat m y assertion from the first chapter that a trans-
disciplinary approach is necessary to m ake sense of the problem s that arise. Further, com parative
analysis of global and regional developm ents is essential in understanding the variations of
solutions that m ay be attem pted.
Before com ing to case study chapters regarding the institutional barriers to resolving som e of the
intractable difficulties in Internet regulation, I will first consider what the political econom y of
the Internet represents: as a global com m unications network, it is what is described as a global
public good. Its regulation needs to consider what the benefits and potential costs of this good
are: if the costs outweigh the benefits then its renationalisation, or re-partition m ight be sensible.
However, to consider this in its broadest sense requires a global and historical perspective on the
political econom y of the Internet. Once this is considered, the new theoretical perspectives on
how to resolve the problem s can be exam ined to conclude this chapter, and m ap the rest of the
book’s attem pt to grapple with the correct questions (if not answers).
3.2. G lobal Regulation and the Internet

The world has changed radically in the past generation, with the Cold W ar dem ise of
Com m unist central planning creating a unipolar geo-strategic landscape, the evolution of m any
corporations into m ore m ulti-state based m ultinational corporations (MN Cs)148, the
extraordinary transform ation of com m unications networks into seam less global webs of digital
inform ation exchange, and the resultant globalization of financial m arkets and consequent
interdependence of capital flows between liberal m arket econom ies149. G lobalization is not
new150. Cynical scholars enjoy debunking the m yths of neo-globalization. But som ething has
changed: technology151. The V ictorian Internet – the telegraph’s dots and dashes152 belonged to
148
Also known as multinational enterprises (MNEs). For our purposes, the two terms are more or less
interchangeable. See UNCTAD (1998) World Investment Report 1998: Trends and Determinants,
UNCTAD, Geneva.
149
See further Vaisse, Justin (2001) L’hyper-Puissance Au Défi De L’hyperterrorisme at
http://www.brookings.edu/views/articles/fellows/20011023vaisse.pdf and Devetak, Richard (2002)
Hyperpower America rewrites the global rules, at
http://www.theage.com.au/articles/2002/12/08/1038950270370.html
150
For a pessimistic scenario, even before September 11, see Samuel P. Huntington (1996) The Clash
of Civilizations and the Remaking of World Order, New York: Touchstone, 1996. For a contrary point
of view, see Francis Fukuyama (1992) The End of History and the Last Man, London: Penguin.
151
This is not to state a technologically determinist or ‘boosterist’ case: for a critical view of ICTs’
effect on society, see Robins, K. and Webster, F. (1999) Times of the Technoculture, Routledge,
London.
152
See Standage, Thomas (1999) The Victorian Internet, Phoenix, NY.
46
an Age when hierarchy and the elite interpreted technologies and com m unications for the m asses.
This has changed – the essential dem ocratization of the Internet m eans that sm all cells of
terrorists are em powered in m uch the sam e way as geeks playing war gam es. Lessig states that for
the first tim e, through cyberspace, we are confronted with the fact that “anyone could be a
m ultinational …routinely living in m ultiple non co-ordinating jurisdictions”.153
G overnm ent response to technology has also changed. The end of the nineteenth century was
responsible for revolutionary regulatory action, notable for trust-busting federal, state and
m unicipal action to regulate the increasingly powerful econom ic concerns that were using the
national networks of railways, electricity, telephone, telegraph to construct a private ordering
considered politically and econom ically disadvantageous to private citizens154. The end of the
twentieth century expanded the gam e internationally, such that the concern with national
networks, assuaged by (often heavy-handed) federal, state and m unicipal regulation, has been
supplem ented by a concern with the international. Replacing the industrial revolution’s concern
with tangible goods and services is the Inform ation Age’s concern with digital data flows,
products and services155. Com m unications is therefore of critical im portance, and previous
concerns with telephone and broadcast networks – the prim ary regulatory concern until as
recently as 1996 – have been expanded into a m uch m ore holistic, if only partially understood,
view of com m unications regulation as underpinning regulation of the entire inform ation
econom y, networks, content and all156. Trachtm an explains that:
Because the technology is so exhilarating, there is a tendency to claim that the changes
we do observe in sovereignty, the state, jurisdiction and law all are caused by cyberspace.
Surely the end of the cold war, the rise of econom ic interdependence, other technological
changes, and a host of historical causes m ust be considered alongside cyberspace … it is
153
Lessig, Lawrence (1999) Code, and Other Laws of Cyberspace, Basic Books, NY at 193-4.
154
In a vast literature, see for instance Freedman, J.O. (1978) Crisis and Legitimacy in the
Administrative Process: A Historical Perspective, Cambridge University Press.
155
See Bell, Daniel (1973) The Coming of Post-Industrial Society, revised paperback edition 1999.
This is acknowledged as the most influential of many 1970s studies of the emerging ‘information age’.
156
For papers exploring these issues in more depth than possible here, see Marsden, C. (ed) Regulating
the Global Information Society, Routledge, New York, 2000. See further the work of Brian Kahin at
the Harvard Information Infrastructure Project, especially Kahin, Brian and Keller, James H.
(eds)(1997) Coordinating the Internet, Cambridge MA: MIT Press; Kahin, B. and Nesson C.
eds.(1997) Borders in Cyberspace: Information Policy and the Global Information Infrastructure,
Cambridge MA: MIT Press; Kahin, Brian and Abbate, Janet eds. (1995) Standards Policy for
Information Infrastructure, Cambridge MA: MIT Press.
47
not the state which has died, but the long-m oribund theory of absolute territorial
sovereignty157.
He dism isses the idealist view of a “world federalist governm ent” explaining that:
the correct allocation of authority is dynam ic, com plex and contingent … contingent
sovereignty – the group of powers society decides to assign to the state at any given
m om ent and in any given circum stance.158
Trachtm an’s concern is also to explain cyberspace in term s of the relative lack of dynam ic
institutional reaction by regulators: “social institutions for allocation of jurisdiction have not
changed to reflect the technological changes brought by the rise of cyberspace.”159 Therefore:
one m ay view the rise of cyberspace as a phenom enon that accentuates the old problem s
to a point where it is worthwhile to devise a m ore substantial institutional solution.160
The greatest, and certainly to a W estphalian161 nation-state-centered universe m ost revolutionary,
challenge for regulation is the increasing co-operation between national, regional and
international networks of regulators, to regulate the G lobal Inform ation Society.
N ote I do not em ploy the term ‘governance’. This is a deliberate oversight by a regulatory lawyer.
First, governance has even m ore ‘soft’ im plications than ‘regulation’, itself a contested term in
legal literature162. Regulation has m any definitions, and I do not intend to draw on a particular
definition to the exclusion of others. However, I do want to explain that m y interest does not
encom pass explicitly ‘m ultistakeholderisation’163, covered by others in this context164, but is m ore
concerned with the political econom y of nation-states and globalising econom ic actors,
m ultinational corporations. That clearly affects m y view of regulation.
157
Trachtman, Joel P. (1998) Cyberspace, Sovereignty, Jurisdiction, and Modernism, Indiana Journal
of Global Studies Volume 5 Issue 2 Spring 1998 at 561, Indiana University School of Law –
Bloomington http://www.law.indiana.edu/glsj/vol5/no2/10tract.html visited 22 July 1999, at 1.
158
Ibid at 3.
159
Op cit at 13.
160
Op cit at 6.
161
The ‘Westphalian’ system, named for the Treaty of Westphalia which established the formal
equality of nation-states in the western European space, is the shorthand used by globalization scholars
for the extended period in which the nation-state was the central object of analysis in international law
and thus international relations. Strange considers this a mythical creation, where totemic equality hid a
continuous asymmetry of bargaining power between nation-states. See Strange, S. (1998) What
Theory? The Theory in Mad Money [November]
www.warwick.ac.uk/fac/soc/CSGR/wpapers/wp1898.PDF visited 10 June 1999
162
Corporate governance is a more careerist concern of business lawyers than Internet governance, and
it is here, particularly in the literature dealing with the Sarbanes-Oxley Act post-Enron/WorldCom, that
legal scholars particularly deal with such issues.
163
In policy terms most explicitly treated as a norm by Baird Zoe (2002)Governing the Internet:
Engaging Government, Business, and Nonprofits, Foreign Affairs, November/December 2002
164
Mueller, Kleinwachter, Hofmann
48
This chapter– and m y previous work – attem pts to interpret the interplay of neo-classical
econom ic and norm ative values in the regulation of cyberspace. Further, I em ploy analysis
developed by Susan Strange. In her final unfinished article165, Strange explains that m ainstream
international relations scholars, based on the exam ple set by international lawyers, delineated
artificial disciplinary walls, failing to explain dom estic-international interaction in regulatory
regim es:
The basic assum ption is that world politics – international relations – are conceptually
different from national/dom estic politics, and m ust therefore be studied separately … the
conceptual wall that was built to define the study of international relations has becom e a
prison wall putting key questions … off-lim its in the study of international politics.
She identifies sanction and enforcem ent of power relations, and therefore international law, in
the sam e constricting conceptual straitjacket, though the case is som ewhat overstated given the
increasing recent attention paid to dynam ic convergence of national approaches to regulation,
both in European law and in broader international com parisons, not least in broadcasting
regulation. Strange explains that, despite the static nature of m uch realist international relations
and legal theory166:
This sharp distinction between international law and dom estic law, and correspondingly
between international politics … and dom estic politics is being widely questioned. The
evidence of overlap and of reciprocal influence is abundant.
I concur fully. If this book has any m essage, it is that regulation (and hence perhaps ‘governance’)
is m ulti-layered, in both actors and geographical reaches.
3.3. Socioo--Econom ic Analysis of Inform ation N etworks as G lobal Public G oods

Spar offers a useful historical perspective on the Internet revolution, com paring it with previous
technological and societal disruptions in the 1400s (trans-oceanic navigation and trade), 1850-
60s (railways and telegraph, to which we can add the steam -powered steel ship), 1900s
(electricity, telephony and autom obiles, to which we can add cinem a), and 1920s (m ass
autom obile ownership, broadcasting and com m ercial aircraft)167. She identifies four phases in the
exploitation of technology:
• pioneers (inventors), the creators of technologies
• prophets (popularisers), the developers of m arket and governm ent support
165
Strange, supra n.19.
166
Strange ibid at 3.
167
Spar, D. (2001) Pirates, Prophets and Pioneers: Business and Politics Along the Technological
Frontier, Random House, NY.
49
• pirates (international regulatory arbitrageurs), who test the lim its of the
technology’s challenge to existing public interest goals and stakeholder vested
interests;
• and finally proprietors, legitim ate businessm en who negotiate rules with
governm ents within which to conduct transactions.
She notes that the 1980s satellite broadcasters, especially Australian-Am erican Rupert Murdoch,
were instrum ental in taking advantage of the choice of jurisdiction that the technology perm itted.
In that case, the Television W ithout Frontiers D irective 1989 was the forerunner of Internet
regulation168.
In these case studies, the distribution-transaction m echanism s went through several generations of
step-changed efficiencies, based on infrastructure upgrade, before assum ing ubiquitous consum er
– and thus societal - penetration. From telegraph to telephone to broadcast was a series of
efficiency gains; from railroads to autom obiles to interstate highways was another series of
upgrades. So it is with the Internet. From supercom puters and m ini-, m icro-, personal and
network com puters, the term inal equipm ent has gained power, functionality and affordability.
The infrastructure has upgraded from analogue copper backbones (1970s), to digital exchanges in
the 1980s, to fibre optic backbone in the 1990s. The next stage developm ent, in the 2000s, are
digital lines to the subscriber (DSL) in the infrastructure, and pervasive voice, text and video
com m unication in the term inals. These changes can be com pared to tarm ac local roads and
interstate highways, with Minis (gam e stations, or interactive TV perhaps), Mustangs (m obile
PCs) and trucks (what we now think of as desktop PCs) travelling at high speeds.
W hile the Internet is an exem plar of an alm ost inexhaustible supply of inform ation sharing, and
is therefore an excellent exam ple of what econom ists call a public good, it is not norm atively
positive. Bell quotes Schum peter in this context, who saw little answer to existential angst in such
consum er supply: “The stock exchange is a poor substitute for the Holy G rail.”169 W e now
exam ine the nature of global public goods.
3.4. Public goods

Public goods have two qualities: non-rivalrous and non-excludable consum ption. The
consum ption of that good by an additional user, who cannot in any case be prevented from so
168
Directive 89/552. I acknowledge my part in contributing to Spar’s satellite case study, as discussed
in her preface.
169
Discussion from Bell supra n.19 at 66.
50
doing, does not detract from the use by all others170. Such public goods are necessary pre-
requisites for m arket activities, requiring collective action in order to secure the benefits of peace,
m onetary exchange, law, efficient transactions, education. All are undersupplied or im possible in
the original position: prim ordial m arkets do not exist in a natural state. W ar and currency
inconvertibility are accepted as econom ically disruptive, as are anarchy, lack of sanction for
contractual breach.
Public goods can be ‘pure’, in their non-rivalry and non-excludability, or ‘im pure’, in that som e
form s m ay be rivalrous or excludable, for instance private education or security provision. In the
developed sectors of the econom y, literacy, analytic ability and num eracy are accepted
requirem ents for intellectual capital. It is broadly accepted that universal healthcare, social
security, and other welfare benefits are necessary public goods in late capitalist society. Access to
rem ote com m unications, whether inform ation or interactive com m unications, is recognized as
having an inform ative and educative value beyond the econom ic transaction itself: consider
postal, telegraph and telephony services.
All these services are underprovided by m arkets, because of the ‘free rider’ problem and prisoner’s
dilem m a. W here individuals cannot be excluded without detracting from the welfare of the
group, as when the cost of im posing excludability on the architecture of the good is so high that
it is im practical to im plem ent, those individuals will be able to parasitically consum e the goods.
Therefore, m arkets tend to undersupply the goods, in order to prevent this m arginal erosion of
their profits. Further, even though it m ay ultim ately be net beneficial to supply those goods in
the private m arket, actors choose not to do so because they cannot be sure that other actors will
also invest in such non-profitable goods: they are im prisoned by this coordination problem .
G overnm ents which supply these services are able to take the broadest econom ic view of the
national econom y, including political econom ic considerations of equity and social justice, which
introduces the norm ative and philosophical elem ent into the public goods argum ent. In narrower
price theory econom ics, note that governm ents m ay extract partial paym ent for its provision of
public goods through taxation. The non-excludability of public goods m eans that those who
object to their deploym ent, whether because of pacifist opposition to m unitions or philosophical-
practical opposition to public education, are not allowed to free-ride: taxation applies to all.
Public participation in electoral politics enables som e indirect control over future assignm ent of
public goods.
170
See Samuelson, Paul A. (1954) The Pure Theory of Public Expenditure, 11 Review of Economics
and Statistics 387-9.
51
All of these public goods were also undersupplied by m arkets because m arkets – no m atter how
integrated across the econom y - could not consider long-term consum er interests within price
theory. Thus sustainable developm ent was enhanced by governm ent supply of public goods, of
which universal education, creation of pure scientific knowledge and protection of intellectual
property for applications of that scientific knowledge, are exam ples of forward planning. This is
not to deny the planning abilities of corporations171, sim ply to adm it to the propensity of
com petitive m arkets to extract rents rather than invest for the distant future, a short-term ism
which is exaggerated in Anglo-Am erican stock m arkets (it m ay be no coincidence that these
m arkets are based in the m ost stable legal, political, social and econom ic m arkets over the long
term ).
3.4.1. Public G oods and Private Actors
Public com m unications m arkets, by which we m ean those intended for use by the general
population and regulated as such by governm ent, are the essential m eans of delivering what is
referred to by econom ists as a ‘public good’ in inform ation. A public good is non-rival, which
m eans that the provision of a public good m ay sim ultaneously benefit m ore than one person. A
public good is also non-exclusive, which m eans that once a public good is produced, it is nearly
im possible to prevent others from sim ultaneously benefiting from its production. The end-to-end
design of the Internet m eans that its content is a public good, with control only exerted by its
final user. Free-to-air broadcasting shares the sam e features, but control over its production and
transm ission by governm ents is possible.
Free Riders: Since a public good is non-exclusive, there is little incentive for anyone to pay for its
production, thus leading to the free-rider problem , in which the optim al strategy for an
individual consum er is to let som eone else pay for production, and since there is non-exclusivity,
free-ride off what is produced. If everyone decides to free-ride, then no one will pay, and no
public goods will be produced. Responsible Internet content providers m ight find that m any
irresponsible, harm ful or even illegal providers are free-riding their attem pts to self-regulate.
The probability of free-riding is proportional to the com m unity's size and com m onality of
interest i.e. the sm aller the com m unity, the m ore im pact a person's action or lack thereof will
171
The most long term of corporations anecdotally tend to monopolistic or oligopolistic markets with
high barriers to entry, including national telephone companies (telcos), energy companies (Standard
Oil, Gazprom, Royal Dutch Shell), and government-granted territorial monopolies (Cecil John Rhodes’
possessions in southern Africa, the British East India Company). These examples illustrate the
tendency to government expropriation of assets where the companies impinge too far on sovereignty. A
modern example is IBM, or even Microsoft.
52
have on the com m unity's other m em bers. Furtherm ore knowledge of m em bers' identities is
im portant; once a com m unity's m em bers can identify the free-riders, the com m unity can exert
social pressure upon them to contribute back to the com m unity. In the case of Internet content
providers, such a com m unity does not exist on a global level, but sub-groups, for instance ISPs at
a national level, arguably constitute sufficient com m unity interest to self-regulate.
G overnm ent Provision of Public G oods: Free m arkets would not adequately address societal and
com m unal needs for public goods. G iven the propensity for free-riders to exploit public goods,
governm ent should regulate to ensure that public goods are being adequately supplied for the
benefit of all. Businesses generally recognize individuals prim arily as consum ers, and their
fiduciary duty to their investors fails to recognize fully non-econom ic issues including the right to
privacy, access to harm ful m aterial, and free speech. If com panies collude in the free-m arket,
they could achieve higher profits at the consum ers' expense. Other businesses m ight engage in
predatory behaviour.172 Even self-regulatory groups of well-resourced com panies m ight choose
high-cost technical or legal solutions to raise those com pliance costs to rivals.
G overnm ent action to encourage private provision of public goods includes the favourable sales
tax and postal charge treatm ent given to printed papers (the press), the distribution of
broadcasting, telecom s, and therefore the Internet. The problem of collective action which the
Internet’s globalization presents is intensified by the capabilities and potential of the technology,
and the characteristics of inform ation m arkets. W here public goods are non-excludable and non-
rival, the privatization of inform ation flows offers possibilities for private m onopoly and sub-
optim al exclusion of social groups and individuals. This is a justification for the strong European
tradition of public service broadcasting. The high fixed and low m arginal costs of inform ation
goods have been exacerbated by the Internet, and the potential benefits of positive externalities
from the wider flow of inform ation are threatened by the closure of this form erly open network.
W here initial European public reaction to the Internet resem bled that associated with
environm ental pollution, and negative externalities are highlighted for public concern173, it is the
potential rem oval of the end-to-end positive externalities of the free flow of inform ation and
172
Goldring John (1997) Netting the Cybershark: Consumer Protection, Cyberspace, the Nation-State,
and Democracy in Kahin and Nesson eds. (1997) Borders in Cyberspace 322
173
See Whitehead, Phillip (1997) Draft Report on the Commission Green Paper on The Protection of
Minors and Human Dignity in Audiovisual and Information Services (COM[96]0483 - C4-0621/96) PE
221.804 of 24 April 1997.
53
innovation which is focused upon by McG owan and Sam uelson174. The excludability of
inform ation from users by strong filtering and cryptography, secure paym ent system s and the like
is essential in order to advance a safer, m ore trusted Internet, but changes the nature of the
m edium from a public to an increasingly private sphere. The regulation of this new global space
is essential to m aintain global public goods, of free inform ation, privacy, interconnectedness and
developm ent.
3.4.2. Public Choice and ‘Races to the Bottom ’
Briefly sum m arized, public choice analysis175 of central governm ent in the U .S.: “diagnosed
regulatory capture and unchecked central governm ent growth as principal ailm ents …[critics]
looked in part to the devolution of regulatory authority to junior levels of governm ent for a
cure.”176 The inform ation asym m etries between governm ent and m arket actors can lead to
regulatory distortions in favour of the m arket actors177. Public choice has a weakness beyond an
over-sim plistic reliance on econom ic resources: it reveals taxpayer preferences where m obility is
assum ed (and hence com petition between geographically fixed regulatory jurisdictions) rather
than the total electorate, and therefore the interests of business (e.g. Internet content providers)
rather than those of the disadvantaged (e.g. children):
Com petition causes the content of regulation and the level of public goods and taxation
to be dictated by the private preferences of a narrow, arbitrarily identified class of
itinerant at-the-m argin consum ers or investors ... com petition can force the pursuit of
policies …rem oved from the public interest178.
Internet policy m ay be biased in favour of m ultinational ISPs, therefore, rather than consum ers.
However, the m obility of com panies, and the weak bargaining position of governm ent, can be
exaggerated. N orth explains that “It is no accident that econom ic m odels of the polity developed
in the public choice literature m ake the state into som ething like the m afia.” 179 He then explains
why the param eters of investigation m ust be broadened180: “Inform al constraints m atter. W e need
174
See Gifford, Daniel J. & David McGowan (1999) A Microsoft Dialog, 44 ANTITRUST BULL 619,
Samuelson P. and Kurt Opsahl (1999) Licensing Information in the Global Information Market:
Freedom of Contract Meets Public Policy, 21 E.I.P.R. 386
175
Moe Terry M. (1997) The Positive Theory of Public Bureaucracy in Dennis C. Mueller (ed)
Perspectives on Public Choice: A Handbook, Cambridge University Press
176
McCahery, et al (1996) at 12.
177
Tiebout, C. (1956) A pure theory of public expenditures, 64 Journal of Political Economy 416-424.
178
McCahery ibid at 15.
179
North (1990) Institutions, Institutional Change and Economic Performance, Cambridge University
Press at 140.
180
Ibid.
54
to know m uch m ore about culturally derived norm s of behaviour and how they interact with
form al rules to get better answers to such issues.” W here m arket failure leads to regulatory
intervention on an institutional basis, it acknowledges that regulatory processes m ay be captured
by incum bent interests. Thus, regulation m ay begin as a response to m arket failure, or to broader
social intervention in m arkets, but becom es a feature of regulatory capture, as regulation is
identified with regulated actors.
It is then entirely explicable that governm ents, responsive to dynam ic technological change,
would adapt regulation, with im perfect use of com petition policy, in order to strategically
influence international m arkets. Such a view of m arket failure on a local basis in both political
and econom ic term s m ay appear cynical. However, inform ation plays an im portant role in
com prehending m arkets. Further, technology has also played an under-em phasised role. To re-
exam ine two problem s with neo-classical econom ics are assum ptions about transaction costs and
perfect inform ation181. The growth of firm s and other m arket institutions is explained by the first,
and intellectual property rights and other non-disclosure the second. W hen one com bines the
two in inform ation technologies, which are both disproportionately strategic and equally tend to
m arket failure on a global scale, one has the ingredients for a com pelling m arket failure scenario.
W here ICT has becom e the prim ary driver of growth, and is ubiquitous in internationally
com petitive industries, one m ay anticipate the intervention of governm ents in m arkets to
increasingly bear the hallm ark of these institutionally based strategic analyses. 182.
3.4.3. G lobal Public G oods

Many goods can have ‘joint product’ qualities, providing both public and private benefits, such as
air traffic control, which both ensures safety of all air travellers, while providing additional
services to com m ercial airlines and m ilitary aircraft, subject to fees for their use. The sam e applies
to U K lifeboats, which offer free rescue as a public good, and salvage of com m ercial vessels as a
private good. Both this exam ples apply to all users within national lim its. They also offer intra-
generational and intergenerational benefits, the first by creating safety in air or sea travel to
181
Joseph E. Stiglitz, Information and economic analysis: A perspective, Economic Journal 95
(supplement), 1985, pp21-41.
182
Geoffrey M. Hodgson, Economics and Institutions: A Manifesto for a Modern Institutional
Economics, Cambridge, UK: Polity Press, 1988; Levy, Brian and Pablo Spiller (1994) The Institutional
Foundations of Regulatory Commitment: A Comparative Analysis of Telecommunications Regulation
10 J. Law, Economics and Organization 2 at 201-246.
55
encourage m arkets to develop, the second by encouraging investm ent in these m edia for transport
and thus continuous im provem ent in services and goods. Consider the taxonom y of K aul et al183:
Table: Private and Public G oods
Rivalrous N on-
on-rivalrous
Pure private good: discrete and N etwork
Excludable Market supplied Club G ood
N on-
on-excludable Freely accessible good subject to Pure public good
congestion or depletion (e.g. air quality) Com m ons
Som e com m ons (e.g. radio spectrum )
I have used the term ‘national’ in connection with governm ent supply of public goods, because it
has generally been the rise of the nation-state or larger political entity which has led to the
creation of such public goods, and the decline of such political entities which has led to the
decline of such public goods. The benefits of Rom an citizenship are one exam ple, the ‘Pax
Brittanica’ in V ictorian tim es another. The loss of the basic public goods of external and internal
security, a stable currency and an efficient legal system led to calam itous declines in living
standards for citizens outside the core of both em pires following their decline, particularly where
those public goods were least established and institutionally em bedded, as for instance in
northern Europe in the late fifth century AD and southern Africa in the late twentieth century
AD (this is not to deny the technological and other factors in the relative decline of these
regions). The ‘W estphalian’ nation-state as a supplier of public goods in the twentieth century
(after the em pires of the nineteenth and earlier) has been the outstanding success of political
organisation in history, aided m assively by the technological revolutions which have created the
extraordinary levels of m aterial consum ption and longevity in the vast m ajority of nation-states.
The decline of the nation-state and its causality with the decline of public goods are
therefore central to study of social science. In the N uclear Age, developed nations no longer have
the option of expansion by warfare, and therefore ‘diplom acy by other m eans’ is increasingly
econom ic as well as political. This expansion into the international sphere is described broadly as
globalisation, and it is both m ore integrated and engages m ore actors than at any previous stage
in history. In contrast to earlier periods of globalisation, international econom ic activity by the
m ultinational corporation (MN C) is not dependent on the supply of Im perial public goods, as in
Chinese, Indian, Rom an or British history. Rather, it is dependent, since the end of the U S
183
Kaul, I., Grunberg, I. And Stern, M. A. (1999) Chapter 1: Defining Public Goods in Kaul et al, eds,
Global Public Goods: International Cooperation in the Twenty-First Century, Oxford: OUP
56
hegem ony in 1969, and particularly after the collapse of Soviet Com m unist expansionism in
1989 (or probably earlier, in the afterm ath of the Afghan debacle, and the m idst of the Angolan),
on ‘global public goods’.
The decline in absolute im portance of the nation-state, caused in part by the increasing
integration of MN Cs globally, and thus their structural power in the global econom y, is not a
uniform pattern. The technological and political integration of national econom ies has advanced
to regional form ations, including particularly the European U nion. Other groupings have not
advanced beyond free trade zones, such as N AFTA, MERCOSU R, ASEAN , EFTA. The
advanced state of integration of Im perial trading zones was far greater than these m ore spatially
logical trading groups, with which they have little institutional sim ilarity: for instance, the groups
m ix civil and com m on law traditions in the enforcem ent of transactions, and working languages
of French, English, Spanish and others. The form er hegem ons have m aintained significant
intragenerational transfers of national public goods legacies of capital and technological advantage
over other states: not only the global hegem ons of the U K and U S, but also European hegem ons
France, G erm any, and East Asian hegem on Japan. These states, particularly where they have
actively encouraged their MN Cs to expand aggressively in the post-W W II period (or to m aintain
legacy operations in form er colonies), have positioned them selves as strategic partners for their
MN Cs, and have therefore arguably dem onstrated not the ‘m yth of nation-state decline’ but the
m yth of hegem onic decline. Arguably, the U S and EU are m ore powerful than at any previous
tim e in the post-W W II period. Those less hegem onic m em bers of the EU ‘club’ have also
leveraged power, notably Sweden, N etherlands, Italy and Spain. These nations have taken
advantage of regional public goods in free trade areas and peaceful coexistence with their
neighbours, which are also ‘club goods’ in that they are excludable from non-m em bers, such as
W arsaw Pact countries (witness the invasions of Hungary, Czechoslovakia, Afghanistan and
Poland by Soviet forces). N evertheless, m any nations have taken free-rider advantage of the club
goods, as for instance Sweden, Finland and Austria from N ATO proxim ity, and Yugoslavia from
both N ATO and W arsaw Pact. Further, security provides an exam ple of joint products, where
foreign m ilitary presence creates investm ent in host countries, an im portant elem ent in the post-
war resurgence of Japan, South K orea, Thailand (for V ietnam ), Malaysia and even G erm any and
Austria (though this is a two-edged sword: witness Laos in 1973 and G erm any in 1919).
57
The taxonom y of public and private goods can therefore be extended beyond depletable
and network-club goods, and intragenerational and intergenerational goods, to take acount of
regional and global goods. K aul dem onstrates this184:
Table: Taxonom y of Public G oods based on good’s characteristics
Tem poral – Spatial Regional G lobal

Dim ensions
Intragenerational Pure: European U nion Pure: W orld Court, W TO
[between present actors] Depleted: Data Distribution Depleted: Spectrum , satellite
over N ational Telephony transm issions
N etworks
Club: Inform ation N etworks Club: Internet, air-sea routes
Joint Product: Peacekeeping JointProduct: Encryption
security
Intergenerational [from past Pure: Patent and IPR Pure: K nowledge Creation
to present to future actors] protection Depleted: Biotechnology
Depleted: Sustained
Developm ent
Club: U niversities and libraries Club: G eostationary orbits
Joint Product: Cultural N orm s Joint Product: IMF-W orld
Bank
Despite this realm of global public goods, the retreat of the state from im perialism , and the
consequent increase in econom ic diplom acy, has created substantial coordination problem s185.
W hereas the U S dollar was considered the reserve currency until 1970, the currency instability
following the U S decision to rem ove this global public good has resulted in greater costs for those
nation-states least able to adapt (to hedge) to this m ore com petitive private m arket for capital.
Further, the withdrawal of U S, Soviet and Maoist forces from nation-state civil unrest has
resulted in a renewed tribal division between form erly ideological enem ies, exacerbated
enorm ously by the econom ic disrupture of the Debt Crisis, which began in 1982 and continues.
A further factor, which has been technological rather than financial or security based, is the
collapse of com m odity prices as a result of the increasing adoption of m icro-processor/integrated
circuit based inform ation processing in the developed econom ies. Though N orth Am erican and
Japanese dependence on crude oil continues, if som ewhat m itigated by the discovery in Europe of
huge natural gas reserves, the resurgence of the early industrial environm ental m ovem ent on a
global scale since the 1970s was spurred by the Oil Shocks of 1973 and 1979, creating a tripartite
– governm ent, business, civil society – m ovem ent towards a balanced energy ecology. The
184
Adapted for information industries from generally environmental examples in Kaul, I. (1999)
Chapter 2: Intergenerational Public Goods, in Kaul et al, supra n.43, at 24-5.
185
Olson, M. (1971) The Logic of Collective Action, Harvard University Press.
58
response to environm ental and econom ic danger of dependency on oil (the latter constantly
underestim ated) has been a rapid m ovem ent towards supply of global public goods: in this case,
environm ental policy, beginning with the Montreal Agreem ent of 1987 on banning CFCs.
Though that response is patchy, and EU environm ental standards have leapt ahead of global
standards (unsurprisingly in a geographically interdependent and resource poor developed
region), it dem onstrates a m ovem ent towards global – or at least international - responses.
N egative externalities are the focus of environm ental policy, and the spillover of pollution to
other nation-states is the graphic illustration of negative externalities. Public goods obviously
contain various externalities, as they are non-rivalrous and non-excludible, which indicates that
they will carry externalities. Most of the public goods given as exam ples in the introductory
discussion contained positive and negative externalities (peace is positive for the nation-state
involved and its neighbours, warfare is difficult to contain, especially where technology perm its
world war), and it is becom ing increasingly plain that technological advances perm it public good
externalities to becom e global in nature. Thus, U S hegem ony in security perm itted securing of
the K uwaiti oil fields in the 1991 G ulf W ar. The positive externality was a stable oil price in the
im m ediate afterm ath. The negative externality possibilities include resurgence of Islam ic
sectarianism , developing country fear of a new W orld Order based on U S hegem ony, the
continued reliance on oil as the basis of energy policy, the failure of oil-producing states to
develop a m ore balanced econom y, and a circularity which resulted in high oil prices returning in
1999-2000. Another exam ple is that high pollution and social security costs in developed
countries have encouraged the trend which was already evident towards ‘hollowing out’ of heavy
m anufacturing industries from developed countries to developing countries. This positive
externality for developing countries also resulted in higher growth in developed countries, though
unskilled m ales did suffer net losses. However, the public good could not be confined, and the
shift in production to the developed world was attacked by environm entalists in the developed
world and by trades unionists representing those unskilled m ales, together with anti-neo-
im perialists in the developed world (despite the positive social benefits of foreign direct
investm ent FDI in higher wages, technology transfer and capital investm ent). Thus, U S plans to
focus on electronic com m erce and inform ation technology in the Seattle Round were derailed by
labour and environm ental protestors from the developed world, as well as developing country
protests at exactly the other protestors!
G lobal public goods are clearly a difficult category, caused by that ‘global’ term . Regional public
goods are broadly identifiable, as well as bilateral and plurilateral public goods. But global public
59
goods? There is no world peace or world currency or world governm ent, and U nited N ations
failures in all these areas are constantly attacked in the U S Senate Foreign Affairs Com m ittee, by
its chair Jesse Helm s and others. It is helpful here to consider the concepts of interm ediate and
final public goods: world peace or an interconnected Internet are final goods, but the
interm ediate organisations and agreem ents which give rise to them are essential186. In focussing
on process (interm ediate goods) rather than outcom e (final goods), critics pay attention to
sym ptom , declining state sovereign independence, rather than cause, interconnectedness at club,
regional and global levels.
This causal effect is a difficult process to analyze, and the leading analysis of the developm ent of
global inform ation goods is the U S sociologist Daniel Bell187. Adding to the spatial and tem poral
dim ensions outlined above, he explains that:
The them e of post-industrialism applies prim arily to changes in the social structure (the
techno-econom ic order) and only indirectly to those in the polity and the culture …
each now operates under axial principles that are contrary to each other.188
In his view, Marx united these three structures in analysing the historical contingency of early
capitalist industrialisation in m id-nineteenth century England, creating a theory of transform ative
socialism which depended on an urban proletariat which no longer exists in large m easure in
post-industrial society. Rather, em ploym ent in post-industrial society involves the application of
scientific principles predom inantly in the creation of services, the application of knowledge to
inform ation in order to create additional inform ation and potentially knowledge, even though
polity and culture will not react equally in all societies to these changes. There is no such “causal
trajectory” as in Marx’ “iron necessity” of the laws of capitalism 189: “But a post-industrial society
is characterized not by a labor theory but by a knowledge theory of value.”190 There thus arises a
knowledge class based on a rough m eritocracy, driving techno-econom ic progress from the
hedonistic m aterialism of late industrial society towards a new series of social structures.
186
Kaul et al (1999) at 13.
187
Bell, D. supra n.19.
188
Ibid at p.xcix.
189
Op cit at pxcvii.
190
Op cit at p.xcii. He shares this insight with Habermas, J. (1970) Towards a Rational Society,
Boston, Beacon Press, p.104.
60
Table: Societal Structure and Resource Inputs191

Techno--econom ic
Techno factors and Resource inputs and consum ption
polity/cultural loci in response characteristics
Pre-
Pre- Industrial Agrarian Econom ics of scarcity; Local socio- Scarce N atural Resources192: Land; W ater;
cultural com m un-ities; Polity based on Sunlight and Heat; Slavery - no consum er
landed wealth and wars of acquisition of class beyond ruling class (hence trade in
natural resources luxuries) – labour/reality of tools and nature
Industrial N ational industrial Econom ics of Accum ulated scarce capital and labour
Disjuncture between supply and dem and; (prolectarian) inputs; m echanisation replaces
socio-cultural com m unities based on labour; electricity replaces steam ; ‘creative
W estphalian nation-state; polity transform ed destruction’ rem edied by collectivisation
into m arket-acquisitive industrialisation: (Soviet and Maoist) and corporatism ; rise of
econom ic im perialism urban bourgeoisie – work/reality of
m achines and factories
Post-
Post-Industrial Econom ics of Inform ation; m ore nearly Inform ation and K nowledge distributed
perfect m arket inform ation; socio-cultural through club networks, often joint products
and political com m unities variable in spatial between public and private spheres –
and tem poral adaptation; pre- and industrial work/reality of inform ation processing in
productivity determ ine post-industrial social-knowledge groups
transform ation
Bell explains that:

“the techno-econom ic changes pose “control” problem s for the political order, [hence]
we find that the older social structures are cracking because political scales of sovereignty
and authority do not m atch the social scales.”193
The scales of techno-econom ic structure are increasingly global, while those of politics and
culture rem ain overwhelm ingly national: “the nation state has becom e too sm all for the big
problem s of life, and too sm all for the sm all problem s.”194 Further “by the end of the twentieth
century one m ajor new feature will have intervened to change the character of the answers – the
increased interdependence of the world econom y and the rise, with new telecom m unications and
jet transport, of the world business corporation.”195 Here he relies on the pioneering work of Ray
V ernon and the Harvard Multinationals Project196, including Spar.
191
Adapted from Bell, pp.lxxxix-xci.
192
Ibid at p.lxxiv and p.486.
193
Ibid at p.lxxxiii.
194
Ibid at p.lxxxi, quoting from Bell (1977) The Future World Disorders, reprinted in Bell, D. (1999)
The Winding Passage.
195
Ibid at p.484.
196
See V ernon Raym ond (1971) Sovereignty at Bay: The Multinational Spread of U nited States Enterprises,
N ew York, N Y, Basic Books
61
3.5. N ew Institutional Econom ics, Public G oods and the Com m ons
The com m ons is defined by V eblen as “collective action in control of individual action”197. He
defines an institution as “a usage which has becom e axiom atic and indispensable by habituation
and general acceptance”198. Recent scholarship has exam ined the issue of how institutions can
‘lock in’ behaviour, bound the rationality of actors via rule-setting and constrain change, thus
creating path dependence that leads to a particular outcom e at a later stage. This has been used by
N obel Laureate N orth to explain the apparently ‘locked in’ behaviour of different national
econom ies, with som e endem ically corrupt and failing, others generating increasing prosperity
through a positive feedback effect from good governance. This re-em phasis on institutions rather
than the neo-classical em phasis on individual actors, especially corporate actors, has created a
broad intellectual agenda in econom ic history and sociology. To som e critics it is a return to
nineteenth century political econom y, to Adam Sm ith and K arl Marx, though all note that it is
inform ed by the post-war m odeling techniques, innovations and m ethodologies of the
quantitative ‘revolution’ in the econom ic and social sciences (such as public choice theory),
enabled by com puting.
Take one critical perspective which sheds light on the debate. Stanfield analyzes in very sim ilar
term s to N orth – but criticizes N orth for reliance on the econom ics of price theory (cost-benefit
transactions)199. He considers structural adjustm ent to technology as ‘institutional adjustm ent’ –
and warns that V eblen’s view of hum an institutions as ossifying in the face of change also perm its
hum an dynam ism where vested interests pursue change200. He sees a ‘critical historical m ethod’201
in J.K . G albraith and Max W eber, and explains that “its evolutionary em phasis introduces social
change and therefore power and culture”202. Further he explains the links between Original
Institutional Econom ics and International Political Econom y. “Bounded rationality, once
articulated in concrete exam ination of decision m aking, leads to rules of thum b, m orally bounded
decisions, and the need for cognition theory” thus m aking institutionalism an older scruffier
version of neo-classical public choice203. “Apparently N orth continues the conventional prejudice
that real incom e is produced only in exchange relations, and hence that it can only be relocated
197
Veblen (1967) Absentee Ownership and Business Enterprise in Recent Times, Boston Beacon Press.
198
Ibid at 233.
199
Stanfield, J. R. (1999) The Scope, Method and Significance of Original Institutional Economics,
XXXIII Journal of Economic Issues 2, 231-256
200
Ibid at 235
201
Ibid at 237.
202
Ibid at 238.
203
Ibid at 247.
62
by redistributive transactions” i.e. the m arket econom y204. W ith Marxism , there is a “shared
vision that powerful corporate interests dom inate politics, culture, and econom ic processes” based
on class and im perialist dom ination205 hence “m uch weight m ust be accorded the liberal concern
for checking and balancing political power and m aintaining an inviolate liberal sphere of
individual freedom ”206 – shares with Haberm asian theory. “Alm ost certainly the G reat Capitalist
Restoration will founder upon the sam e endem ic social problem s that led away from nineteenth
century capitalism : m acroeconom ic instability, ecological destruction, social inequality and
political upheaval”207.
Applications of institutionalism which reintroduce norm ative judgm ents into form erly neo-
classically founded binary (efficient/inefficient and therefore right-wrong) tests are apparent in all
the social sciences. In com petition regulation, McChesney suggests that if m arket failure is self-
correcting, antitrust m ust be a political instrum ent208. In econom ics, Schum peter suggests that
“Econom ically … the case against concentration of econom ic control … m isses the salient point
[i.e.] the political consequences of concentration.”209 G lobally, W ood suggests the dilem m a is to
“solve the problem of effective regulation of international transactions without binding us to a
new worldwide antitrust bureaucracy”210. Meesen claim s this leads to a “com petition of
com petition laws”211.
Slaughter suggests that institutional theory teaches interdisciplinary exam ination – for the thick
narrative of law to add to theories from International Relations, which can be applied to IPE212;
Slaughter uses liberal International Relations theory to argue for a transnational law, grounded in
individuals as well as states213. This shares a perspective with K oh214, as well as Jessup’s earlier
work215. Slaughter states that the interaction between dom estic politics and international
behaviour provides “new worlds to conquer, or at least to analyze”216. This restates her interest in
204
Ibid at 248.
205
Ibid 250
206
Ibid at 251.
207
Ibid at 252.
208
Fred McChesney (1999) Economics versus Politics in Antitrust, 23 Harvard J. L. and Public Policy
1 at 133-144
209
Schumpeter, J. (1942) Capitalism, Socialism and Democracy, at 150, quoted in Bell, D. (1999) at 66
210
Wood, Diane P. (1999) International Law and Federalism: What is the reach of regulation? 23
Harvard J. L. and Public Policy 1 at 97-110, at 110.
211
Karl M. Meesen (1989) An International Antitrust Challenge: Competition of Competition Laws, 10
Int. L. and Bus. 17.
212
Slaughter, A-M., Tulumello, A. S., Wood, S. (1998) 92 Am. J. of Int. Law 3 at 367-397
213
Slaughter, A-M. (1995)International Law in a World of Liberal States, 6 Eur. J. Int. Law 503
214
Koh, Harold A. (1996) Transnational Legal Process, 75 Neb, L. Rev 181, and Koh, H.A. (1997)
Why do nations obey international law? 106 Yale L.J. 2599
215
Jessup, P. (1956) Transnational Law.
216
Supra n.20 at 383
63
the ‘disaggregated state’, consisting of various actors, including N G Os, MN Cs and others, in
voluntary (i.e. self-regulatory) codes – what she term s “soft law”. Abbott term s this “international
governance theory”217: with three fields of enquiry:
1. institutions above the state internationally; but note that “the concept of a dom ain of
unitary and functionally identical rational state actors interacting ‘above’ the level of the
state is deeply artificial and ultim ately counterproductive” (at 384)
2. “social construction through shared norm s”;
3. liberal agency theory, exam ining the role of the state in enforcing international law.
Study of these fields leads on to regim e theory, and even process design – Strange’s state-firm
diplom acy m odel is com m ented upon briefly b y Slaughter218, but the absence of consideration of
econom ic governance219 is a depressingly fam iliar one in public international law, reflecting the
grip of liberal International Relations theory. Trachtm an220, com m enting on Judith G oldstein’s
IPE two-level analysis of dom estic and international regim es:
“seeks to explain the role of dom estic politics in international econom ic cooperation …
utilizing som e of the tools of [IPE] including regim e theory. W hile this set of tools has
not kept pace with the developm ent of the new institutional econom ics….”.
There is clearly a research agenda here of m uch interest for the international lawyer – and it seem s
inevitable that the regulatory lawyer concerned with issues of the Internet is playing their part in
expanding the range of international regulatory law, whether they explicitly acknowledge the
non-national status of their work or not. I have previously stated this as ‘Regulating the G lobal
Inform ation Society: Cyberlaw and International Political Econom y’. Seven years after I and
others first expressed the problem in this way221, the challenges for regulatory lawyers in
explaining their techniques as part of the global attem pt to create institutions and practices to
m eet the networked global com m unications system are growing. The rem ainder of this book
attem pts to elaborate on som e of the research challenges and potential routes to answering those
challenges. W e begin that journey in the following chapter with the physical infrastructure of the
Internet: the telecom m unications system .
217
Abbott, K.W. (1996) presentation to Yale Law School; see Slaughter et al at 384 their note.114
218
See at n.125, p386: “a whole new research program” can be embarked upon based on Strange’s
work.
219
For instance Arrow. K. (1995) Barriers to Conflict Resolution.
220
Trachtman, J. (1999) Book Review: the WTO as an International Organisation ed. Anne M.
Krueger, London University of Chicago Press, 1998, 92 Am. J. of Int. Law 3 pp598-601 at 599.
221
See author list in Marsden (2000) supra, and speaker list at Warwick-OECD conference
‘Communications Regulation in the Global Information Society’ (June 1999): http://www.digital-
law.net/warwick/
64
Chapter 4: Telecom m unications Regulation – Towards Behavioural Regulation
“An antitrust price squeeze test…yields no inform ation about consum er

welfare.” Crandall and Singer (2005)222
This chapter identifies m ajor trends and dem onstrates evolving com petition principles in the EU
m edia sector by discussing cases and literature in the deploym ent of broadband carriage. The
chapter finds prim arily that institutional barriers to reform of com petition in broadband create
bottlenecks in any policy reform process. This is in part because incum bent actors have used
com petition law to stym ie reform , to create a ‘juridification’ of the system preventing rapid
regulatory action223. Policy discussion in telecom s needs to be broadened to consider structural
flaws in the institutions that govern these regim es. The chapter takes a broad N orthian224 view of
institutions to encom pass governance via m arkets, state and society in order to provide this view.
4.1. Introduction: Institutional Failures in Com petition Policy
In addressing failures in digital m edia com petition law and policy, I will not take a standard
European com petition law approach, exam ining individual statutes and cases. N or will I m odel
the U S standard: econom ic analysis of the law225. Instead I begin from policy for two reasons. It is
essential to view m arkets as: dynam ic and com plex; political, econom ic, legal and social226.
Broadband as a sector is very hard to m onitor discretely in econom ic term s. The key question is
whether special sectoral regulation should continue for long run bottlenecks, or whether these are
222
Crandall, Robert W. and H.J. Singer (2005) Are Vertically Integrated DSL Providers Squeezing
Unaffiliated ISPs (and Should We Care)? at http://ssrn.com/abstract=710601
223
See Scott, C. (1997)The Proceduralization of Telecommunications Law: Adapting to Convergence
JILT [Journal of Information, Law and Technology] in:
http://elj.warwick.ac.uk/jilt/wip/97_3scot/default.htm
224
North, Douglass (1990) Institutions, Institutional Change and Economic Performance, Cambridge
University Press
225
See Buiges, P. and P. Rey eds (2004) The Economics of Antitrust and Regulation in
Telecommunications, Edward Elgar; Crandall, Robert W. and Flamm, Kenneth (eds)(1989) Changing
the Rules: technological change, international competition, and regulation in communications,
Brookings Institute, Washington D.C. A broader interdisciplinary treatment is Einhorn, Michael (2004)
Media, Technology, and Copyright: Integrating Law and Economics, Edward Elgar
226
Larouche P. (2000), Competition Law and Regulation in European Telecommunications, Hart
Naftel, Mark and Spiwak, Lawrence (2001) The Telecoms Trade War, Hart Publishing,
http://www.phoenix-center.org/telindex.html
65
becom ing unnecessary as m onopoly rents are short-lived?227 Typically, practitioners claim that
convergence of carriage and content m eans that we should adopt a com petition-based approach,
but I will argue to the contrary, that without structural change in the sectors, such an approach is
bound to fail. My analysis is based on standard vertical decom position of digital m edia m arkets,
which identifies creation and distribution – ‘content’ and ‘carriage’228. This article builds on
previous analysis published variously in the period 2001-3229. There are very com plex vertical
linkages, which creates particular problem s in com petition evidence: today’s com plainant is
tom orrow’s supplicant230.
The m ost far-ranging exam ination of the behavioural abuses conducted by a m onopolist in
defence of its dom inance is often held to be that of Microsoft, in the U S case which ran from
1995 to 2001, and the ongoing European Com m ission case231. However, the evidence collection,
enforcem ent and scrutiny afforded to dom inant telecom s com panies dwarfs that afforded to
Microsoft232. In the case of the longest-regulated incum bent in a com petitive m arket, it even
resulted in the structural separation of the com pany’s regulated access bottleneck, as we will see at
the end of the chapter.
The developm ent of m arkets in real property (local loop and radio spectrum ) and m edia IPRs
(Intellectual Property Rights), has been severely ham pered by the failure to delim it and efficiently
transfer property rights233. It is not an exaggeration to state that the developm ent of the concept
227
Laffont J.J. and Tirole J. (2000), Competition in Telecommunications, MIT Press; Marsden C.
(2000) ‘The European digital convergence paradigm: from structural regulation to behavioural
competition law?’ in Erik Bohlin et al (eds.) Convergence in Communications and Beyond, Elsevier
Science, chapter 20,
228
Cowie, Campbell and Marsden, C.T., 1999, ‘Convergence, competition and regulation’, INFO issue
1, pp.53-66
229
Marsden C. (2003a) Chapter 8 in Noam, Eli M. et al (ed.), 2003, Television over the Internet:
Implications for Networks, Infrastructures, Strategy, Policy and Content, Lawrence Erlbaum
Marsden, C. (2003b) Keynesian-Korean Strategies for Recovering from the Great TMT Depression 31
Intermedia 2 at 38-42
230
Geradin, D. and O’Donoghue, R. (2005) The Concurrent Application of Competition Law and
Regulation: The Case of Margin Squeeze Abuses in the Telecommunications Sector, GCLC Working
Paper 04/05, College of Europe, Bruges
231
See U.S. v. Microsoft Corp., No. 98-1233 (U.S. District Court for the District of Columbia),
Findings of Fact, ¶¶39-41. Available at www.usdoj.gov/atr/cases/f3800/msjudgex.htm
232
For evidence from the early modern period of telecoms regulation, see Noam, E. (1989)
International Telecommunications in Transition, pp257-297 in Crandall, Robert W. and Flamm,
Kenneth (eds) Changing the Rules: technological change, international competition, and regulation in
communications, Brookings Institute, Washington D.C.; Robinson C. (ed), Successes and Failures in
Regulating and Deregulating Utilities, Edward Elgar.
233
For discussion of the theory, legislation and practice, see variously: Ungerer, H. (2002) ‘Media in
Europe: Media and EU competition law’
http://europa.eu.int/comm/competition/speeches/text/sp2002_006_en.pdf;
Mayer-Schoenberger, V. (2003) Crouching Tiger, Hidden Dragon: Proxy Battles over P2P File
Sharing, mimeo; Fallenböck, Markus (2003) ‘On the technical protection of copyright: the Digital
Millennium Copyright Act, the European Community Copyright Directive and their anti-circumvention
66
of property rights, together with a consistent and m easured exam ination of the public interest in
regulating and assigning those rights, are the prim ary challenges for both governm ents and
m arket actors, beyond even the extraordinary pace of technological innovation which is creating
the space within which those rights will be exercised234. N orth and W illiam son have
dem onstrated that property rights are the basis for transferable wealth and therefore econom ic
developm ent235. The latter following Coase236, has shown that where transaction costs in property
rights are sub-optim al, corporations will be form ed, internalizing those rights within
organizations. Failing property rights transfer, econom ies are reduced to barter, in which roughly
equally valued goods and services are exchanged without m onetization.
This paradigm , that without property rights being efficiently assigned, m onetization of transfers
is inefficient where possible, and replaced by barter, is the situation pertaining to m uch of the
broadband m edia m arket. This applies to traditional broadcasters and Hollywood studios, but
also to the m usic industry, and to broadband infrastructure owners. The solutions adopted by
m arginal consum ers denied access to broadband access and content has been to share resources
freely: via peer-to-peer networks such as W iFi237, peer-to-peer applications such as voice over
broadband, and peer-to-peer content such as Creative Com m ons licensed, or m ore frequently,
illegally downloaded content. Reinstituting fair pricing by corporations and fair use by consum ers
provisions’ International Journal of Communications Law and Policy, issue 7 at

http://www.ijclp.org/7_2003/ijclp_webdoc_4_7_2003.htm
Geist, Michael (2000) iCraveTV and the New Rules of Internet Broadcasting, 23 University of
Arkansas at Little Rock Law Review 123; Hugenholtz P.B. (2000) ‘The Great Copyright Robbery.
Rights allocation in a digital environment’, paper presented at New York University School of Law, 2
April, at http://www.ivir.nl/publications/hugenholtz/PBH-Ecology.doc
Hugenholtz, Bernt, 2002, ‘Copyright and Freedom of Expression in Europe’, in N. Elkin-Koren &
N.W. Netanel (eds.), The Commodification of Information, Information Law Series, Vol. 11, Kluwer
Law International, pp. 239-263.
Jaszi (2000) Video On The Internet: Icravetv.Com And Other Recent Developments In Webcasting
Hearing Before The Subcommittee On Telecommunications, Trade, And Consumer Protection Of The
Committee On Commerce House Of Representatives - One Hundred Sixth Congress; Second Session:
February 16, 2000 Serial No. 106-94. Testimony of Peter Jaszi, at:
http://comnotes.house.gov/cchear/hearings106.nsf/a317d879d32c08c2852567d300539946/8d45454ad2
93f0db85256965006e67c1/$FILE/94.pdf
JumpTV (2001) Copyright Board Denies Application to Suspend JumpTV Proceedings June 4 at
http://www.jumptv.com/mediaroom/pr2001-06-04.1.html
Yu Peter K. [2005] Intellectual Property And The Information Ecosystem Michigan State L.Rev. 1.
234
Figueiredo, R.J.P. and Spiller, P. (2000) Strategy, Structure and Regulation: Telecommunications on
the New Economy, 2000 L.REV. M.S.U.-D.C.L. 1 at pp253-285
235
Williamson, O. (1975) Markets and Hierarchies: Analysis and Antitrust Implications, Free Press; for
telecoms see Cherry, B. and Wildman, S. (2000) Preventing Flawed Communications Policies by
Addressing Constitutional Principles, 2000 L.REV. M.S.U.-D.C.L. 1 pp55-107, at 96-99, for e-
commerce Burk, Dan L. (1999) Virtual Exit in the Global Information Economy, 73 Chicago Kent Law
Review 4, at 943-995.
236
Coase, R.H. (1937) The Nature of the Firm Economica 4 386-405
237
See Croxford, I. and Marsden, C. (2001) I Want my WiFi! The Opportunity for Public Access
Wireless Local Area Networks in Europe, from Re: Think!, London, at www.re-think.com
67
requires (but is unlikely to result in) governm ent intervention. This is not the case in East Asia.
The claim is that K orea, China and Japan’s governm ents have been able to force incum bent
m onopolies in the direction of a broadband future for two predom inant institutional reasons:
they had less invested in the analogue past; they have m ore reasons to coordinate carriers and
software actors, because the econom y and polity pays m ore interventionist attention to hardware
exporters, such as Sam sung, than com m unications and m edia service providers.
Table 1: Broadband Household Penetration Top 10 Countries238
Country % Household Penetration 12/2004
South K orea 76
Hong K ong 72
Taiwan 56
Israel 52
Canada 49
Singapore 49
N etherlands 45
Japan 42
Denm ark 42
Switzerland 39
In Europe, the explosive growth of Internet access to households followed what becam e known as
the Freeserve m odel – nam ed after then-Dixons plc, now France Telecom subsidiary
Freeserve.com . Offering ‘free’ ISP service by revenue sharing per-m inute charging with the
incum bent telco, a wave of ISPs quickly achieved a m illion subscribers or m ore: Freeserve
launched in Septem ber 1998 and achieved 1m illion in February 1999. The econom ic m odel was
transitory because it depended on per-m inute charging, and of the early 1999 successes, Freeserve
was purchased by France Telecom and W orldonline/LibertySurf were bought by Tiscali, all in
2000-1. The m odel had lasted less than 18 m onths. This extrem ely weak and fragm ented239
European ISP m arket was dom inated by U S m ultinationals that grew by acquisition in the late
1990s: AOL (buying Com puserve 1998 and N etscape 1999), U U N et and Microsoft. The latter
dom inated through its parent com pany’s size and assets, but the first two through acquisitions
(finally m erging with Tim e W arner and MCI W orldCom respectively in 1998-9 at the peak of
the bubble).
By 2000, two critical decisions were m ade that created conditions for com petitive ISP m arkets in
Europe. In the U K , flat rate Internet access was m andated by Oftel, the telecom s regulator,
238
Source Broadband Trends (2005) http://broadbandtrends.com/Fast%20Facts.htm
239
Creating the potencial for large trade inequalities with the US: see Noam, E. (2000) The Coming
Trade Wars in Erik Bohlin et al (eds) Convergence in Communications and Beyond, Elsevier Science.
68
against the opposition of British Telecom , and in response to com plaints by W orldCom 240. In the
rest of Europe, Local Loop U nbundling241 perm itted ISPs to offer access over ISDN lines (digital
lines at 15-130% faster speeds than analogue phone lines) which could be upgraded to
broadband Digital Subscriber Lines242.
Table: Developm ent of European Consum er Internet Market
Date Developm ent Com panies Result Legacy

1995-
1995 - US Com puserve >1m illion European
HH Early adopter
date investm ent AOL U U N et subscribers m arket and tech-
Phase 1 MSN savvy ISPs
W inter ‘Freeserve’ Freeserve 1m illion to 10m illion Mass m arket
1998-
1998 -9 m odel Tiscali European HH subscribers consum er ISPs
Phase 2 LibertySurf
W orldOnline
2000-
2000 -2 FRIACO Freeserve Increased tim e online, reduced Mature m arket
Phase 3 AOL charges – down to under 10 and e-com m erce
Yahoo euros unlim ited per m onth
2000-
2000 - LLU and Tiscali Broadband Internet, potential V oice call revenue
date broadband – FastW eb for rich content – served by decline and
Phase 4 V OIP Tele2 video, viruses and P2P capital-intensive
copyright theft in absence of BSPs: regulation?
legal services
There is now a m ore m ature ISP m arket in Europe, dom inated by incum bent ISPs. Strong
entrants include AOL, W anadoo (outside France where it is incum bent), and Tiscali.
Because the TMT sector linked capital and technology m arkets so tightly on a global basis never
before seen, it created an investm ent ‘bubble’ in developed countries, transm itted the irrational
exuberance and easy capital availability of W all St over other interlinked capital m arkets, and
collapsed the bubble with extrem e rapidity in 2000-1243. A com bination of technological change
and financial m arket revaluation caused ‘the perfect K ondratieff W ave’ – a technology burst
240 See Oftel’s Direction of May 2000 at http://www.ofcom.org.uk/static/archive/oftel/publications/internet/fria0400.htm

241 The lines are bought wholesale from the incumbent and resold to ISP customers. The incumbent’s ISP competes for end-customers, theoretically from the
same wholesale price. Independent ISPs continue to allege that incumbent wholesale prices to their retail ISP allow illegal subsidy to be recovered. See
www.ectaportal.com for latest developments in competitor’s regulatory strategy.
242 In the UK, British Telecom was permitted by Oftel to charge and provision at sufficiently unattractive terms that LLU produced almost no commercial
interest until 2003, a total of only 7600 lines at September 2003, see:
http://www.ofcom.org.uk/static/archive/oftel/publications/internet/internet_brief/broad1003.pdf at p22. OfCom’s priority on replacing the abolished Oftel on
28 December 2003 was to adjust the terms of LLU: see http://www.ofcom.org.uk/consultations/current/wbam/wbam.pdf
243
Abramson, Bruce (2005) Digital Phoenix: Why the Information Economy Collapsed and How It
Will Rise Again, MIT Press; Noam, E. (2003) The Effect of Deregulation on Industry Concentration:
An Analysis of the Telecoms Act of 1996 and the Industry Meltdown, Columbia Science and
Technology Law Review at http://www.stlr.org/html/volume4/noam.pdf
69
bubble effect244. W hy were the broadband supply and dem and curves so uncoordinated? I want
to suggest governm ent and policy com m unity could and should acted in two long-term ways.
How to ensure there were less bottlenecks in supply? How to prim e dem and via m edia policy?
The first part is about carriers, the second about copyright, and both are about property
m onopolies in the inform ation age which have to be opened up for fair and necessary
com petition to em erge (for a contrary view, see Sidak245). For carriage, I will talk about local loop
unbundling and telcos. Issues explored in telecom s – including the critical question of access to
networks246 – are not yet converged with those arising in m edia regulation, which begins from a
rights-based approach. Further, the apparently unregulated arena of com puting is actually a blend
of m edia and telecom s regulation, though this truth is often obscured by the relative paucity of
regulation to this point, and the tendency towards self-regulation, whether de facto or through
industry-wide agreem ent.
In conclusion, I will consider the prospects for European and N orth Am erican econom ies to
escape the throttling em brace of their dom inant analogue m onopolists and break into a
broadband future. The overall argum ent is sim ply that the W est has failed to overcom e its
m onopolies in carriage, whose policy networks are so deeply em bedded that it is im possible to
overcom e these institutional barriers to innovation and change247.
4.2. Broadband Carriage: Infrastructure Deploym ent

The investm ent required for the developed world to achieve high-speed access to an always-on
Internet is enorm ous248. A m ixture of asym m etrical regulation of the form er national m onopolist
through price control, introduction of com petition in m obile, long distance and international
services, and social pricing of access to ensure provision of service to the econom ically
disadvantaged, has generally succeeded in reducing the price of telecom s services. This has been
m ost dram atically the case for business in econom ies which liberalized earliest. Arguably, this has
244
A description from Hal Varian’s keynote closing speech to ITS Europe 2003, Helsinki Finland.
245
Sidak J.G. (2004), “The failure of good intentions: the collapse of American telecommunications
after six years of deregulation”, in Robinson C. (ed), Successes and Failures in Regulating and
Deregulating Utilities, E. Elgar, 1-15
246
Brennan T. (2003), “The FCC and policy federalism: broadband Internet access regulation” pp173-
199 in G. Madden (ed.), International Handbook of Telecommunications Economics V. III, E. Elgar;
Cave M. (2004), “Economic Aspects of the New Regulatory Regime for Electronic Communication
Services”, in P. Buiges and P. Rey (eds), The Economics of Antitrust and Regulation in
Telecommunications, E. Elgar, 27-41
247
Levy, B. and Spiller P. (1994) The Institutional Foundations of Regulatory Commitment: A
Comparative Analysis of Telecommunications Regulation 10 J. Law, Economics and Organization 2 at
201-246
248
Shelanski, H. (1999) The Speed Gap: Broadband Infrastructure and Electronic Commerce, 721-744
in 14 Berkeley Technology Law Journal 2
70
been a trium ph for price caps and social regulation of access, rather than the result of free m arket
com petition, which has only em erged in the m ost profitable long distance or m etropolitan sectors
of business, international and long distance calling. The real wholesale com petition which will
drive down local telephone costs has to em erge in the local loop. Local loop com petition is not
yet in place in any m ajor European m arket except Benelux and Scandinavia. Further forwards,
broadband Internet access is increasingly possible through ‘third generation’ (3G ) m obiles and
voice-over-Internet-Protocol (V oIP). The extent to which com petition has survived is very m uch
a secondary consideration to these m ore intrusive regulatory factors. Barnes argues that the basic
policy choices are political, not econom ic, and com petition policy is not yet a dom inant them e in
telecom s249.
Consider governm ent incentives to encourage broadband local loop investm ent. The gatekeepers
face two property rights challenges, which have becom e especially profound in the European
m arket: third generation (3G ) wireless spectrum rights and local loop unbundling (LLU )250.
N ational regulatory differentials caused chaos in the auction of 3G licences for m obile telephony.
The failure to coordinate a com m on standard for the European 3G auctions is one of the twin
tragedies of m em ber-state regulation of telecom s in 2000. The other is the initial failure to
develop alternative local loop broadband services, by divesting national telcos of their cable TV
division in advance of the 1998 liberalization, to perm it upgrading of services from analogue to
digital. A partial answer has been to ‘unbundle the local loop’ by co-locating rival operators’
switching equipm ent in local telephone exchanges, perm itting them to use the higher bandwidth
elem ent in the copper wire lines for Digital Subscriber Line (DSL) services. W here the
com bination of com petitive infrastructure being least advanced and regulators m ost advanced,
LLU has been a partial success. Com petition is seen to be em erging in broadband via these two
routes. The regulator is opening access to the assets of the telcos, which previously had been
com m itted to m arketing ‘m idband’ 128K b/s Integrated Service Digital N etworks (ISDN ). The
telcos rationally feared that DSL would cannibalize their investm ent in ISDN revenues. The
property rights sham bles proves the poverty of regulatory zeal, with spectacularly high m obile
auction prices, intransigent telco (and cable) m anagem ent and national regulation of LLU , which
have delayed broadband roll-out.
249
Barnes, Fod (2000) ‘Commentary: when to regulate in the GIS? A public policy perspective’,
Chapter 7 in Marsden, C. (ed) Regulating the Global Information Society, Routledge, New York
250
Marsden, C. ed.(2000) Regulating the Global Information Society, Routledge, New York
71
Regulators have responded typically via requirem ents such as these introduced by Irish regulatory
Com Reg and supported by the European Com m ission in the case of joint dom inance in the retail
m obile telephony m arket251. Obligations im posed were:
• “Obligation to provide network access following a reasonable
• seeker
• N on-discrim ination obligation
• Obligation of price control to be im plem ented by way of cost orientation
• Obligation to prepare separated accounts
• Obligation to im plem ent appropriate cost accounting system s.”
These are m inim al requirem ents under the European regulatory fram ework for operators holding
Significant Market Power (SMP). However, im position of m ore onerous conditions than these
have produced relatively little com petition in the m arket and broadband rollout as we will see, in
the following em pirical section.
petiti
4.3. Com parative Broadband Penetration: Q uantitative Evidence of Com petition
There have been several recent com parative studies of telecom regulation in Europe252, all of
which point to extrem ely large disparities in im plem entation of telecom liberalization in the
period. This is reflected in current levels and types of residential DSL penetration. The argum ents
between infrastructure com petition and unbundled network elem ents-local loop unbundling drag
on253. Then-U K E-Com m erce Minister Mike O’Brien declared at Europe-Asia Sum m it on 21
February 2005: “The U K now has the m ost extensive broadband m arket in the G 7.”254 U K has
good coverage: 50% cable reach at m idband, 99.6% DSL; BT Retail DSL had only 26% of new
connections in Q 4 2004. However, in LLU , the U K had 27,800 lines unbundled out of 4.2m
total: 0.6% of the total, lower than all other EU -15 countries including Ireland and Belgium (see
Table 2 below). Both in Europe following the 1998 opening of telecom s to full (sic) com petition
and in the U S following the Com m unications Act 1996, the incum bents generally showed how
251
See Barroso, Jose Manuel (2005) Case IE/2004/0121: Access and call origination on public mobile
telephone networks in Ireland: Comments pursuant to Article 7(3) of Directive 2002/21/EC Letter of
20 January to ComReg
252
See British Institute of International and Comparative Law (2004), Effective Access and Procedure
in Telecommunications Disputes in Europe BIICL, London at
http://www.biicl.org/admin/files/BIICLTelecoms20Report202004.pdf
ECTA (2005) http://www.ectaportal.com/uploads/BroadbandScorecardQ42004.pdf
Also various Commission studies by consultants at
http://europa.eu.int/information_society/topics/ecomm/useful_information/library/studies_ext_consult/i
ndex_en.htm#2003;
253
Sandbach, Jonathan (2001) Levering open the local loop: shaping BT for broadband competition, 3
info 1 at 195-202
254
Ovum 2005 http://www.theregister.co.uk/2005/02/22/govt_boradband_g7/
72
to use regulatory procedure to destroy the intent of legislation. Only in Denm ark, where the
regulator had full governm ent backing and substantial technical regulatory resource from
consultants, did LLU work255. Once the com petitors had gone bust in 2000-2001, the
incum bents finally showed how to drive DSL penetration. The sam e occurred in the U S, the
difference being that whereas European cable operators were governm ent-owned and starved of
capital to upgrade to digital, in the U S the m ore m ature industry achieved real success ahead of
DSL incum bents.
Table 2: Residential DSL Types by EU Mem ber States
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
xe a ia
Ita d
Sw pa a
Ire ar y
th Ma rg
re y
Po nd s
ov ia
bo a
m e
Sl tug d
ov a l
ed in
De Re s
en
Cz C ium
t rk
n p.
G ra n d
ng e
la a
nl a
ly
lg i a
UK
G an
S ni
er c
r n
m ni
h ru
Hu ec
F an
Fi oni
Lu Lith a tv
er lt
Sl a k
u
E sm a
Be str
la
P o la
e
ec yp
u
Au
Ne
Retailed by indep. ISP Retailed by incumbent LLU
Several econom etric attem pts explain when the W est was lost, and East Asia’s enorm ous success
in m ulti-m egabit true broadband256. An independent report257 for the U S governm ent has
pointed out six separate research areas into the different rates of broadband penetration in OECD
countries:
1. lessons from broadband deploym ent in leading countries – Canada, South K orea;
2. elasticity of broadband dem and, residential price-sensitivity;
255
Andersen, J. (2005) Regulatory Challenges in a Converging and Liberal Telecoms World, at
http://www.ero.dk/42ACC9DD-B775-4948-8B63-CA7EA18B0329?frames=no&
256
Bauer Johannes M., Junghyun Kim and Steven S. Wildman [2005] An Integrated Framework For
Assessing Broadband Policy Options Michigan State Law Review 21
See the work-in-progress at http://userpage.fu-
berlin.de/~jmueller/its/conf/helsinki03/papers/papers.htm Bauer’s modelling based on real data
suffered from the intermittent level of official data gathering, whether by ITU or OECD. The models
run with contemporary data promise to illustrate more than simply that narrowband users are predictive
of later broadband users. However, qualitative studies using few countries may produce more insights
than quantitative using 30 or more countries (even if, in Bauer’s case, they are self-selectively
developed economies).
257
Balkovich (2003) Balkovich, E., Baer, Walter S., Vollard, Bfor Office of Science and Technology,
Rand Corporation Science and Technology Issue Paper ‘Research Topics for Informing Broadband
Internet Policy’ at http://www.google.co.uk/search?sourceid=navclient&ie=UTF-
8&rls=GGLD,GGLD:2003-48,GGLD:en&q=rand+broadband+policy
73
3. lessons from 1970s-1980s cable deploym ent in the U S;

4. wireless broadband deploym ent (not including W LAN or 3G W AN );
5. prospects for technology leapfrogging to e.g. fibre to the hom e258;
6. do ‘killer applications’ for broadband exist?
It is clear that price-sensitivity has created enorm ously increased dem and for broadband – in the
U K , for instance, a very late deployer of broadband DSL (if not cable) broadband take-up
increased hugely in 2003-4 as prices dropped from £40 to under £15 for a 512K b/s connection.
The price of broadband in Japan proved dem and was hugely sensitive, in a leapfrogging high-
speed deploym ent in 2002-4259. In K orea, price and online gam ing proved im portant dem and
issues260.
258
See Special Issue 2 info 2 (2000)
259
Shinoda (2004) BB Access Network Deployment in Japan Past Present and Future: Reasons and
Issues, Presentation to INET 2004, Barcelona, May, available from this author
260
Brunel for Department of Trade and Industry (2002) Investigating Broadband Technology
Deployment in South Korea, Brunel University Uxbridge, at
http://www.broadbanduk.org/reports/SKorea_report.pdf
The Brunel University study of South Korea claimed six reasons for its success:
Geography and demographics with most population in densely populated urban areas.
Government vision, strategy and commitment, which has proved to be a key factor.
Strong competition, with a number of competitors providing broadband infrastructure.
The ‘PC bang’ phenomenon where large numbers of people use an Internet gaming café
Providing service at an affordable price.
Clear user benefits resulting from broadband access. Within the education sector all schools have a
broadband connection and there is strong support with Korean society for providing high quality
education
74
In countries with greater cable deploym ent – Canada, N etherlands, Denm ark, U S – the option
for intra-m odal com petition has created greater price and service com petition. Adding unbundled
local loops to this existing com petition drives lower pricing and therefore greater penetration –
see France, Japan and now U K 261. Finally, from the exam ples of N apster and K azaa, it is clear
that the killer application at least for early adopters, is access to copyrighted m aterials via peer-to-
peer networks. The increasing network effect of voice over the Internet (V oIP) via services
including Skype prom ises to becom e a new killer application for internationally social groups. In
East Asia, prices are lower and speeds faster than in W estern European rivals.
The U K is far behind the leaders in local loop unbundling as the three following figures
dem onstrate. The first is the October 2005 prediction by the independent dispute adjudicator of
the ability to m ove to local loop unbundling, dem onstrating the continued slow growth, with the
blue line the real perform ance and the green and red lines showing the predicted perform ance.
Figure: Proposed Acceleration of U K Local Loop U nbundling 2004-6262
The second is the projection for broadband penetration by RAN D Europe based on OECD data
for 2000-5. The figure dem onstrates that for several m ajor Mem ber States, saturation will be
achieved by 2008-2010. U nder this assum ption a revised Directive263 will not affect broadband
adoption by existing Internet users. W hat it also indicates is that the U K has rem ained ahead of
261
Cadman, R. and Dineen, C. (2005) Broadband Access Markets in Europe: Why Regulation should
Promote Competition at http://www.spcnetwork.co.uk/uploads/20040315_investment_paper.pdf
262
Source: Office of the Telecom Adjudicator (2006, January 13) LLU Forecast at
http://www.offta.org.uk/charts.htm
263
Broadband speed demand is impacted by rich media application use as well as other factors. A
1Mb/s connection is obviously less useful for video files than 100Mb/s. However, guaranteeing speed
throughout the IP transport is a highly technically challenging task that head-end theoretical access
speed only partially reflects.
75
France in broadband penetration absent significant local loop unbundling, and that broadband
penetration in the U K is increasing faster than in other m ajor m arkets.
Figure. G rowth of broadband penetration towards saturation 2000-2010264
1 00%
9 0%
8 0%
7 0%
6 0%
5 0%
4 0%
3 0%
2 0%
Belgium D enm ark
Finland France
1 0% G erm any Spain
Sw eden U nited K ingdom
U nited States O EC D
0%
2 001 2 002 2 003 2 004 2 005 2 006 2 007 2 008 2 009 2 01 0
N on-users of the Internet indicate a variety of reasons for non-adoption of broadband, am ongst
them cost, sim plicity of use, and harm ful or m isleading content. Their reasons for adopting
broadband can be anticipated to diverge from those of existing narrowband users, and be m ore
com patible with the public safety m easures in the ‘Television W ithout Frontiers’ Directive than
existing narrowband users265.
The third figure shows the extent of broadband globally at Septem ber 2005. N ote that only one
EU country is in the top 7, and that the U nited States rem ains above the five large European
nations, though behind Canada, Japan and South K orea. It is notable that the top nine nations
all have significant alternative network com petition to telecom s (DSL) but that except for
U S/U K /Japan the following countries have little or no cable com petition. G erm any, France and
Italy lag particularly in this respect. Marcus suggests that the U S retrenchm ent from com petition
within telecom s has enabled the European nations – m ost notably the northern European
264
Saturation of Internet population estimated at 1 citizen in 3 over the age of 18. In most European
households, 2.4-2.7 adults inhabit. Taking account of the 10-20% of adults who have not used, nor
intend to use the Internet, 1 in 3 is saturation of the adult Internet population. Source: RAND Europe
Projections based on 2005 OECD dataset. See RAND Europe (2006) at europa.eu.int
265
See for example Oxford Internet Survey (2005) at
http://www.oii.ox.ac.uk/research/oxis/oxis2005_results.pdf. This is part of the 24-country survey, the
World Internet Project, led by the Pew Center: http://www.pewinternet.org/trends.asp
76
m arkets – to overtake the U S266. Prelim inary figures suggest U K and France have overtaken the
U S in overall penetration267.
Table: Point Topic Statistics on Broadband and DSL Household Penetration for Septem ber
2005
Country Penetration % % DSL Total DSL

1. South K orea 80 44 6,628,000
2. Hong K ong 76 42
3. Israel 62 42
4. Taiwan 61 51 3,535,000
5. Singapore 59 32
6. N etherlands 54 32 2,292,000
7. Canada 52 25 3,040,000
8. Switzerland 51 33 1,010,000
9. Denm ark 50 31
10. N orway 45 37
11. Japan 43 30 14,456,000
12. Belgium 43 na 1,192,000
13. Finland 42 32
14. Sweden 39 na 1,031,000
15. US 38 16 17,310,000
16. UK 37 27 6,381,000
17. France 36 34 8,397,000
18. Spain 32 25 3,486,000
19. Estonia 32 na
20. Australia 31 24 1,791,000
Italy 28 27 5,525,000
G erm any 22 21 8,150,000
Poland 10 na 1,041,000
China 10 7 24,655,000
The availability of broadband does not equate to service, as cost and bandwidth vary significantly,
and in this case there is again a clear correlation between network com petition and
price/bandwidth. The ITU report ‘The Internet of Things’ dem onstrates this in a chart in the
statistical appendix.
Table: ITU ‘Broadband Bargains’ 2005268
266
Marcus, Scott (2005) Is the US dancing to a different drummer? 60 Communications et Strategies
pp39-58.
267
Burstein, Dave 92006) DSL Prime Newsletter, January.
268
Source: ITU (2005) The Internet of Things, Appendix at
http://www.itu.int/osg/spu/newslog/Broadband+Bargains+2005.aspx
77
The exponential difference between Belgian and Japanese broadband prices – which is constant
over the past three years – suggests real structural differences in the m arkets.
4.4. European Behavioural Com petition Policy
The current European consensus on com petition in com m unications is based on a shared vision
of contem porary m arket and technological developm ents. The position, outlined in the 1994
Bangem ann Report and the 1997 European G reen Paper on Convergence269, claim s the
following:
• Due to the ‘end of spectrum scarcity’ and the convergence between
telecom m unications, com puting and broadcasting, m uch of the sector specific
regulation that encum bers the com m unications sector can be gradually rem oved.
• W here previous technological constraints m ade som e form s of natural m onopoly
and m arket failure inevitable, both in telecom m unications and broadcasting,
state ownership and regulation will becom e less necessary.
269
See Blackman, Colin, 1998, ‘Convergence between telecommunications and other media. How
should regulation adapt?’, Telecommunications Policy, vol. 22, no. 3, pp. 163-170
Marsden, Christopher T., and Verhulst, Stefaan G., 1999, Convergence in European Digital TV
Regulation, London, Blackstone Press.
78
• In the com plex m arkets that em erge, individual and producer choices articulated
through m arket m echanism s will deliver the widest and m ost efficient choices,
optim um social welfare and increase overall com m unications sector innovation
towards com puting levels.
This went with the ideological grain and the technological and econom ic exuberance of the
period. N o such consensus existed in the detailed discussion of broadcasting and m edia policy in
1997/8, which were deliberately excluded from the 1999 Com m unication which led to the
Electronic Com m unications Services (ECS) Directives of 2002270. The ECS Directives should
have been im plem ented in m ost EU m em ber states by m id-2003 with a process of m arket review
based on those Directives im plem ented by national regulators by January 2005271. The Directives
aim ed not only to harm onise com petition fram eworks but, with a nod to public service
broadcasters, to ‘prom ote the citizen interest’.
W hen it com es to the fram ework for com m ercial broadcasting and the position of
publicly funded broadcasters, and various form s of state aid, som e posit a future of deregulation
in which com m unications are eventually treated as m arkets like any other, whilst others argue
that m any of the public policy objectives in the sector will require perm anent sector specific
regulation, and even that the com petition fram ework currently being im plem ented should be
subordinated to citizen not consum er interests.
W hy is com petition policy not working effectively in term s of the key aim s of prom oting
choice, m arket entry and dealing with abuse of dom inance? A point of consensus is that
270 Commission of the European Communities, 2002a, Eighth report on the implementation of the telecommunications regulatory package COM(2002) 695
final, Brussels, December, at: http://europa.eu.int/information_society/topics/telecoms/implementation/annual_report/8threport/finalreport/annex2.pdf
Commission of the European Communities, 2002b, Directive 2002/21/EC of the European Parliament and of the Council on a common regulatory framework
for electronic communications networks and services, (“Framework Directive”), OJ L 108, at:
http://europa.eu.int/information_society/topics/telecoms/regulatory/maindocs/index_en.htm
Commission of the European Communities, 2002c, Directive 2002/20/EC of the European Parliament and of the Council on the authorisation of electronic
communications networks and services, (“Authorisation Directive”), OJ L 108, at:
Commission of the European Communities, 2002d, Directive 2002/19/EC of the European Parliament and of the Council on access to, and interconnection of,
electronic communications networks and associated facilities (“Access Directive”), OJ L 108, at:
Commission of the European Communities, 2002e, Directive 2002/22/EC of the European Parliament and of the Council on universal service and users’ rights
relating to electronic communications networks and services (“Universal Service Directive”), OJ L 108,
at:http://europa.eu.int/information_society/topics/telecoms/regulatory/maindocs/index_en.htm
Commission of the European Communities, 2002f, Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the
processing of personal data and the protection of privacy in the electronic communications sector (“Directive on privacy and electronic communications”), OJ
L 108, at: http://europa.eu.int/information_society/topics/telecoms/regulatory/maindocs/index_en.htm
Commission of the European Communities, 2002g, Decision No 676/2002/EC of the European Parliament and of the Council of 7 March 2002 on a regulatory
framework for radio spectrum policy in the European Community (“Radio Spectrum Decision”); OJ L 108,
271
The Commission is taking legal action against those late-implementing or late-reviewing states as
we write in May 2005.
79
com petition policy is facing huge challenges in im plem enting com petition in the m edia. Cave
gives an account of m edia com petition policy that m ay give rise to som e pessim istic reflection
three years into the process of im plem entation of the 2001 Directive package: Cave outlines the
failure to deal with the problem s with bundling of the m ost valuable live football rights, and
sim ilar difficulties in conditional access and other m arkets. I place the problem of com petition in
the context of the behaviour of m arket dom inant players in hindering infrastructure com petition
and developm ent272. U ntil these problem s are resolved, we will not arrive at a m ulti-channel
‘utopia’ of choice.
Is com petition policy actually the m ost suitable response to regulatory questions
involving m edia and especially broadcasting? Som e challenge the econom ic policy orthodoxy that
com petition is the m ost appropriate tool to regulate European m edia, and exam ine som e of its
im plications in recent regulatory policy. G ibbons, in a political analysis of OfCom ’s intentions
and philosophy, questions why econom ic analysis of the m edia should be the predom inant
them e, when dem ocratic pluralism and cultural heritage are rival them es which do not necessarily
com plem ent com petition analysis273. G ibbons raises questions not only about the perform ance of
com petition policy in the service of narrow aim s of increasing com petition and prom oting
m arket entry, com pared with the broader social welfare associated with broadcasting.
Com petition objectives, m ethods and term inologies com e into play alongside other regulatory
tools.
I challenge the potential of com petition when the industry needs radical structural
reshaping of existing m onopolies before any effective com petition policy can be introduced at the
m icro-level. Those that stress the fundam ental shift to com petition regulation tend to stress
connectivity: U ngerer outlines the shift from analogue – the ‘Hertzian age’ - to the ‘age of nearly
unlim ited channel capacity’274. The content problem s on the other hand such as the key role that
prem ium content plays in this world, and the enduring econom ies of scale in key quality
production sectors including news, will in this vision be dealt with by the com petition and access
fram ework. U ngerer’s view is that the rise not only of m arkets but of highly com plex vertically
integrated m arkets in broadcasting m akes this new role for com petition a necessity.
272
Cave, Martin (2005) Competition and the exercise of market power in broadcasting: a review of
recent UK experience 7 info 5 at pp. 20-28
Marsden (2005) Free, Open or Closed – Approaches to the Information Ecology 7 info 5 at 6-19
273
See Gibbons (2005) Competition policy and regulatory style – sigues for OfCom, 7 info 5 at 42-51
274
See Ungerer, Herbert (2005) Competition in the media sector - how long can the future be delayed?
7 info 5 at pp. 52-60
80
Though the focus of this chapter is the U K m arket, which has fallen behind in fixed line
telecom s developm ent but still leads in the transition to digital terrestrial broadcasting275 (at least
am ongst larger European m edia ecologies), the lessons are applicable to overall European policy.
OfCom is trying to regain the regulatory initiative with its strategic reviews into both public
service broadcasting and fixed line telecom s in 2004-5, which will conclude in 2006. How
European com petition policy responds to the changes in this highly strategic networked sector
will be crucial to its em ergence as an Inform ation Society, and its response to the challenges of
achieving som e parity with m ore advanced N orth Am erican and East Asian inform ation
econom ies. W hether it can achieve this and m aintain the European socio-cultural m odel depends
in large part on its response to the enduring policy questions analyzed.
4.5. The N ew Experi
Exp erim ent: Telecom Regulation and Co-
Co -Regulation
The outcom e of the three-part OfCom Telecom Strategic Review (TSR) was a negotiated
settlem ent that resulted in British Telecom separating its regulated national wholesale telecom
network into a separate subsidiary called ‘Open Reach’, which began activities in January 2006276.
This radical dism em bering of the form er incum bent’s access lines business was undertaken
voluntarily by BT. In truth, any attem pt to enforce this action via a referral under Section 2 of
the Com petition Act 1998 would have required a political intervention by the incom ing
Secretary of State for Trade and Industry after the May 2005 G eneral Election, a referral to the
Com petition Com m ission for a full sectoral enquiry, and potentially judicial review of the
outcom e. This would have prevented any rem edy until at least 2009, and exhausted the
regulator’s resources.
A precedent for m uch lesser action had been the decision to attem pt to regulate the wholesale
cost of calls from fixed line to m obile telephones, in which an investigation begun in 1997
resulted in a Com petition Com m ission referral in 2001, a year-long inquiry in 2002, a judicial
review in 2003 and finally a rem edy in 2004, though retroactively applied. This seven-year cycle
of action is fam iliar to other national regulators, and for som e the period from European
liberalisation in 1998 until the deadline for im position of new regulation in 2005 was a seven-
year period in which no binding court decisions were enforced against the regulator. W ith appeal
to higher courts and all the legal and econom ist resources available to the incum bent, it is little
surprise that G erm any and the U nited States have seen regulatory im passes result from their
respective 1996 telecom s legislation, as Marcus indicates.
275
See Noam, Eli M. et al (ed.), 2003, Television over the Internet: Implications for Networks,
Infrastructures, Strategy, Policy and Content, Lawrence Erlbaum
276
See ofcom.org.uk for details of the Review and its outcome.
81
The aim s of telecom s policy are radically shifting in two respects:

• The m ove towards V oice over Internet Protocol m eans that revenues from calls are
falling rapidly, and that access pricing is therefore even m ore critical. The security and
consum er safety im plications of this m ove are attracting significant controversy as
Marcus describes.
• The incum bents are rapidly building out all-IP networks, described as ‘N ext G eneration
N etworks’ (N G N s)277, which typically have far less ‘Points of Presence’ (POPs) at which
to interconnect with com petitors, than the traditional telephone exchanges (in the U K
125 POPs as opposed to 6500 local exchanges). As a result, the network typography will
change very rapidly and com petitors will need to change their network investm ent to
m atch the typography offered by the incum bent.
The U K regulator has adopted a co-regulatory approach to negotiating the term s and conditions
of interconnection to the U K N G N being built for 2010 by BT. The OfCom use of a co-
regulatory structure for standard-setting in the interconnection regim e follows an old pattern: the
first telegraph m essage sent along the G reat W estern Railway's com m unications system in 1847 -
which caught a m urderer as he fled from W indsor to Dover to escape to leave the country - was
interoperable with other system s by use of Morse Code278. The European Telecom m unications
Platform - an advisory body to the European Com m ission states that: [the U K ] 'could be an
appropriate m odel for deploym ent elsewhere. Due to its far-reaching consequences for the
regulated firm the appropriateness of the m odel elsewhere has to be evaluated very carefully'279. It
further explains "U K interconnection standards have previously been produced by the N ICC
(N etwork Interoperability Consultancy Com m ittee). The N ICC bases its work on that of ETSI.
W hilst there are on-going discussion on how N ICC should evolve to be suitable for the evolving
m arket place, N ICC has em barked on a program m e to develop N G N interconnect standards.
The first phase of this program m e is to develop IP Interconnect specifications (based on ETSI
standards) to support PSTN /ISDN services." N ICC is a very technical com m ittee, heavily
preponderant towards the best-resourced m arket actors, especially the incum bent.
It is therefore rem arkable that a solution was achieved in the U K to interconnection, with the
voluntary agreem ent of the incum bent. In the Microsoft litigation, it was suggested that the
277
See www.itu.int for details of the NGN Working Group.
278
See Standage (1999).
279
See ETP 2006 (01) On the technology, business models and regulatory aspects of NGN
http://www.etp-online.org/downloads/06_01_ETP%20public%20NGN%20final.pdf ETP was formed
in 1998 and combines "Open Network Provision Co-ordination and Consultation Platform (ONP-CCP),
founded in 1991, and the European Interconnect Forum (EIF)" http://www.etp-
online.org/html/masterset_public.asp
82
diversion from innovation produced when an incum bent chooses to deliberately raise entry
barriers rather than attem pt a Schum pterian innovation-led com petitive strategy was preventing
the m ost efficient outcom e for the com pany, the m arket and the consum er280. In this case, the
telecom s incum bent appears to have (grudgingly) accepted the case for innovation and separation
of the regulated access business from its value-added retail businesses. This m ay set a precedent
for other m arkets where regulators have sufficient intellectual and political capital, and
incum bents sufficient stim ulus to m ove away from the regulated business. It does only provide an
extrem ely partial structural solution to the various institutional barriers preventing real
com petition in both access and content businesses.
In the following chapters we will see that in Internet content, in control over root servers and
dom ain nam es, and in the standards and underlying logical protocols that underlay the physical
transport layer (telecom s) that this chapter discussed, there is m uch that is institutionally set to
prevent real com petition em erging to the existing order.
280
Also see discussion in Gifford, Daniel J. & David McGowan (1999) A Microsoft Dialog, 44
ANTITRUST BULL 619; Lemley, M. (2000) ‘Will the Internet remake antitrust law?’, Chapter 12 in
Marsden, C. (ed) Regulating the Global Information Society, Routledge, New York; Lessig, L. (1999b)
Brief as Amicus Curiae U.S. v. Microsoft, 65 F.Supp.2d 1 (D.D.C. 1999) (No. Civ. 98-1232 (TPJ), Civ.
98-1233 (TPJ) at < cyber.law.harvard.edu/works/lessig/AB/abd9.doc.html>,
Liebowitz S.J. and Stephen E. Margolis (1999) Winners, Losers & Microsoft: Competition And
Antitrust In High Technology; Marsden, C. (2001) Cyberlaw and International Political
Economy:Towards Regulation of the Global Information Society, 2001 L.REV. M.S.U.-D.C.L. 1 at
pp253-285; McGowan, David (2000) ‘The Problems of the Third Way: A Java Case Study’, Chapter
13 in Marsden, C. (ed) Regulating the Global Information Society, Routledge, New York.
83
Inter--G overnm ental Regulation of the Internet

Chapter 5: Renewed Inter
“W e are at an inflection point, a revolution point. W e m ight just be at the point where the
utility of the Internet stalls -- and perhaps turns downward. My goal in calling for a fresh
design is to free our m inds from the current constraints, so we can envision a different
future. The reason I stress this is that the Internet is so big, and so successful, that it seem s
like a fool's errand to send som eone off to invent a different one."
David D. Clark, Massachussets Institute of Technology and IETF (2005)281
5.1. Introduction: Is Internet regulation broken?

An unstoppable force is colliding head-on with an im m ovable obstacle. Legitim ate, accountable
bureaucratic governm ents are standing in the way of the dynam ic, innovative, laisser-faire
Internet ‘netizens’. After a honeym oon period in which governm ent acted as enabler of Internet
developm ent across boundaries and encouraged its increased use, security and com petitive
provision, the new world order post-9/11 has resulted in a renewed attem pt by governm ents to
im pose control on the newly pervasive Internet282. W ith a billion citizens online, and over a
billion soon to join via m obile telephones, both developed and developing countries are asserting
national sovereignty over Internet content, allocation of Dom ain N am es, and design. From
Australian and French governm ent intervention to prevent hate speech and pornography online,
to the ‘G reat Firewall’ of China designed to stop the global Internet contam inating the one-party
dictatorship in the world’s m ost populous nation, to U nited States anti-piracy rules that
crim inalise children and innovators, to attem pts to preserve local cultures in the face of the global
281
See Talbot, D. (2005) ‘The Internet is Broken’ Technology Review, 19 December 2005,
http://www.technologyreview.com/InfoTech-Networks/wtr_16051,258,p1.html
Research Fellow, Programe on Comparative Media Law and Policy, Oxford University. Email:
ctmarsden@ yahoo.co.uk
I thank Richard Higgott and Ben Zissimos, formerly of the Warwick Centre for the Study of
Globalisation and Regionalisation, for collaboration on the project from which this paper developed,
Debora Spar, Eli Noam, Stuart Schwarzstein and Jonathan Aronson for appropriately worldly insights
into the impact of technologies on business and politics, Michael Hodges and Susan Strange for earlier
inspiration, and colleagues on the board of ‘info’, the Working Group on Internet Governance, Adam
Peake, Jeanette Hofmann and others, and at the Harvard Ruschlikon annual seminars from 2001-5 for
current inspiration. All errors and omissions remain my own.
282
Though Mueller argues that property rightsholders were already using the trademark resolution
system to reassert control: Mueller, Milton (2000) Technology and Institutional Innovation: Internet
Domain Names, 5 International Journal of Communications law and Policy at
http://www.ijclp.org/5_2000/ijclp_webdoc_1_5_2000.html
84
onslaught of the ‘Californication’ of values283, the siege of the Internet has begun. W e can date it
to the first stage of the W orld Sum m it on the Inform ation Society in Decem ber 2003, a
withdrawal from prior laisser-faire attitudes established in the G 8 Bonn Sum m it of 1996284.
As the im m ovable object determ ines its strategy towards the Internet, we can pose a series of
governance questions. G overnm ents intend to tam per with the fastest growing and m ost powerful
inform ation resource in the history of the world, a seam less end-to-end web of com puters, data
and people. W ill they find a new flexibility in their bureaucracies to cope with the evolution of
this extraordinary organism ?
1. Can the governance response be dynam ically legitim ate and can the Internet be
perm itted to continue its legitim ate dynam ism ?
Problem : governance by legislation and international treaty involves a six-year policy cycle, and
six years prior to W SIS, in 1997, broadband was still a laboratory experim ent. Em bracing the
speed of Internet developm ent is a m ajor problem for largely static governm ent bureaucracies.
2. Can the governance response be innovatively accountable, and will the Internet pioneers
be accountably innovative?
Problem : governm ents derive their legitim acy from periodic general elections, approxim ately
every four years, from which they build coalitions of interests at national and international level
to regulate in the interests of the m ajority. Four years prior to W SIS, in 1999, the global Internet
population was only 100m illion, m ainly in N orth Am erica. In 2006, it is a billion, with the
largest constituency in Asia. Representation in governance is therefore an enorm ously challenging
task. On the other hand, the builders of the Internet were a group of m iddle-class white m ales
from Anglo-Saxon cultures with a clearly libertarian political ethos, and they continue to govern
the architecture of the Internet, no m atter how ethnically, sexually and age-diverse its growing
population has becom e.
3. Can the governance response be a benevolent control, and can Internet users be
responsibly free?
Problem : governm ent is rules-based, and the bureaucratic m ethod is to system atize the object to
be controlled within bureaucratic definitions: what system s theory after N iklas Luhm an describes
as an autopoetic self-referential system 285. In brief, governm ent is a ham m er, and to a ham m er
everything else looks like a nail. For governm ent to learn how to regulate reflexively, in the
283
A reference not only to the state of incorporation of ICANN, but to the cultural influence of
Hollywood music and movie majors: see Lessig (2004) for examples.
284
See Chapters 1-2 for narrative.
285
See Teubner (1986).
85
expression of pioneering regulatory theorists Braithwaite and Ayres, is a critical test of their
com petence. Internet users m eanwhile, expressing all the freedom but none of the responsibility
of this globalising network, have filled it with spam , porn, hate speech and pirated m edia. Again,
to the anonym ous Internet users, all control appears to be a ham m er blow to their freedom s,
when in fact som e responsible self-regulation is necessary in any public space. Sim ilarly m edia
com panies try to regulate online user behaviour with legal restrictions: legal code trying to defeat
or dim inish software code. The key test for users, governm ents and corporations is to arrive at
acceptable joint solutions to produce innovation and dynam ic freedom with greater
accountability and legitim acy to society. Hacking, flam ing, spam m ing and surfing for sm ut are
no longer solitary anonym ous and am oral activities, if they ever were.
5.2. Five W aves of Internet Regulation, and U nited States Leadership
Internet regulation has been the subject of several waves – or policy cycles speaking m ore
technically. Cam p previously stated these in 2000 as: technocracy-adhocracy-bureaucracy, but
since there have been two further waves – ‘m ultilateral kleptocracy’ (as the U nited N ations is
often characterised) and now gerontocracy (the old succeeding the young, both inside the
Internet’s self-regulating instititutions and outside, via political control). More explicitly the latter
two are the regaining of m ultilateral sovereignty and legitim acy, followed by the increasing
political coalition building around the ‘deradicalisation’ of the Internet’s control and design.
If this is correct, then the current stage, which grew from the previous three, is in its relatively
early stages. N ote that in the years 2001-5, the Internet grew extrem ely rapidly not so m uch in
sheer num ber of users (in any case an im possible figure to obtain with any accuracy) but in
num ber of broadband connections.
86
Table: Five W aves of Internet Regulation in Developed Countries
Descriptive Label Dates Milestones K ey Policy For a and Decisions

Technocratic U ntil 1962: Architectural IETF, W 3C, IAB form ed initially as
1993 principles by Paul Baran academ ic partnerships
1969: ARPAN et
1989: W W W developed
by Tim Berners-Lee
Political 1993-5 1993: AOL founded Bangem ann Report 1994,
pre-em ption 1994: N etscape browser Clinton ‘vision thing’
released Com m unications Decency Act 1996
1995: N etscape Initial (January),
Public Offering K ey escrow encryption,
floatation EU 1995 Privacy Regulation
1995: Microsoft Internet Soete ‘bit tax’ April 1996
Explorer bundled into
W indows Office suite
Business-oriented 1996-8 1996-2000: Internet Bonn G 8 accord Septem ber 1996,
regulatory ‘dot-com ’ bubble W 3C PICS 1995-8
forebearance and 1996: Telecom s Act U S A CLU v Reno 1995,
‘deregulatory creep’ results in widespread European ‘Convergence G reen Paper’ 1997
m arket entry and debates
1998: European G IIC July 1995-2000 (approx.)286
telecom s liberalisation ICAN N form ation 1998
results in widespread DCMA 1998
m arket entry ‘Safe harbour’ EU -U S privacy agreem ent
Status quo and 1999- 1999-2000: W ave of ITU Telecom 99, October
internationalisation 2001 global G 8 Digital Opportunities Taskforce July
telecom /Internet/m edia 2000
m ergers, biggest AOL- U N ICT Taskforce March 2001
Tim eW arner 2000 ICAN N election October 2000
2001-2: Technology and E-Com m erce Directive 2000
stock m arket bubble G BDe 1999-2002
bursts
Public safety 2002-4 9/11/2001: global Patriot Act 2001,
regulation security paradigm Microsoft security im pulse,
generational shift CoE Cybercrim e Convention 2001
2002-4: growth of Data Privacy Regulations 2002
viruses, spam and 2003 CAN -Spam Act
‘botnets’
‘Convergence’ into 2004- 2004: G oogle IPO W SIS (Dec 2003, N ovem ber 2005)
existing regulatory 2005: Stock m arkets EU debates reform ing ISP liability 2005
paradigm s recover to 1998 levels London Action Plan on spam 2004
CoE Cybercrim e Convention Part II 2004
286
http://www.giic.org/about/faq5.asp
Its achievements in 1995-8 were stated as:
the development of guiding principles on a variety of global policies and applications that helped
define the GII;
a focus on the role of the private sector on issues relating to investment, market access, standards,
technology and applications.
the engagement of developing country private sector leaders on GII issues;
interaction between public officials and the private sector, bringing expanded dialogue and
understanding of diverse perspectives and solutions; and
the redefinition of the roles of international organizations in the development of the GII
87
and Senate ratification 2005

U S ‘duopoly’ telecom s policy 2005
Here, the residential broadband figures becam e extrem ely international, and in absolute term s
N orth Am erica lags behind other world regions.
Further, the ubiquity of broadband connections in m any countries m eant that the term
‘Inform ation Society’ began to have a descriptive rather than predictive or polem ical m eaning.
N ote that the U SA ‘hom e of the Internet’ is outside the top 15 countries in penetration at end-
2004.
In a world without borders, the hyperglobalized individual is as m uch a fact as is the ‘hyper-
power’: the U nited States287. These global citizens, spam m ers, phishers, pornographers, bloggers,
287
It is claimed that the US is the world’s first ‘hyper-power’ – in a unipolar interconnected world, the
US is the superpower without equal in the history of the world, in military, financial, corporate, cultural
terms. Even the British Empire, Alexander of Macedonia’s Empire, the Roman Empire, were
counterbalanced by others, including the Chinese Empire, Germanic political groups, Russian or
88
illegal peer-to-peer file-sharers and legitim ate Internet users raise the greatest global security and
developm ent challenges of the 21st Century288.
5.2.1. ICTs, Regulation, and G lobalization

In assessing the im pact of ICTs on globalization, we assess the intersection of science and social
sciences, to attem pt to bridge the greatest trans-disciplinary gulf in academ ic analysis. It is
unsurprising that there have been relatively few such attem pts, and a basic finding is that social
science literature has relatively neglected:
• study of international (regional and global) econom ic, social, political and legal
phenom ena;
• the dom estic response to such international (and therefore) exogenous restraints;
• the role of technology in the internationalization of production, finance, inform ation

flows, and security structures;
• the step change in international integration of econom ies, nations and societies caused in
the past three decades by com puter-m ediated inform ation transfer, satellite and fibre-
optic transm ission and digital switching-com pression technologies, and latterly the
ubiquitous penetration of the Internet and m obile telephony into the developed sections
of the global econom y and society.
The om ission of research from nationally oriented agendas due to funding and resource
constraints, is com pounded by the disciplinary gulf between social scientists and com puter
scientists. There are therefore both financial resource and knowledge resource gaps.
The research gaps identified above (crudely: parochialism , econom ic disparities, knowledge
disparities, neo-Luddism or technophobia289, the neglect of the international study of social
im pact of ICTs in consequence, and the Am erican challenge) are broadly those which can be
identified in the ‘real’ international political econom y. In this, academ ic research reflects the
Mongol power. The term is from the French ‘hyper puissance’ in use since at least 1998, see Hubert
Védrine (1998) Jeune Afrique, 24 February, quoted at footnote 6 in Nossal, Kim Richard (1999) Lonely
Superpower or Unapologetic Hyperpower? Analyzing American Power in the Post-Cold War Era,
South African Political Studies Association, Saldanha, Western Cape at
http://post.queensu.ca/~nossalk/papers/hyperpower.htm
288
This conceptualized dilemma has recently been explored in-depth in for instance Ball, K. and
Webster, F. (2003) The Intensification of Surveillance: Crime, Terrorism and Warfare in the
Information Age, Pluto, London.
289
For a robust defence of neo-Luddism, and a critique of ICTs creating a ‘global panopticon’ of
control, see Robins, K. and Webster, F. (1999) Times of the Technoculture: Information,
Communication and the Technological Order, Routledge, London.
89
policy arena. I therefore address these them es in both assessing the current state of knowledge,
and in assessing the ‘state of play’ in international governance term s290. I identify:
• G LOBALIZ ATION : Continued under-estim ation of the linkage of international

developm ent to the national policy arena, no longer tenable in the face of international
com m unication flows;
• EXCLU SION AN D ACCESS: An exponentially widening gulf in resources between

sm all clusters of financial plenty (e.g. the hyper-globalized bankers of the City of London
or Manhattan) and broad swathes of the excluded even in their own societies: the
‘working poor’ and non-em ployed;
• LIN K AG ES IN COMMU N ICATION TECHN OLOG Y: A gulf between scientific and

technological realities in which interlinked global production, finance and knowledge
structures are m ore pervasive than ever in hum an history (consider the phenom enal
growth of both Internet access and m obile phone penetration) and an academ ic denial of
the irreversibility of globalization;
• HYPERPOW ER: An international econom ic order in which U S-based corporations

have becom e so dom inant in world ICT trade, that the early 1990s predictions of the
decline of U S hegem ony, the growth in stature of Japan and the EU , have been proved
(decisively) false.
• HYPERG LOBALIZ ED IN DIV IDU ALS: A broader societal view of globalization and
ICTs, in which the word ‘globalization’ has been dem onized, yet where the student
generation travels, em ails and m anipulates global cultural artifacts m ore ubiquitously
than all but utopians could anticipate only a decade ago – with occasionally deadly
results as Muslim extrem ists educated in W estern European and Pakistaniuniversities
tragically dem onstrated;
• FOCU S AN D TARG ET: This ‘hyperpower’ provides the focus for global social
injustices and youth rebellion energy: just as the 1960s provided Fidel Castro, Che
G uevara, Ho ChiMinh and Abdul N asser, so the 2000s provide Bin Laden,
Com m andante Marcos, and Saddam Hussein.
The constellation of these forces is represented diagram m atically in the Figure below. The
double-edged sword of globalization, m ost shockingly represented in the digital television
290
See generally Marsden, C. [ed] (2000) Regulating the Global Information Society, Routledge, NY.
90
coverage of terrorist activities in both W estern capitals (London, Madrid, N ew York) and Arab-
Muslim cities (Istanbul, Sharm el Sheikh, Ram allah, Baghdad) has recently begun to dom inate
consideration of the effect of com m unications-driven globalization.
Hyper
Power:
Financial,
Military
Technology
ICTs Hyper- Exclusion

Information globalized and Access:
Linkages Groups v. Perceived
Hyperpowe Injustice
r
Globalizati
on: faster,
deeper,
wider and
profound
Previous waves of technological innovation have increased com m unication possibilities for non-
state actors, from the G utenberg Press to the telegraph to telephony291. N ational governm ents
have succeeded in preventing contam ination of their system s by globalized individuals through
m ore or less crude censorship of the networks carrying these technologies. The privatization and
globalization of technologies such as m obile phones, satellite transceivers and especially the
Internet m ean that the phenom enon of foreign cells operating in-country is far easier than ever
before, raising huge new problem s for both the controllers of inform ation (the state), the citizens
living with these new freedom s and threats, and developing states with no tradition of
inform ation freedom and therefore a m uch greater task in encouraging dem ocratic civil society292.
291
For an excellent comparative business history view, see Spar, D. (2001) Pirates, Prophets and
Pioneers: Business and Politics Along the Technological Frontier, Random House, NY.
292
Political economy literature provides some evidence of cultural differences from multinational
management studies, for instance Trompenaars, A. and Hampden Turner, C. (1997) Riding the Waves
of Culture: Understanding Cultural Diversity in Business, Nicholas Brealey, London; Trompenaars, A.
and Hampden Turner, C. (1993) The Seven Cultures of Capitalism: Value Systems for Creating Wealth
in the United States, Japan, Germany, France, Britain, Sweden, and the Netherlands, Doubleday,
London; Ohmae, K. (1991) The Borderless World: Power and Strategy in the Interlinked Economy,
Harper Business, NY.
91
5.2.2. Researching ICTs and W hy History Matters: Path Dependency

W hilst the spread of the Internet is usually interpreted in a broadly positive light, significant
questions and concerns rem ain. Param ount am ongst these is the global dom inance of U S online
content and standard setting. This has been criticized in a num ber of respects. On an econom ic
level, it has led to large-scale redistribution of telecom m unications revenues from developing
nation com panies to U S carriers. The enduring dom inance of the content itself is seen by
Schiller to be driving the cultural hegem ony of export-oriented U S m edia products and
consequently social values293. Although there is political econom ic investigation of ICTs per se, it
begins from a prem ise largely founded in sociology and cultural studies, with a norm ative base of
freedom of com m unication and social relations, rather than assignm ent of intellectual property
rights [IPRs].294 Mansell and Steinm uller have explored ICTs and globalization issues m ore fully
from a political econom ic standpoint295, with m uch useful work on international standard setting
conducted by political econom ists296.
W riting about ICTs and globalization over five years since the stock m arket peaks of m id-March
2000 - after the party is over - is a useful exercise in perspective. This is not only because too
m uch ‘globaloney’ was talked during the 1990s foreign direct investm ent ‘gold rush’ into the
Asian Tiger econom ies, m ass privatizations, the Latin Am erican revival (prior to the Argentine
tragedy). It is not only because the ‘e-enem a’ applied by the pricking of the speculative bubble in
Internet, telecom s, m edia and technology stocks has rem oved the ICT euphoria. It is also not just
because the sites of resistance to ‘globalization’ – i.e. unfettered free m arket capitalism – were
evident in the 1990s in for instance the Z apatista rebellion in Mexico, Taliban Afghanistan or the
failed states of equatorial Africa (Sierra Leone, Som alia, Rwanda, Dem ocratic Republic of Congo,
Liberia and by extension Haiti). The rise of fundam entalist terrorism was the latest, m ost
dram atic episode in a cycle of m odernizing resistance to W estern-im posed hegem ony in the
newly post-Marxist, post-colonial, post-nationalist, and Islam ic worlds which stretches back
centuries, and in its international form at least to the wave of anti-im perialist m ovem ents which
sparked the G reat W ar of 1914 (for instance, Com m unist, Young Turk, Boxer rebellions in
293
Schiller, D. (1999) Deep Impact: The Web and the changing media economy, 1 info 1 at 35-52.
294
Mansell, R. and Roger Silverstone, R. eds (1996) Communication by Design: The Politics of
Information and Communication Technologies, Oxford University Press, Oxford. Also W.H. Dutton ed
(1997) Information and Communication Technologies: Visions and Realities, Oxford University Press,
Oxford. W.H. Dutton ed (1998) Society on the Line: The Information and Communication Revolution,
Oxford University Press, Oxford.
295
Hawkins, R. R. Mansell, W. E. Steinmueller (1999) Towards Digital Intermediation in the
Information Society, Journal of Economic Issues, Vol. XXXIII (2): 383-391.
296
Blackman, C., M. Cave, P.A David (1996) The new international telecommunications environment.
Competition, regulation, trade and standards 20 Telecommunications Policy 10 at 721 – 724.
92
respectively Europe, the Ottom an and Chinese em pires). Since the 9/11 and 3/11 attacks on
N ew York and Madrid, it is clear that the downsides of network effects in globalization are
affecting m arkets as m uch as the giddy heights of the ‘dot-com ’ bubble positive network effects of
the late 1990s.
The wheels have been com ing off the latest version of globalization for several years. I argue in
com m on with m any that it is the U S failure to lead a stable N ew W orld Order which has caused
this297. As U S elections are decided in large m easure by financial interests in fund-raising for
candidates, and in the advertising on television stations, and control of news agendas on those
sam e stations298, it is therefore private forces which control U S politics, and increasingly dom inate
the econom ic (if not social) agenda in the European Com m ission, and foreign-investm ent and
export dom inated econom ies in satellites, in East Asia, Latin Am erica, W estern Europe, and W est
Asia. The influence of the dom inant hyper-power is greater than ever before, driven by ICTs.
The thesis that the U S has failed to exercise its leadership role effectively, as evidenced by the
breakdown of the ‘Seattle Round’ of world trade talks in 1999, the failure of the IMF and W orld
Bank to address poverty in highly indebted countries, the m arginalization of the U nited N ations
after U S withdrawal from the K yoto Accord, predates the G eorge W . Bush presidency. W hat I
hope to contribute by two case studies is how ICTs, an agent for transform ing power relations in
the international political econom y (IPE), have instead contributed to a new digital divide which
exacerbates the N orth-South Divide m ost fam ously described by the Brandt Report of 1972 and
the U N G eneral Assem bly declarations of a N ew International Econom ic Order (N IEO) and
N ew W orld Inform ation and Com m unications Order (N W ICO) of 1964 and 1982 respectively.
5.3. Internet governance at an Inflexion Point

The Internet as we know it has reached a crossroads. It is no longer a U S-dom inated m edium and
has reached the status of a critical infrastructure. Around the world approxim ately a billion users
now rely on the Internet for a wide range of social, political and com m ercial transactions. But
alongside this change in size and reach the Internet appears to be increasingly vulnerable. This
puts into question the highly inform al and hyper com plex governance m odus operandi the
Internet relies on. A m ultitude of actors, standard m aking bodies, international institutions and
297
For instance Soros, G. (2004) The Bubble of American Supremacy, but much earlier and
continuously since, Weidenfeld and Nicholson, NY, Vidal, G. (2003) Dreaming War: Blood for Oil
and the Cheney-Bush Junta Vidal, G. (2002) Perpetual War for Perpetual Peace: How We Got to Be
So Hated, Causes of Conflict in the Last Empire, Clairview; Vidal, G. (1994) United States: Essays
1952-1992, Abacus, NY; Vidal, G. (1969) Reflections Upon a Sinking Ship, Heinemann, NY.
298
Terry M. Moe (1997) The Positive Theory of Public Bureaucracy in Dennis C. Mueller (ed)
Perspectives on Public Choice: A Handbook, Cambridge University Press, Cambridge, at pp456-9.
93
com m ercial enterprises are involved in m aintaining the Internet as we know it, but can they also
reply effectively to an increasing dem and for a m ore secure and reliable Internet?
In the afterm ath of the first phase of the W orld Sum m it on the Inform ation Society com peting
visions of the future of Internet governance begin to em erge and will create new policy paradigm s
guiding the future of the Internet. The questions are not new, but will need to be answered with
m ore urgency. W hich parts of the Internet need to be regulated:
• G lobally, regionally, plurilaterally, nationally and locally,
• In public-private m ulti-stakeholder partnerships,
• Bottom up or top down,
• V ia governm ental regulation, oversight or via self-regulation.
• Do we need to rethink old rule m aking procedures and establish novel form s of
governance?
In the recent past m ost notably inform al technical bodies like the Internet Engineering Task
Force, the W 3C, or m ore recently ICAN N , the non-governm ental California corporation
charged in 1998 with m anaging critical parts of the Internet's infrastructure299, have tried to
coordinate the evolution of the Internet and have m aintained a rem arkably open, decentralized
and globally interoperable com m unications system . All these rather inform al organizations300 are
under increasing pressure from big m edia conglom erates, software giants, governm ents, civil
society and intergovernm ental organizations like the ITU to form alize the governance of the
Internet.
This section of the chapter attem pts to lay out the m ain challenges the Internet faces and explore
what style of governance seem s to be the m ost appropriate. It will attem pt to better delineate
different layers of Internet regulation and try to define accordingly what governance approaches
they require.
5.3.1. Innovation, Politics and G overnance

W hat is wrong with the Internet? W hy does it need different governance m odels? Calls for
regulation have focused on two problem s:
• its dom inance by the hyper-power U nited States, and
• m oral panics associated with illegal and/or im m oral content.
299
For European reaction, see Select Committee on European Scrutiny (2000) 21st Report, Annex 7,
Management of the Internet, Commission Communication COM (99) 02 at
http://www.pubications.parliament.uk/pa/cm199900/cmselect/cmeuleg/23-xxi/2310.htm
300
See Gould (2000) Chapter 11 in Marsden, Regulating the Global Information Society, Routledge.
94
These are two excellent reasons not to regulate, because unless harm is shown, the precautionary
principle is m ore im portant in Internet policy than others. It is ‘born global’, in the sense of
m ultinational com panies rather than truly global phenom ena, and therefore has been adopted
internationally with U S technological and social norm s. Accom panied by U S free speech
philosophy and to som e extent engineering to reinforce that end-to-end uncensored m ode (due as
m uch to its university origins), its overall effect is what critics m ight describe as a
‘Californication’ of the Internet.
So the Internet continues to be an exam ple of global deregulatory phenom ena, but is increasingly
relevant for students of international organizations, with the struggle over the ‘W ashington
consensus’ now taking place in the U nited N ations. This new com bination of m ulti-stakeholder
self-appointed consum er-citizen lobbies, and the challenge to U S hegem ony over Internet design,
is used by other states, especially those in the Security Council or denied such entry (Chins,
France, Brazil, India, the European Com m ission), to pressure for reform of the system .
It is noteworthy that the criticism s of the U S governm ent com e from both those who see it failing
to protect their fundam ental rights (particularly privacy in the whois directory and elsewhere
from rapacious copyright and e-com m erce com panies using the U S as their ‘client’ state) as well
as those who would prefer states to reassert som e collective governm ent at the expense of som e
individual rights (for instance, in allocating IPv6 addresses, in preventing hate speech, in
preventing terrorist speech). The U S is caught between a rock – its existing broadly deregulatory
position – and a hard place – perm itting policy to be set by a bureaucracy even m ore broken than
either W ashington or Brussels, nam ely the U N agencies.
My intention is not to m ap such a debate – that had been done exhaustively by others – nor to
broaden it to encom pass the debate in the U N as an exam ple of m ulti-stakeholdersim – again
exhaustively docum ented. Rather, I want to return to first principles: what is wrong with the
Internet? W hat regulatory structures can perform m ore efficiently and accountably than the
devolved heterarchical self-regulatory structures that currently exist? W hich governm ent or
com bination could as efficiently exercise the precautionary principle with the DN S (Dom ain
N am e System ) as that of the U S? U nless we can find a cure better than the disease, I suggest that
change is neither inevitable nor beneficial.
Two reform s have recently been proposed by governm ents: m ore U nited States G overnm ent
(U SG ) control by Michael G allagher in June 2005, and m ore international supervision and
oversight, by the U N W G IG in July 2005. W hich is m ore legitim ate? Do we judge that
95
legitim acy by its operational standards or by its procedural efficiency? Mathiason and Mueller
state301:
“An analysis of the [W G IG ] report suggests that W G IG has indeed m oved us further,
but we m ust now be m ore analytical if we are to know where to go. W G IG has m oved us
forward by clearly dem onstrating that there are Internet governance problem s that need
to be addressed. This consensus provides a starting point …it is clear that there is
unfinished business with respect to ICAN N ’s supervision that m ust be addressed.” (at 3)
They explain this succinctly in their response to the N TIA’s Statem ent of principles that included
a continuing oversight role of the Dom ain N am e System (DN S): “If nothing changes, the U S
role will continue to inflam e political criticism of Internet governance for years to com e.”302 They
suggest five associated links:
1. Set in m otion a tim e-lim ited process for negotiating a set of authoritative agreem ents
that will determ ine the basic principles and norm s for global Internet governance.
2. Focus policy research, dialogue and negotiation on the process through a m ulti-
stakeholder forum .
3. Review the progress of the process on a regular basis, with the involvem ent of
stakeholders.
4. As problem s that affect the security, stability and openness of the Internet develop whose
solution cannot wait, build them into the process.
5. Provide adequate secretariat support and use it to help focus and facilitate.”
The U S governm ent response to W G IG was relatively bland, supporting bottom -up private
sector-led for a generally, but including useful analysis of what they see as solutions to com plex
problem s303:
“Spam : Increasingly, spam is, in large part, a security issue: spam is one way in which viruses
and other security threats can be delivered to com puters. Industry m ust play a lead role in
developing technical tools to address this problem . In addition, m any of these security threats
often result from crim inal conduct. The Convention on Cybercrim e provides a
com prehensive fram ework to address these threats. In 2003, the U nited States enacted an
anti-spam law established a fram ework of civil and crim inal enforcem ent tools to help
301
See Mathiason and Mueller (2005) Quo Vadis? At http://dcc.syr.edu/miscarticles/IGP-quovadis.pdf
302
Mueller et al (2005) The Future US Role in Internet Governance: 7 Points in Response to the US
Commerce Department’s
“Statement of Principles http://dcc.syr.edu/miscarticles/IGP-USrole.pdf at 4
303
Comments of the United States of America on Internet Governance, 15 August 2005, at
www.state.gov
96
Am erica’s consum ers, businesses, and fam ilies com bat unsolicited com m ercial e-m ail.
However, the U nited States does not believe that the statute alone will solve spam . The
U nited States approach to com bating spam relies on a com bination of legal tools for effective
law enforcem ent, developm ent and deploym ent of technology tools and best practices by the
private sector, and consum er and business education. W e believe that work undertaken to
com bat spam should ensure that em ail continues to be a viable and valuable m eans of
com m unication. G overnm ents have a role to play in educating consum ers and enforcing
spam laws. To this end, governm ents should encourage spam enforcem ent agencies to join
the London Action Plan on international spam enforcem ent cooperation.” (at 3)
Clum sy U S governm ent reactions to international pressure have reasserted the control over
ICAN N and top level dom ain nam es exercised by the Departm ent of Com m erce304, with notably
political action to prevent the prior policy of establishing a ‘safe harbor’ for pornography with a
new ‘.xxx’ dom ain. In response the Internet G overnance Consortium , a leading academ ic
research group, has stated that “W hether intended or not, this intervention has the potential to
becom e a turning point in the history of Internet governance.”305
The European U nion has stated:
“That stability, dependability and robustness of the Internet rem ain a high priority;
security and spam are im portant issues in this field. A global com m on understanding of
the issue of Internet security m ust be developed. This includes the use of security policies
in general at all relevant levels.
W ith regard to spam there is a need to adopt com m on principles of action concerning
cooperation in this field. Anti-spam efforts should not be based only on legislation and
cross border enforcem ent, but also on industry self-regulation, technical solutions,
partnerships between governm ents and the Internet Com m unity, as well as awareness-
raising.
304
McCullagh (2005) US Reasserts Control over the Internet, 1 July – Asst. Com Sec Michael
Gallagher, “maintain its historic role in authorizing changes or modifications to the authoritative root
zone file” http://news.zdnet.co.uk/Internet/0,39020369,39206653,00.htm Reporting on
http://www.ntia.doc.gov/ntiahome/domainname/USDNSprinciples_06302005.htm at Wireless
Communications Association, 30 June 2005.
305
See http://dcc.syr.edu/miscarticles/STATEMENT-XXX.pdf
97
The im portance of ICTs for the com petitiveness of industry and therefore encourages
active involvem ent of the private sector in the Internet governance discussions during the
second phase of W SIS.”306
This suggests that m ost developed country governm ents agree the Internet regulatory m odel is
not fundam entally inappropriate but needs adaptation. Arguably, a less political solution to
technocratic solutions, rem oving oversight of the U S governm ent but avoiding its replacem ent by
the ITU or another U nited N ations body, would be m ost appropriate. An attem pt to reform the
Internet encom passing at least a passing interest in Regulatory Im pact Assessm ent is m ade by
Auerbach307. He concludes that various of ICAN N ’s politically captured functions can be
undertaken at lower cost out of political influence.
Auerbach also states that “The com m unity of Internet users is ignored by the W G IG report”308.
Two substantial failings of the U N process to date have been those of vested expert interests
which did not encom pass the Internet: the W G IG experts were either expert in international
telecom m unications policy-m aking (N OT engineering) or in international m ulti-stakeholder
negotiating. Both are valuable, but both dem onstrate a com m itm ent to reform ing the Internet to
m ake it m ore dem ocratically and bureaucratically accountable. N o m em ber of the Com m ittee
took a contrarian position309. In particular it was not possible to nom inate a civil society
representative who opposed the U N ’s role in Internet governance. On the contrary, m em bers
seem ed captured by the process, inasm uch as their m ain recom m endation was to extend the
group’s processes in various form s in the future. Below I set out a m ore technocratic scenario
planning exercise.
5.3.2. Three prelim inary conclusions:
1. Those who designed the Internet as a unique com m unication tool want it to stay unique,
those who are newbies to the Internet but earned their crust from extracting m onopoly
rents from old m edia want it to becom e like TV or m obile or any other regulated
m edium . Sym pathy is still with the Internet's pioneers - possibly because of bias in
favour of innovation over fossilisation.
2. Civil society is truly m obilised by the power of U N sum m itry - with advantages and
disadvantages. If they can drag governm ents over the dem ocratic coals, perhaps they can
306
ITU (2005) 1 August, WSIS-II/PC-3/CONTR/19-E, Initial comments by the European Union and
the acceding countries Romania and Bulgaria, on the report of the Working Group on Internet
Governance.
307
Auerbach, K. (2004) at http://www.itu.int/osg/spu/forum/intgov04/contributions/governance-
structure-ITU-Feb26-27-2004.html
308
Auerbach, K. (2005) at at http://www.cavebear.com/cbblog-archives/000188.html
309
To my knowledge based on expert interviews.
98
save the netizen libertarians from them selves by m aking the whole U N process too
nakedly obvious a powergrab by censorious governm ents. But that m ay be optim istic -
perhaps m uch m ore likely they are also a Trojan horse for governm ent intervention in
the nam e of bureaucratic legitim acy.
3. It's about the m oney and power, of course - the Internet m eltdown has fatally weakened
the entrepreneurial zeal of Internet pioneers in m ost countries outside U S and
Scandinavia - it's now all about ccTLD and telco m onopolies - so governm ent is not
strangling a new form of econom ic organization at birth but instead m oving in to
regulate (i.e. m onopoly tax and m aintain) established dom inant actors. There's no
libertarian tendency for governm ent to be m ystified and im pressed by any m ore. V oIP
prom ises to wreck their interconnection advantages from international telecom s so
plenty of reasons to be active - ICAN N as Trojan horse for ITU settlem ent rates.
The Internet forces us to think, act and com prom ise faster than any previous m edium - but
m ulti-stakeholderism is bound to slow the process by introducing civil society into the form erly
corporatist m ix. If that is neo-corporatism , perhaps it will grind along as slowly as corporatism
did310: if the Internet is a global public good, that can change its governance characteristics for
good - just as the environm ent and financial services. However, the global warm ing debate and
the AIDS debate show that such global public goods - technologically and globally determ ined -
have appalling consequences in public policy when populist politicians decide to ride the wave of
short term electoral success.
5.4. A Suggested Research Agenda
To set out a research agenda for future Internet policy. First, above, good regulation dem ands the
precautionary principle and a regulatory im pact assessm ent based on robust em pirical evidence.
However, even in the absence of regulatory reform , the Internet is sufficiently im portant in socio-
econom ic and public safety term s, that continual m onitoring m ust be undertaken. This should in
m y view take the form in part of future scenario planning.
This contribution is deliberately provocative: I have reservations about the U S First Am endm ent
absolutism which perm its N azi m em orabilia auctions on the Internet and arrogantly inform s
nation states that they cannot prevent their citizens partaking. As a Council of Europe consultant
on application of Article 10 of the European Convention of Hum an Rights to digital networks
310
See Grant 1998
99
including the Internet311, I properly see an interm ediary role for states in ensuring the widest
access for citizens to im partial and inform ative sources. However, I believe in the individual’s
right to self-determ ination consistent with the rights of others, not the N IEO and N W ICO
version that determ ines the states’ rights before its citizens. They can no m ore stop the Internet as
stop television, m obile telephones, individual am bition, or the tide. Creating a pluralist and stable
evolution to a hyper-globalized interdependent Internet planet is a huge policy challenge.
This leads to a real research project. How do we assess the critical future of the G lobal
Inform ation Society, in order to m ake predictions beyond the political expediency and warm
words of a U nited N ations Sum m it? The answer m ay be illum inated by specialist futures analysis,
described below.
5.5. Proposal to Create Internet Developm ent Scenarios and Futures

Futures planning – using scenario prediction, foresight networks and Delphi analysis312 as tools,
am ongst others - has becom e increasingly instrum ental in governm ent policy planning for the
m edium /long-term . These developm ents are m irrored in m any developed countries, and the
OECD International Futures Program m e docum ents these313. Futures planning – or Long Term
Policy Analysis (LTPA) – has m ade use of various techniques over the course of its developm ent.
It does not always intend to exercise policy predictions – in fact the longer term , m ore
interdependent and m ore com plex the environm ent, the m ore necessary networks of experts and
m ultiple predictive directions becom e to the flexible and adaptive planning of futures.
N evertheless, between the com ical m istakes which dem onstrate the com plexity314, predictive
qualities have been dem onstrated, especially in technology foresight. To take exam ples of
successful futures prediction, W are in 1967 predicted315 as follows:
311
See Marsden, C. (1999) Pluralism in Multi-Channel Digital Communications: Suggestions for
Regulatory Scrutiny, MM-S-PL [99] 12 def 2, available at IJCLP Web-Doc 5-4-2000 at
http://www.ijclp.org/4_2000/pdf/ijclp_webdoc_5_4_2000.pdf
312
Delphi, Dalkey, 1967, RAND P-3704: the Delphi technique of long-range forecasting separately
elicits and refines the opinions of a group of advisers without contact among them, and calculating a
statistical "group response." The procedure was designed to overcome the disadvantages common to
committees and small groups. The experts reply to written questionnaires or an online computer,
receive statistical feedback through formal lines of communication, and resubmit their estimates.
Where the response is a number (such as a date or amount), the most useful index has been the median
of the individual estimates. During the process, opinions do converge; where answers can be checked
against reality, it is found that the median response tends to move in the direction of the true answer.
Self-confidence is not correlated with individual performance, but the subgroup with the highest self-
ratings for competence will consistently perform slightly better than the group as a whole.
313
http://www.oecd.org/document/59/0,2340,en_2649_34219_2410299_1_1_1_1,00.html
314
As much Louis B. Mayer of MGM as Thomas Watson of IBM. Mayer famously said ‘Never make
predictions, especially about the future’. Watson predicted a global market for five computers, in 1945.
315
Ware (1967) The Computer in Your Future P-3626
100
“Com puting power will be readily available, like telephone service.

• Com puters will be sm all, powerful, inexpensive, and easy to use, and will accept natural
language and graphical input.
• Perhaps a broadband com m unication cable will carry all the inform ation services into
each hom e and workplace: voice, video, facsim ile newspaper, data transm ission, and
com puting service.
• Design prototypes and prelim inary m odels will be replaced by com puter sim ulation
studies.
• The com puter will touch m an in every part of his life, m odify his career, transfer blocks
of jobs from one industry to another, and force rapid and frequent change upon him .
• Retraining and reeducation will be the way of life for everyone except those retiring
before 1972.”
More fam ously, Paul Baran in 1962 predicted the design of the Internet316.
In addition to Delphi analysis and com putational em pirically forecast m odels, scenario planning
is also used to cluster predictive destinies, and to draw a fact-based narrative to engage
policym akers’ im agination in futures work. A type m uch used is XLRM – to use em pirical
analysis to construct a sm all num ber (m axim um three) of scenarios based on four elem ents:
• A set of key uncertainties (X
X factors) about the future under consideration
• Policy or strategy Levers (choices) for the m ajor stakeholders
• Relationships linking the uncertainties and the policy or strategy choices, and
• Measures of im pact.
The types of scenario planning can be grouped into two:
• norm ative scenario building and
• em pirically based projections into the m edium -long-term future.
Each has m erits and effective scenario planning is likely to incorporate elem ents of both. The
result is that the policy-m aker is able to separate knowledge of current policy analysis from a
belief system in the inherent virtue (and post hoc rationalisation of the logic system ) of that
developm ent. Futures scenarios should stretch the internalised logic of an organisation and
challenge it to pursue m ore lateral thinking.
Foresight planning based on system s analysis has been used for over 40 years in governm ent317.
The earliest RAN D reference is to a report on a pilot study of the prediction of social and
316
P. Baran. 1962 On Distributed Communications Networks. P-2626
101
technological events published in 1949318. The D elphi project which developed Delphi analysis
was begun in 1951. The Futuribles project in the early 1960s was a path-breaking exam ple319.
N eiswender wrote in 1967 of a conference320:
of a group of specialists whose interests are focused on a system atic exploration of the
future, their opinions on the advantages and disadvantages of life in the 21st century,
and their varying approaches to predicting and planning for the future. Am ong the
forecasting m ethods described is the Delphi technique. In the opinion of Realites the
tim e has com e to establish both national and international institutes for prediction and
planning.
Adoption of scientific futures planning by governm ent cham pions such as U S, Japan and France
resulting in what has becom e a ubiquitous deploym ent of futures planning at strategic levels.
These em ploy com puter m odeling and em pirical projections, which have resulted in advanced
m odels with diverging outcom es321.
However, that deploym ent has been criticized as lacking:
• em pirical focus on governm ental policy outcom es,

• holistic im pact analysis from socio-econom ic as well as technological bases, and
• rigorous cost-benefit analysis to evaluate the additionality of futures projects to policy
outcom es.
This criticism and the im provem ent of futures planning as a policy tool can be traced through the
three existent and two further iterations of technology foresight practised by the Foresight U nit,
as identified by G eorghiou322:
317
Brunner and Bewer 1972, Policy and the Study of the Future: Given Complexity, Trends or
Processes? P-4912 http://www.rand.org/publications/bib/SB1019.pdf). See further Quade 1969, The
Systems Approach and Public Policy, P-4053, Dalkey 1969, Predicting the Future, P-3948, Dror 1968,
Futures in Government, P-3909, Rescher 1967, The Future as an Object of Research, P-3593
318
A. Kaplan, A. L. Skogstad, M. A. Girshick (1949) The Prediction of Social and Technological
Events. P-93
319
P-3045 Futuribles. B. DeJouvenel. January 1965. An analysis of Project Futuribles, an international
venture which operated on the premise that the future is not a given unknown but one of many possible
futures. Futuribles gathered expert speculation as to possible future developments in various areas,
which were presented in 83 published essays.
320
R. Neiswender. February 1967 The Exploration of the Future [Realites] No.245, June 1966, Pp 50-
58.
321
Best-known are Kahn (1976)The Next 200 Years and Meadows (1972) Limits to Growth.
322
Georghiou, L. (2003) Foresight: Concept and Practice as a Tool for Decision Making, paper
presented to the Technology Foresight Summit, Budapest, March 27-29.
102
Figure: G enerations of Foresight Projects
Foresight Project Stage323 Im provem ents Made
1: 1994-8 Em pirical projections based on Delphianalysis. Led by

Technology foresight OST science inputs
2: 1998-2002 Includes im plem entation planning by m arket actors.

Industry Foresight Included governm ent stakeholders.
3: 2002- Includes socio-econom ic and political im pacts.

Stakeholder Foresight Cham pioned by departm ental sponsors
4: N ear future: N etworked and therefore responsive linkages of

Distributed N etworked Foresight foresight analysts em bedded in flexible networks in
sponsoring organisations
5: Medium Term : Integrated role in identifying drivers of innovation

Em bedded Innovation Policy inside organisations, with loose ubiquitous external
linkages between foresight practitioners. Full evaluation
m echanism s
G eorghiou characterised evaluation of the U K ’s Foresight program m es was “grand plans that
degenerated into piecem eal efforts.”
5.5.1. Objective of an Internet Futures Study

A study of Internet futures would have this m ain objective:
1. to assess the literature, m ethodological and policy docum entation that details future
Internet study, including:
• historically predictive analysis (such as com parisons between Internet and telegraph)
• Delphianalysis,
• technology foresight,
• group narrative (non-Delphi) including stakeholder (i.e. non-technical expert) and
• em pirical scenario m ethods,
a. both com putational and
b. XLRM-based
323
http://www.foresight.gov.uk/About_Foresight/The_Previous_Foresight_Rounds_A_Brief_History.html
103
c. U sing scenarios as an aid for gam ing exercises

in order to assess the state of the art in this field;
It is recognised that futures studies have to date produced inform al and form al networks of
educated ‘futurologists’, thus creating critical m ass for further integration of futures work into
departm ental and other large organisational strategy and im plem entation. The objective is –
absent these networks whose recognisable value is known and whose audit is m ade m ore difficult
by the voluntary m em bership of m any participants over tim e – to analyse m ethodological tools
m ore m inutely than by reference to networks and shared learning outcom es.
W e can posit the m ajor problem s to research below.
In the afterm ath of the first phase of the W orld Sum m it on the Inform ation Society com peting
visions of the future of Internet governance begin to create new policy paradigm s guiding the
future of the Internet. The questions are not new, but will need to be answered with m ore
urgency.
1. W hich parts of the Internet need to be regulated globally, regionally,
plurilaterally, nationally and locally? In public-private m ulti-stakeholder
partnerships? Bottom up or top down? V ia governm ental regulation, oversight
or via self-regulation?
2. Do we need to rethink old rule m aking procedures and establish novel form s of
governance?
3. W hat constitutes ‘innovation’ when referring to Internet developm ent?
4. Is innovation threatened by change to the original Internet architecture from a
global, anonym ous and end-to-end m edium ?
Is change necessary to protect the network from its inherent design flaws, such as unrestricted
anonym ity?
As the im m ovable object determ ines its strategy towards the Internet, we can pose a series of
governance questions. G overnm ents intend to tam per with the fastest growing and m ost powerful
inform ation resource in the history of the world, a seam less end-to-end web of com puters, data
and people.
1. W ill they find a new flexibility in their bureaucracies to cope with the evolution
of this extraordinary organism ?
2. Can the governance response be dynam ically legitim ate and can the Internet be
perm itted to continue its legitim ate dynam ism ?
104
Lag Problem : governance by legislation and international treaty involves a six-year policy cycle,
and six years prior to W SIS, in 1997, broadband was still a laboratory experim ent. Em bracing the
speed of Internet developm ent is a m ajor problem for largely static governm ent bureaucracies.
3. Can the governance response be innovatively accountable, and will the Internet
pioneers be accountably innovative?
Constituency Problem : governm ents derive their legitim acy from periodic general elections,
approxim ately every four years, from which they build coalitions of interests at national and
international level to regulate in the interests of the m ajority. Four years prior to W SIS, in 1999,
the global Internet population was only 100m illion, m ainly in N orth Am erica. N ow the largest
constituency is in Asia. Representation in governance is therefore an enorm ously challenging task.
On the other hand, the builders of the Internet were a group of m iddle-class white m ales from
Anglo-Saxon cultures with a clearly libertarian political ethos, and they continue to govern the
architecture of the Internet, no m atter how ethnically, sexually and age-diverse its growing
population has becom e.
4. Can the governance response be a benevolent control, and can Internet users be
responsibly free?
Bureaucracy Problem : governm ent is rules-based, and the bureaucratic m ethod is to system atize
the object to be controlled within bureaucratic definitions: what system s theory after N iklas
Luhm an describes as an autopoetic self-referential system . In brief, governm ent is a ham m er, and
to a ham m er everything else looks like a nail. For governm ent to learn how to regulate reflexively,
in the expression of pioneering regulatory theorists Braithwaite and Ayres, is a critical test of their
com petence. Internet users m eanwhile, expressing all the freedom but none of the responsibility
of this globalising network, have filled it with spam , porn, hate speech and pirated m edia. Again,
to the anonym ous Internet users, all control appears to be a ham m er blow to their freedom s,
when in fact som e responsible self-regulation is necessary in any public space. Sim ilarly m edia
and software com panies try to regulate online user behaviour with legal restrictions on intellectual
property: legal code trying to defeat or dim inish software code. The key test for users,
governm ents and corporations is to arrive at acceptable joint solutions to produce innovation and
dynam ic freedom with greater accountability and legitim acy to society. Hacking, flam ing,
spam m ing and surfing for sm ut are no longer solitary anonym ous and am oral activities, if they
ever were.
105
G lobal Public G ood Problem : The Internet as we know it has reached a crossroads. It is no longer
a U S-dom inated m edium and has reached the status of a critical infrastructure. Around the world
approxim ately a billion users now rely on the Internet for a wide range of social, political and
com m ercial transactions. But alongside this change in size and reach the Internet appears to be
increasingly vulnerable. This puts into question the highly inform al and hyper com plex
governance m odus operandithe Internet relies on.
• A m ultitude of actors, standard m aking bodies, international institutions and
com m ercial enterprises are involved in m aintaining the Internet as we know it,
but can they also reply effectively to an increasing dem and for a m ore secure and
reliable Internet?
Multistakeholder G overnance Problem
1. How can Internet governance balance the need for innovation with questions of
legitim acy and accountability?
2. Does ‘rough consensus and running code’ provide a better 21st Century policy
m odel than indirect dem ocracy via governm ents and inter-governm ental
organisations (IG Os)?
3. To what extent do national, regional or local bureaucratic structures stifle or
stim ulate Internet innovation?
4. Do civil society advocates play an effective legitim ising role?
Convergence Problem : It is recognised that regulatory (and even cultural) convergences are
occurring at differential rates between regions and sectors, such that telecom s infrastructure and
television content, to nam e two m ass m edia, are converging around global standards
asynchronously.
1. Are Internet governance structures converging between national traditions?
2. Is Internet governance converging faster than its predecessors?
3. If so, is it easier to produce cross-national solutions to governance problem s?
4. W hat kinds of “cross-fertilization” between business, ‘N etizens’ and governm ent could
be m utually beneficial and how can this be fostered?
5.6. Conclusion: Internet G overnance After Tunis
The second stage of W SIS in Tunis produced no real conclusion except agreem ent not to
challenge im m ediately the U S role in Internet governance. This result was produced from the
pre-Tunis political pressure put on the EU Presidency holder, the U nited K ingdom , by the U S
106
Secretary of State Condeleeza Rice and Secretary of Com m erce, Carlos G uiterrez, in a now-
fam ous private letter. In part they state:
The success of the Internet lies in its inherently decentralized nature, with the m ost
significant growth taking place at the outer edges of the network through innovative new
applications and services. Burdensom e, bureaucratic oversight is out of place in an
Internet structure that has worked so well for m any around the globe. W e regret the
recent positions on Internet governance(i.e., the “new cooperation m odel”) offered by
the European U nion, the Presidency of which is currently held by the U nited K ingdom ,
seem s to propose just that - a new structure of intergovernm ental control over the
Internet... The history of the Internet’s extraordinary growth and adaptation , based on
private-sector innovation and investm ent, offers com pelling argum ents against burdening
the network with a new intergovernm ental structure for oversight. It also suggests that a
new intergovernm ental structure would m ost likely becom e an obstacle to global Internet
access for all our citizens. It is in this spirit that we ask the European U nion to reconsider
its new position on Internet governance and work together with us to bring the benefits
of the Inform ation Society to all.324
This restates the June 2005 ‘Dom ain N am es: Principles of Internet G overnance’325 stated by
Michael G allagher of the N TIA:
The U nited States G overnm ent intends to preserve the security and stability of the
Internet’s Dom ain N am e and Addressing System (DN S).
S). G iven the Internet's
im portance to the world’s econom y, it is essential that the underlying DN S of the
Internet rem ain stable and secure. As such, the U nited States is com m itted to taking no
action that would have the potential to adversely im pact the effective and efficient
operation of the DN S and will therefore m aintain its historic role in authorizing changes
or m odifications to the authoritative root zone file.
G overnm ents have legitim ate interest in the m anagem ent of their country code top level
dom ains (ccTLD). The U nited States recognizes that governm ents have legitim ate
public policy and sovereignty concerns with respect to the m anagem ent of their ccTLD.
As such, the U nited States is com m itted to working with the international com m unity to
324
Rice, C. (2005) 7 November Letter to Jack Straw, at The Register:
http://www.theregister.co.uk/2005/12/02/rice_eu_letter/
The letter has been seen by many journalists who vouch for the accuracy and authenticity of this copy.
325
See http://www.ntia.doc.gov/ntiahome/domainname/USDNSprinciples_06302005.htm
107
address these concerns, bearing in m ind the fundam ental need to ensure stability and
security of the Internet’s DN S.
ICAN N is the appropriate technical m anager of the Internet DN S. The U nited States
continues to support the ongoing work of ICAN N as the technical m anager of the DN S
and related technical operations and recognizes the progress it has m ade to date. The
U nited States will continue to provide oversight so that ICAN N m aintains its focus and
m eets its core technical m ission.
Dialogue related to Internet
Internet governance should continue in relevant m ultiple fora. G iven
the breadth of topics potentially encom passed under the rubric of Internet governance
there is no one venue to appropriately address the subject in its entirety. W hile the
U nited States recognizes that the current Internet system is working, we encourage an
ongoing dialogue with all stakeholders around the world in the various fora as a way to
facilitate discussion and to advance our shared interest in the ongoing robustness and
dynam ism of the Internet. In these fora, the U nited States will continue to support
m arket-based approaches and private sector leadership in Internet developm ent broadly.
W hether that leaves any real room for serious discussion in future U N m eetings is doubtful.
Certainly the ITU response to U S intransigence on international telecom s settlem ent rates in the
1990s has been developing-country led im potent protest in the face of the hyper-power. A m ore
likely response is cyber-crim inals using poorly policed a nd under-resourced developing nations to
deliver viruses and spam at the network, as seen in N igeria, N orth K orea, Iran and other
countries in 2005.
Larry Lessig described and forewarned326 that the libertarian critics of regulation would continue
to dom inate Internet regulation in the U S, in an unholy alliance of open source
cyberlibertarians327 and red-in-tooth-and-claw corporate interests, the latter om inously
represented by the Motion Picture Association of Am erica, the single greatest lobbying power
known in W ashington D.C. European constitutional lawyers are aware of the awesom e power of
these corporate interests, even if expressed in the ‘public’ broadcasters and telecom m unications
operators who stand in the way of com petition. Lessig declares it the ‘Age of the Ostrich’, when
social libertarians will not join with supporters of governm ent provision of public goods and
exclusion of public harm s, for fear of the dead hand of bureaucracy. I have increasingly concluded
326
Lessig, L. (1999) ‘Code and Other Laws of Cyberspace’
327
For a critique of the attempts to ‘enforce’ free software, see Hoeren, Thomas (2004) The first-ever
ruling on the legal validity of the GPL - A Critique of the Case at
http://www.oii.ox.ac.uk/resources/feedback/OIIFB_GPL3_20040903.pdf
108
that it m ay hearken back to an earlier, ‘G ilded’ Age of Am erican capitalism , in which governm ent
inaction and private corporate power grew to such uncontrollable levels that the eventual crash
shook the debtor and dependent econom ies of Europe, Latin Am erica, East Asia and others328.
‘W ho are EU ?’ asked the late Susan Strange in 1998 in a landm ark article329 – how will
Europe achieve its Lisbon 2010 aim s to becom e the m ost productive and innovative Inform ation
Age econom y in the world? The success of ‘Silicon V alley Code’, of creative destruction, of
venture capital financed high technology, of individual entrepreneurship, created a cluster of
success factors including hum an capital sucked out of hierarchical institutions, such as old-style
telephone and com puter hardware com panies, universities and governm ents. The abundance of
this environm ent created thousands of ‘Microserf’ and other m illionaires, in their paper share
options. In tim e, the ‘W intelciscos’ em erged from m erger and other com petitive litigations to
dom inate parts of the personal com puter and Internet industries. They created a superhighway
for the owners of content. The abundance created by digital networks, which com bine m any of
the functions of print publishers and libraries, railways, autom obiles, telegraphs, telephones,
electric and gas power, radio and television, led to m erger m ania and a huge investm ent bubble.
This is consistent with m arket behaviour in previous boom s in the industries cited.
At a certain point in tim e – for convenience let us describe it as the high noon in 1998-2000 of
the U S Microsoft antitrust trial, though in truth it was a continuous tightening of the regulatory
noose as industries m atured, m erged and interoperated one with the other – regulation becam e
an instrum ental part of these boom ing industry sectors, as dom inance and abusive behaviour,
towards com petitors, consum ers, suppliers, investors and others, were perceived by governm ent
through the labyrinthine channels of lobbying in a com plex federal representative dem ocracy. In
consequence, this ‘Silicon V alley’ code – innovate, capitalise, exploit opportunity – was joined
with ‘Hollywood’ or ‘Madison Avenue’ Code – intellectual property and tradem ark law, antitrust
and First Am endm ent law. The boundaries are som ewhat artificial, but the lobbying
com m unities are not. Hence the Sonny Bono Copyright Term Extension Act and Digital
Millenium Copyright Acts of 1998 (the extension was not to an entire m illenium , however
com pliant Congress in what Yochai Benkler has term ed the Mickey Mouse Copyright Act, after
‘Hollywood Coders’ the Disney-ABC Corporation330). Again, the ability of incum bents to
328
‘Gilded Age’ refers generally to the enormous increase in American wealth and power created by
mass immigration, industrialisation and the transcontinental networks created in the 1880s-1920s, and
more particularly after the transfer of capital dominance from Europe to Wall Street in the 1914-18
war.
329
Strange 1998
330
See Lessig (2004)
109
m aintain power, whether it be content com panies in the Internet Age, film studios in the
television age, or m erchant banks in the railway age, should not be surprising to political
scientists, however bewildering to ostrich libertarians. The closest to any kind of serious
m oney/power discussion, a ‘Bretton W oods’ for the Inform ation Age, has been the G 8 m eeting
in Brussels in April 1995, repeated in Okinawa in July 2000, G enoa 2001 and Edinburgh in
2005.
But has the U S led? Is there political leadership or public support for an expanded role? Though
not the im m ediate project, it was these questions which repeatedly em erges in the Internet
governance debate, an ‘innovation’ in international governance of inform ation. The crisis of
legitim acy in the Internet self-governance com m unity is striking, but the U nited N ations
interm ediary, K ifi Annan’s e-Envoy Markus K um m er, a Swiss diplom at, has an im possible job
when m eeting V int Cerf, of ISOC. W ere these two heads of state or of international
organisations, the protocol can be im agined. The Internet ‘protocol’ of their m eeting was a
sim ple ‘How do you do’. It dem onstrates both the inform ality and the lack of form al hierarchy of
the current m erit-adhoc m ethods of regulating. Much will be lost as well as gained with the
increasing form alism which appears to be em erging. W itness ICAN N elections – Esther Dyson
m ade clear in a London panel on Internet governance 25th Septem ber 2004 her discom fort in the
role and wish to tell the ‘real’ ICAN N story, though others dism issed ICAN N ’s ‘neo-liberal
cesspool’ as ‘Californication and original sin’ given ICAN N incorporation as a private com pany
in the U S W est Coast. The question posed is:
W ill Am erica lead in the Internet Age, or will it allow private corporations to set agendas to
which legislators respond purely on the captured business case it presents for their clients?
If Disney, Cisco, Motorola, IBM, G oldm an Sachs and others exert international leadership as
proxies for their governm ent, that m ay be a return to the collapsed international system under the
farcical League of N ations in the 1920s (public choice case studies par excellence)331.
Meanwhile the real action in global Internet governance continues to take place in standard-
setting bodies. It is to these that we now turn: first content in the fastest growing industry sectors
(com puter gam es and m obile content); then technical standard-setting.
331
As much was hinted by Kenn Cukier in discussion on an outstanding panel at the www.tprc.org
2004 conference on 5 October, with bleak references to the crisis of United Nations agencies, of
developing countries’ policies, and of Internet self-regulatory legitimacy.
110
Chapter 6: Co-
Co-Regulation of Internet Content
“In broadcasting, it is possible to im pose additional rules beyond the straightforward

crim inal law because broadcasters require regulators’ perm ission to operate... For som e
form s of new m edia distribution, sim ilar pegs m ight be found - for instance for m obile
content. But for pure, internet-delivered content it is difficult to see how any m eaningful
licensing controls could be im posed and hence how any sanction could be enforced.
These problem s arise even if the regulatory instrum ent of choice is a co-regulatory
schem e in which industry operates against a long-stop of possible enforcem ent action by
the regulator.”
David Currie, Chair, OfCom (2005) 332
6.1. W hat is co
co--regulation?
Co-regulation expresses a form of regulation which is neither state com m and-and-control
regulation in its bureaucratic central or IRA (Independent Regulatory Agency) specialized
functions333, but is also not ‘pure’ self-regulation as observed in industry-led standard setting. The
state, and stakeholder groups including consum ers, are stated to explicitly form part of the
institutional setting for regulation. Co-regulation constitutes m ultiple stakeholders, and this
inclusiveness results in greater legitim acy claim s. However, direct governm ent involvem ent
including sanctioning powers m ay result in the gains of reflexive regulation – speed of response,
dynam ism , international cooperation between ISPs and others – being lost. It is clearly a finely
balanced concept, a m iddle way between state regulation and ‘pure’ industry self-regulation.
Responsive regulation reflects a m ore com plex dynam ic interaction of state and m arket, a break
with m ore stable previous arrangem ents. Teubner states: “European conceptions of law as
“m oving away from the idea of direct societal guidance through a politically instrum entalised law
332 Lord Currie, Speech 21 September 2005, available at: http://www.ofcom.org.uk/media/speeches/2005/09/liverpool_conf

333 Baldwin, R. et al (1998) Socio-Legal Reader on Regulation, at 3 explain that “At its simplest, regulation refers to the promulgation of an
authoritative set of rules, accompanied by some mechanism, typically a public agency, for monitoring and promoting compliance with these rules.”
334 See a summary of de Cockborne’s Montreux speech in Adam Watson Brown (1999), Industry Consortia and the Changing Roles of Standards
Bodies and Regulators, 35 Inst. Prospective Tech. Stud., June 1999, available at http://www.jrc.es/pages/f-report.en.html
335 P.P Craig at 197 in Baldwin, R. and McCrudden, C. (1987) Regulation and Public Law.
111
… Instead, reflexive law tends to rely on procedural norm s that regulate processes, organisation,
and the distribution of rights and com petencies”337. This applies to other globalising phenom ena
than digital TV and the Internet, for instance financial and environm ental law338, where negative
externalities are highlighted for public concern. In advertising co-regulation, protection of m inors
and consideration of broadcast regulation’s extension to new m edia including the Internet and
3G /U MTS m obile phones, co-regulation is a vitally im portant concept to define, refine and
exam ine against a rigorous m ethodological tem plate.
The term ‘co-regulation’ encom passes a range of different regulatory phenom ena, which have in
com m on the fact that the regulatory regim e is m ade up of a com plex interaction of general
legislation and a self-regulatory body. The varying interests of actors result in different incentives
to cooperate or attem pt unilateral actions at the various points of the value chain. W ithout
regulation responsive to both the single European m arket and the need for constitutional
protection of freedom of expression and protection of m inors at national levels, co- and self-
regulatory m easures cannot be sufficiently responsive to econom ic and cultural environm ents to
be self-sustaining.
6.1.1. W hat can be learnt from existing studies in analysing co

co--regulation?
There have been m any studies of self- and co-regulation in the m edia sector since Boddewyn’s
pioneering 1988 study of advertising341, notably those of Oxford U niversity342, of PCMLP faculty
and associates independently343 and with collaborators344, while shorter country- or sector-specific
contributions have been published in the past five years in IRIS346.
337 G. Teubner, '

The Transformation of Law in the Welfare State'in G. Teubner (ed.), Dilemmas of Law in the Welfare State (Berlin: W. de Gruyter,
1986), at 8.
338 See for instance Gaines, Sanford E. and Cliona Kimber (2001) Redirecting Self-Regulation Env. Law 13(157)
339 See Whitehead, Phillip (1997) Draft Report on the Commission Green Paper on The Protection of Minors and Human Dignity in Audiovisual
and Information Services (COM[96]0483 - C4-0621/96) PE 221.804 of 24 April 1997.
340 A communitaire legal justification for national application of EC competition law under the Treaty of Rome is provided in Temple-Lang, J. (1998) The
Duty of National Authorities under Community Constitutional Law 23 European Law Review 109 at 119
341 Boddewyn J. J. (1988) Advertising Self-regulation and Outside Participation. See also Bollinger, Lee C. (1976) “Freedom of the Press and Public Access:
Toward a Theory of Partial Regulation”, 75 Michigan Law Review 1
342
PCMLP (2004) documents from European Commission study at www.selfregulation.info
343
Price, M., Verhulst, S. (2004) Self-Regulation and the Internet; Larouche, Pierre (, 2001),
‘Communications convergence and public service broadcasting’, at
http://infolab.kub.nl/uvtweb/bin.php3?id=00011353&mime=application/pdf&file=/tilec/publicatio
ns/larouche2.pdf
344 See also Humphreys, Peter J. (1996) Mass media and media policy in Western Europe; Harcourt, A. (2004) European Institutions and Regulation of the
Media Industry.
345 Marsden, C. (1999) Pluralism In The Multi-Channel Market: Suggestions For Regulatory Scrutiny, Council of Europe Human Rights Commission, Mass
Media Directorate, MM-S-PL [99] 12 Def 2.
346 See special issues of IRIS in 2002-3, notably IRIS Special (2003) Co-Regulation of the Media in Europe, Strasbourg, Council of Europe at
http://www.obs.coe.int/oea_publ/iris_special/2003.html.en Schneider, Andrea “Child Protection on German Television – The Voluntary Television Review
Body (FSF)”, IRIS 1995-3:7/13.
112
Schulz and Held have investigated co-regulation in the G erm an context, specifically in the case of
protection of m inors347. In their view, self-regulation in Anglo-Am erican debate is concerned with
“reconciliation of private interests” whereas their form ulation – regulated self-regulation348 – is
indirect state regulation based on constitutional principles. It is the com bination of “intentional
self-regulation” – the actions of m arket actors, whether in social or econom ic settings – with the
state sanction in reserve which results in self-regulation which is ‘regulated’ by the possibility of
state intervention. At the Birm ingham ‘Assizes Audiovisuels’ in 1998, the form ulation used was:
“Self-regulation that fits in with a legal fram ework or has a basis laid down in law”.
The term ‘co-regulation’ also gives a sense of the joint responsibilities of m arket actors
and state, in the activity under investigation. It has been used by the U K ’s telecom regulator to
suggest a state role in setting objectives which m arket actors m ust then organize to achieve – with
the threat of statutory powers invoked in the absence of m arket self-regulation349. However, co-
regulation is used in such a wide variety of circum stances that its specific m eaning m ust be seen
in the national, sectoral and tem poral context in which it is used.
Schulz and Held suggest that ‘regulated self-regulation’ can be any of these categories:
co-regulation, intentional self-regulation, or a third category - ‘audited self-regulation’.
Independent audit of self-regulation is a U .S. concept of using an independent standard or
professional body to audit a self-regulatory organization or individual com pany according to pre-
set standards. In the case of ISPs, audited self-regulation m ight involve at least a standard being
set that an audit firm could certify organizations against (or at least that organizations could self-
certify with reporting requirem ents), but could involve the setting of an international standard, as
increasingly occurs in accountancy, for instance. At a m inim um , dedicated budgetary and
personnel resources, with activity reports, would be required to dem onstrate regulatory
com m itm ent. The G erm an concept of regulated self-regulation gives the state a role when basic
constitutional rights need to be upheld: “The extent of possible delegation [to self-regulation]
depends … on the relevance … in term s of basic rights”.
Co-regulation in the European context m ust also be proportional to the aim s of the legal
instrum ent, as well as conform ing to the com petition law of the European U nion. Enforcem ent is
the ultim ate responsibility (‘the safety net’) of the state. In Schulz and Held’s case study,
347 Schulz and Held (2001) Regulated Self-Regulation as a Form of Modern Government.
348 See Hoffman Reim, Wolfgang (1996) Regulating Media.
349 See Thomas, Richard
(2000) "Better business practice: how to make selfregulation work for consumers and business" National Consumer Council
(UK) at http://www.ncc.org.uk/pubs/pdf/self-regulation_gpg.pdf and OFCOM (2004) Consultation Document ‘Criteria for Transferring Functions to co-
Regulatory Bodies’ http://www.ofcom.org.uk/consultations/past/co-reg/?a=87101
113
Australia, practical self-regulation is illustrated in the application of the 1997 Telecom s Act and
1992 Broadcasting Services Act, where four types of regulatory schem e can be identified:
Regulatory Type State Role
1. Intentional or ‘Pure’ Self-regulation N o state IRA involvem ent
2. Industry Codes Registered with the state IRA
3. Industry standards Mandatory codes set in the absence of pan-

industry code agreem ent
4. Com m and-and-control Set by state IRA pre-em pting attem pts at self-
regulatory action
However, there are clearly nuanced approaches that industry can take, in choosing the m enu of
‘regulated self-regulation’ to adopt. Multinational actors can choose to engage in gam e-playing
with the regulator and other actors in order to secure their preferred environm ent for regulation,
and it is here that national ‘regulated self-regulation’ according to national constitutional
standards are arbitraged by m ultinational actors seeking a m ore co-ordinated international
regulatory fram ework. That is not to suggest the extrem e and celebrated case in which Yahoo!
attem pted to substitute U .S. First Am endm ent speech standards over French governm ent
controls on illegal N azi m em orabilia auctions (clearly a Type 4 regulatory discussion), but to
suggest that Types 1-3 above afford substantial latitude for rational m ultinational actors to seek
to arbitrage regulators in favour of their ‘hom e state’ standards for speech and com m ercial
freedom (which in the U .S. coalesce m ore frequently than in Europe, which does not protect
com m ercial speech to the sam e extent).
However, there are clearly nuanced approaches that industry can take, in choosing the m enu of
‘regulated self-regulation’ to adopt.350 The Oxford PCMLP project researched in the fifteen pre-
2004 Mem ber States in the areas of: broadcast co- and self-regulation; m obile telephony and
child protection; Internet self-regulation; com puter gam es and video cassette ratings schem a;
print news m edia self-regulation. Based on the www.selfregulation.info report and other prior
work, they offer som e tentative initial hypotheses351. Three options suggest them selves:
350 See Reidenberg, J. (2004) States and Law Enforcement, 1 Uni.Ottawa L.& Tech.J. summarising Reidenberg, J. (2002) Yahoo and democracy on
the Internet, 42 Jurimetrics 261.
351 PCMLP refers to its conclusions, final chapter and executive summary for the IAPCODE project, which contain detailed options for co-
regulation in the media co-regulatory area, including specifications for regulatory audit.
114
1. Adopting best practice in self-regulatory approaches taken from the U S and potentially
U K m odels;
2. Developing and extending a sophisticated version of co-regulation such as that found in
Australia or G erm any, with a pan-sectoral focus.
3. Extending either practice to a pan-European role, as in the Internet sector, where
hotlines (IN HOPE), ISP Codes of Conduct (EuroISPA) or video gam es ratings (PEG I) have
adopted a successful m odel. However, the role of free speech, cultural diversity and the
enforceability of such regim es rem ain problem atic.
These three options are in addition to nation-specific and sector-specific status quo options,
which one m ight term Option 0.
In considering the range of self-regulatory solutions across Europe, it is necessary to reflect on
exactly why there is a range of responses, and whether it is possible to conceive of a European
m odel of m edia co-regulation:
• W hat is the m ost im portant national factor with regard to co-regulation, and what are
the barriers to international cooperation?
• Is it legal and constitutional and the im plications for co-regulation or rather the
differences in cultural content standards?
• Is it rather a m ore com plex set of factors relating to institutional political econom y?
To place this survey in the context of country-level differences and EU -wide changes that im pact
on m em ber states in contrasting ways, the level of analysis m ust be useful for:
• understanding co-regulation on the national level,
• for policym aking that is concerned with coordinating national m edia approaches across
sectors, and
• for evaluating prospects for convergence in practices on the EU level.
There are difficulties in assessing changing political cultures. Cultural as well as econom ically
rational m otivations differentiate state and m arket actors. Pan-European options present further
com plexity: m ultilateral solutions m ay therefore be theoretical solutions to intractable real-world
problem s. Yet, when co-regulation is put into practice this is often first done on the national
level, and here attention to econom ic governance, political culture, civil society and institutions
in general m ay m ake a crucial distinction in assessing which co-regulatory schem es succeed and
which fail.
115
6.2. Self
Self--regulation of the electronic gam e industry
The electronic gam es industry is not new – having existed offline for over thirty years – but is
becom ing a m uch m ore significant part of the m edia. Form erly a gam es console based discrete
activity, it is rapidly becom ing a broadband-enabled online activity, with m assive m ultiplayer
gam ing having originated in broadband pioneer South K orea. According to PriceW aterhouse
Coopers352, the global entertainm ent and m edia industries are expected to see accelerated growth
of 7.3 percent and will reach $1.8 trillion in annual sales by 2009. The global video gam e m arket
is expected to be the fastest-growing com ponent and will increase from $25.4 billion in revenue
in 2004 to alm ost $55 billion in 2009, growing at a 16.5 percent com pound annual rate. The
introduction of the next-generation broadband-enabled consoles, as well as the continued growth
of handheld and m obile gam ing, will be key drivers of growth. The PC gam e m arket will
continue to shrink with gam e sales projected to decline from $771 m illion in 2004 to $655
m illion in 2009.
Figure: OECD (2004) W orld Com puter G am es Market to 2008353
Europe m ay be only the third largest gam es m arket (after East Asia and N orth Am erica) but it is
the largest publisher of gam es, with about 40% of gam es authored in the European U nion.
Additionally, Europe and East Asia have m uch greater adoption of m obile gam es than N orth
Am erica. A corpus of legal scholarship concentrates on the virtual world’s relationship with the
352
Price Waterhouse Coopers (2005) Global Entertainment and Media Outlook 2005-9 from
http://www.pwcglobal.com/extweb/industry.nsf/docid/8CF0A9E084894A5A85256CE8006E19ED?ope
ndocument&vendor=none#IAAS
353
See OECD (2005) Digital Broadband Content: The Online Computer And Video Game Industry
DSTI/ICCP/IE(2004)13/FINAL at pp53-54, at http://www.oecd.org/dataoecd/19/5/34884414.pdf
116
law, with the ‘State of Play’ conference an annual event in N ew York354. Much of Lessig’s
discussion in the sem inal ‘Code and Other Laws of Cyberspace’ considers the etiquette of virtual
environm ents. The effect of violent gam ing on youth has been recently m uch debated in the
m edia, and has led the Am erican Psychological Association to pass a 2005 resolution which states
in part:
“BE IT FU RTHER RESOLV ED that APA engage those responsible for developing violent video
gam es and interactive m edia in addressing the issue that playing violent video gam es m ay increase
aggressive thoughts and aggressive behaviors in children, youth, and young adults and that these
effects m ay be greater than the well docum ented effects of exposure to violent television and
m ovies.
BE IT FU RTHER RESOLV ED that APA recom m end to the entertainm ent industry that the
depiction of the consequences of violent behavior be associated with negative social consequences.
BE IT FU RTHER RESOLV ED that APA (a) advocate for the developm ent and dissem ination
of a content based rating system that accurately reflects the content of video gam es and interactive
m edia, and (b) encourage the distribution and use of the rating system by the industry, the
public, parents, caregivers and educational organizations.”355
There is a great deal of pressure on the industry, whether or not one accepts the APA’s finding
that video gam es can be m ore harm ful to young people than television. Content rating is
instituted in Australia under the co-regulatory schem es established by the N ational Classification
Code of 1995, under the powers in the Broadcasting Services Act 1992. In 2000, Section 5 of the
Act becam e operational, extending the co-regulatory system to the Internet356. Further co-
regulatory m easures have recently been introduced in China and South K orea (to prevent gam es
players’ obsessive playing resulting in harm to self and others offline).
European concerns with com puter gam es have reached the stage of pan-European legislation to
regulate the content of gam es, in the new proposed Recital 12 of the new Recom m endation on the
Protection ofM inors and H um an D ignity as am ended:
This Recom m endation covers new technological developm ents and com plem ents
Recom m endation 98/560/EC. Its scope, on account of technological advances, includes
audiovisual and on-line inform ation services, such as newspapers, m agazines and,
354
See http://www.nyls.edu/pages/2713.asp
355
American Psychological Association (2005) Resolution on Violence in Video Games and Interactive
Media at http://www.apa.org/releases/resolutiononvideoviolence.pdf released August 17 at
http://www.apa.org/releases/videoviolence05.html
356
See Wright, A. (2005) Coregulation of Fixed and Mobile Internet Content, paper at ‘Safety and
Security in a Networked World: Balancing Cyber-Rights and Responsibilities’ conference, Oxford,
September, at http://www.oii.ox.ac.uk/research/cybersafety/?view=papers
117
particularly, video gam es, m ade available to the public via fixed or m obile electronic
networks357.
The dom inant m odel for self-regulation of the video gam e industry on the pan-European level is
the PEG I rating system (the Pan European G am es Inform ation), im plem ented in the spring of
2003 by the Interactive Software Federation of Europe (ISFE)358. The rating system is a result of
a period of collaboration and negotiation between stakeholders from national self-regulatory
organizations and the industry, and the project also received either advice or support from m ajor
video console m anufacturers, experts in the field, and relevant stakeholders within the European
Com m ission. As of Septem ber 2005, 16 European states were participating in the schem e and the
m ost notable country not included was G erm any (due to its new legislation on protection of
m inors that includes video gam es and the Internet within the larger national self-regulatory
fram ework).359
PEG I was updated in August 2005 to take account of m obile and online gam ing, extending its
rem it360. It has been referred to as a m odel exam ple of pan-European regulation by Com m issioner
Reding in speeches in 2003 and 2005361. In its first two years of operation, PEG I statistics show
the following im plem entation.
357
European Parliament (2005, 7 September) legislative resolution on the proposal for a
recommendation of the European Parliament and of the Council on the protection of minors and human
dignity and the right of reply in relation to the competitiveness of the European audiovisual and
information services industry (COM(2004)0341 – C6-0029/2004 – 2004/0117(COD)) at
http://www.europarl.eu.int/omk/sipade3?PUBREF=-//EP//TEXT+TA+P6-TA-2005-
0330+0+DOC+XML+V0//EN&L=EN&LEVEL=0&NAV=S&LSTDOC=Y&LSTDOC=N
358
See Calvert, Justin (2002) New age rating system for games in Europe, ‘Gamespot’ 25 October at
http://news.zdnet.co.uk/hardware/emergingtech/0,39020357,2124493,00.htm
359
The participating countries included Austria, Belgium, Denmark, Finland, France, Greece, Ireland,
Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United
Kingdom.
360
See PEGI Info newsletter No.7 at http://www.pegi.info/pegi/download.do?id=12
361
See presentation by Secretary General Patrice Chazerand Liverpool 21 September 2005 at
http://www.fco.gov.uk/Files/kfile/AudiovisualConference_Draft_Programme.pdf
118
Figures: G am es Rated in Years April 2003-March 2005
The PEG I age categories are 3, 7, 12, 16 and 18, and all the gam es also receive up to six content
descriptors, warning that gam e’s content includes discrim ination, drugs, fear, bad language, sex,
or violence..
Figure: PEG I content descriptors
violence bad language fear sex drugs discrim ination
The three m ost prom inent self-regulating bodies of the electronic gam e industry by 2003
included the Entertainm ent Software Rating Board (ESRB, a program of the Interactive Digital
Software Association in the U nited States), the V ideo Standards Council (V SC, a program m e of
the European Leisure Software Publishers Association in the U nited K ingdom ), and the
K ijkwijzer (a program m e of the N etherlands Institute for the Classification of Audio-visual
Media in the N etherlands). From analysis of the three system s, it becom es clear that the PEG I
system resulted from the initiative of the U K ELSPA, and was im portantly influenced by the
119
British self-regulatory experience. However, the addition of content descriptors and m ore reliance
on the m anufacturers are only two of several im portant influences from the Dutch N ICAM
m odel. In that sense, the ISFE’s PEG I rating system is a self-regulatory m echanism that
incorporated the institutional and historical lessons of British self-regulation in the areas of film s
released on video, and later video gam es, with new and original thinking behind the Dutch
experim ent of rating m edia content regardless of the m edium in which it is found.
Because PEG I is the only pan-European system , it has a potential of serving as a best-practice
m odel for other industries in search of a coordinated schem e of self-regulation across EU m em ber
state boundaries. The experience in this industry sector indicates that national self-regulatory
agencies can serve as im portant advocates of im plem enting a pan-European system , but their
initiative has to be accom panied by a consensus that there is indeed a need to im plem ent a system
that transcends national boundaries. In the case of the video gam es, the pan-European system was
also m ade possible by the fact that the U K ELSPA system was well developed and that the vast
m ajority of the other m em ber states either lacked any system of video gam es ratings or had only
started to think about adopting their other m edia rating m echanism s to deal with this relatively
new and expanding m edia industry.
However, the exem ption of one of the world’s largest national m arkets for video gam es,
G erm any, from the PEG I system serves as an im portant rem inder that even the m ost successful
‘pan-European’ self-regulatory m odels m ay not live up to all the Mem ber States’ concerns or
national legal requirem ents. G erm any’s newly-im plem ented laws on the protection of m inors
placed video and com puter gam es within the overall m edia ‘co-regulatory’ fram ework, with a
shared responsibility between a self-regulatory body on the federal level and länder governm ents
on the local level. Such a national schem e, a priori, one can argue, precludes G erm any from
relying on an entirely voluntary pan-European system of video ratings.362
V ideo and com puter gam es were am ong the m ost recent m edia to be subjected to public and
governm ent scrutiny over suitability of their content for children. Drawing on experiences of the
video and film industries, the response of several electronic gam e industries in both Europe and
N orth Am erica has been to introduce self-regulatory m echanism s of content rating according to
age suitability. This is not to forget that som e m em ber states, such as G erm any and until recently
France, relied m ore heavily on the governm ent to participate in determ ining what video gam e
content should be available to the m inors.
362
However, one could envisage a system where local governments could delegate the rating
responsibility to a pan-European self-regulatory agency.
120
The three self-regulatory m echanism s that predated the PEG I ratings showed considerable
diversity of approaches to constitution, content, coverage, com m unication and com pliance
elem ents in their codes of conduct and practices. The U nited K ingdom was the m ost difficult
case as the V ideo Recordings Act only has lim ited (but still significant) application to the
electronic gam e industry. One way out of these com plexities would have been to negotiate safe
harbour provisions (as was the case with the ESRB and the Federal Trade Com m ission in the
U .S.) or to am end legislation to introduce a governm ent-endorsed self-regulatory body (as was
the case with the Dutch Media Act and N ICAM). The solution, however, was found neither in
following a strictly Am erican or a strictly Dutch approach, but instead adopting the ideas of the
British-dom inated ELSPA to a new pan-European system .
Another im portant decision that was to be m ade before PEG I was to com e in force was whether
to pursue video self-regulation within a national m edia self-regulatory schem e (i.e. G erm any) or
to separate the video gam es industry and address it through its own rating system (eventually the
PEG I system ). The experience of the V ideo Standards Council showed the difficulties of applying
the sam e rating m echanism to videos and gam es without applying it to film s and m usic; with
developm ent and convergence of technology, the preferred approach is either the universal or the
highly specialized. This problem is com pounded by the challenge of interactive gam ing, with
developm ent of consoles that allow playing against (or with) anonym ous players, who can
com m unicate during the gam e.
PEG I itself claim s to be m ore co-regulatory and self-regulatory:
One key feature of the PEG I system - probably the m ain driver of its success - is its
unique com bination of business and governm ent input (on this ground, som e would call
it co-regulation rather than self-regulation).363
Its com position, with an advisory board m ade up alm ost entirely of governm ent representatives,
14 out of 15 m em bers, certainly supports that interpretation364. The ratings system has provided
a m odel for the U K Independent Mobile Classification Body (IMCB)365 – see Chapter 10.
Chazerand m akes the point that the system works for three specific reasons:
363
Chazerand (2005) Liverpool speech supra at http://www.europa.eu.int/comm/avpolicy/revision-
tvwf2005/docs/ip6-isfe.pdf. See also ISFE (2005) Position paper on the Communication from the
Commission “Challenges for the European Information Society beyond 2005” at http://www.isfe-
eu.org/index.php?PHPSESSID=804e6e3cfdc7506303904aa81afe850e&template[0]=matrice.html&tem
plate[1]=rubrique.html&oidit=T001:dfb8bc4a6dcf3052747cea0839f32f75
364
See IFSE (2005) Appendix to ISFE’s comments on Issue 6: Protection of Minors and Human
Dignity, September at http://www.europa.eu.int/comm/avpolicy/revision-tvwf2005/docs/ip6-isfe-
encl.pdf
365
See UK Independent Mobile Classification Body (2005) IMCB Guide and Classification
Framework for UK Mobile Operator Commercial Content Services at p6 at http://www.imcb.org.uk/
121
All the m ajor industry players joined the system at its political inauguration in 2001-2, urged by
the European Com m ission and Mem ber States;
It has been able to take advantage of the classification system expertise of N ICAM and the
appeals procedure expertise of V SC, in a telling exam ple of pan-European system s using national
expertise for com parative advantage;
The continued success of the system has produced a virtuous circle of im proved procedures (for
instance appeal) and pan-sectoral im pacts (for instance, m obile and online m arkets).
Research com m issioned by PEG I in 2004, fifteen m onths after launch, suggested that the pan-
European system had resulted in greater player awareness of ratings than in Australia’s system
which had been developed in 2000:
Com pared with num bers taken in Australia, a territory with a sm aller population and a com m on
language, 3 years after the (Australian) OFLC launched its own age rating system , those com ing
out of the N ielsen survey look pretty encouraging: 59% of respondents said that they were aware
of a gam e rating system in Europe against 42% in Australia. Am ong the players’ com m unity, the
results are pretty close: 69% for PEG I vs 74% in Australia366.
This highlights the com m unications and m edia literacy elem ent of any Code of Conduct or
ratings system . W ithout consum er awareness, and retailer education to actually enforce the
ratings, the system is an em pty vessel. IFSE is em barking on a m edia literacy course in 2005-6 to
further publicise the ratings system . In the U K , for instance, a special ‘gam es sum m it’ was held
on 5 Decem ber 2004, calling on retailers and parents to ensure that they did not allow children
to play 18 rated gam es367.
It appears that PEG I has proved the success of a co-regulatory system in a highly dynam ic and
rapidly converging industry. U sing the exam ples taken from film classification and video
classification, it has form ed a pan-European system (absent G erm any) that has itself been
acclaim ed by the European Com m ission and adopted in part by the m obiles industry. Its effects
are extending to online and m obile m arkets through the ratings system itself. Its success m ay be
m ore an exception than the rule, however, as the lack of institutional inheritance and prior art in
regulation in this field perm itted com puter gam ing to take a m ore self-regulatory approach than
m ore established m edia. The sim ilarities with m obile content self-regulation are striking, but the
366
See IFSE Press Release (2004) 14 December: PEGI Celebrates Its First Two Years Of Operations at
http://www.isfe-
eu.org/index.php?PHPSESSID=ea3fccbb3951ccab88f9ca9a568e1772&template[0]=matrice.html&tem
plate[1]=rubrique.html&oidit=T001:96dec7f314175b346499b34f5ad64fda
367
See Video games warning (2004) http://www.politics.co.uk/domestic-policy/video-games-warning-
$7240904.htm
122
differences com pared to m any m ore established industries m eans that the view of PEG I as a
blueprint for all content regulation online is sim plistic.
6.3. Mobile Internet Services and Protecti
Protection of Minors from Adult
Adult Content
Mobile services have been used to serve web pages to European users since approxim ately 2000.
The first generation of m obile Internet devices used W ireless Application Protocol (W AP) to
deliver specially program m ed, norm ally sim plistic and graphic-poor pages over narrow-band
G lobal System for Mobile (G SM) networks. These 2G networks delivered data at about 65% of
the speed of the m odem s used for fixed line com puters circa. 1994/5. The screen for G SM-only
devices is typically very sm all and m onochrom e, and the pixilation (granularity) of the screen
m eans that photographic im ages are cartoon-like. Text services (Short Messaging Services or
SMS) on G SM devices have developed as 160-character text m essages, rather than W AP-enabled
chat or listserve.
Figure: From G SM Digital Mobile Handsets to Advanced Internet Access Devices368
368 Source Croxford and Marsden (May 2001)

I Want My WiFi White Paper at
http://www.croxford.org/ivan/rethink/index.html, typo in original.
123
Table: Explaining the Evolving G enerations of Standards and Handsets for Mobile Internet
2 G SM 2.5 G PRS 3G W iFi & Sm art Phones

Bluetooth
Downloa 9 26 64-384 c.5000 (W iFi); Dependent on
d speed 50,000 network
kb/s (W iFi5)369
Typical
phone
Content SMS, W AP G raphics, V ideo All – including Public Internet

types surfing and photos, gam es stream ing and DV D quality
ringtones downloads and capture, video
MP3 broadband download
applications
The second G eneration of handsets, for G PRS services, offers 64k colour screens, access at up to
27kb/s to 2.5G networks, and larger screen size. Text is enhanced by im ages. The 3rd G eneration
– so-called Sm artPhones – and the data-card connected Personal Digital Assistants and laptop
com puters – are all enabled to receive web pages without re-coding for W AP. These are therefore
the first portable Internet devices. Accessing the W W W at G PRS speeds, increasing to 3G
(perhaps 384K b/s) and W iFi (up to several m egabits), they can approxim ate the wired Internet
use experience. W ith larger full-colour screens, they are fully specified Internet devices for im age,
sound and video.
From the consum er's point of view, the m ain differences between old and new generation
m obiles are characterized by the different applications they facilitate, which can be sum m ed up as
follows:
• 2G allows short m essaging services (SMS), W ireless Application Protocol (W AP) and
SMS applications;
• 2.5G allows Multim edia Messaging Service (MMS) including low-resulution video, Java
and BREW gam es;
• 3G allows rich m edia, stream ing, full-m otion high-resolution video.
369 Depends on ‘backhaul’ DSL speed – see generally Croxford and Marsden (2001).
124
This m eans that the concerns raised by the need to protect children from harm ful m aterial
accessed via PCs m ight also be raised by m obile access to the Internet.
Such technological advances have however also led to the developm ent of new business m odels
for network operators, which focus largely on collecting revenue from online content. The
m odels range from :
• V ertical Integration: network operators offering their own content (e.g. V odafone Live!)
to
• Interm ediation: network operators allowing third parties to provide content (paym ent
authorized by the network operator) to
• Transit only: providing open access to the Internet (with paym ent, if any authorized by
third parties such as Bango.net).
In the U K , m obile operator 3 is currently the only operator offering open access to the Internet
for adult subscribers, whereas the other operators em ploy walled gardens. It is im portant to
recognize the existence of these different m odels, as the possibilities for co- and self-regulation
associated with the content delivery m odel will clearly vary according to the sources of content.
6.3.1.Legal and Regulatory Fram ework: Mobile Service
Mobile networks in Europe have very lim ited com petition, with between 3 and 6 networks in
m ajor m arkets. The costs of term inating calls on m obile networks, previously unregulated, have
recently been exam ined and regulated in the U K 370. The first national com m ercial broadband
network for m obile is that of Hutchison W ham poa’s 3 service in the U K and Italy (as at
01/03/04). By end-2004, m ost EU m em ber territories have m etropolitan broadband wireless
services – by 3G and W iFi ‘hotspots’ – which m eans that custom ers with handsets have a real
Internet experience.
The European Com m ission has reported on Mem ber States’ regulation of m obile content:
G erm any, France and Finland indicated that transm ission via m obiles, in particular
through U MTS [3G ], is covered by regulation. Sweden considers that its legislation on
illegal content is in principle applicable to m obile phone transm issions, but m entioned
that this had not been tested in the courts. The N etherlands argued that the self-
regulatory provisions had been drafted in a technologically neutral way, but were lim ited
to “hosted inform ation”.371
370 See http://www.ofcom.org.uk/static/archive/oftel/publications/mobile/ctm_2002/docs_index.htm for a full list of documents submitted by Oftel, or the full
Competition Commission report at http://www.ofcom.org.uk/static/archive/oftel/publications/mobile/ctm_2003/ctm2.pdf
The Report is 15 chapters with 9 appendices and took a year to research and publish.
371 COM (2003) 776 Final at 12: http://europa.eu.int/comm/avpolicy/legis/reports/com2003_776final_en.pdf
125
In addition to these ‘blanket’ regulations of fixed content extended to m obile, the U K and
N orway responded by pointing to codes of conduct being developed to learn from fixed ISP self-
regulation. The U K code is the central case study in this chapter.
6.3.2. Self-
Self -Regulation in Practice in Mobile Content
Mobile networks already have two exam ples of regulated self-regulation in place before Codes of
Conduct are considered. These are:
• an om budsm an service for custom er com plaints over pricing and service, and
• a prem ium rate regulator (for instance RegTel in Ireland or ICSTIS in the U K ).
Both are shared with fixed line telephony, and are m andated under European law.
Om budsm an Schem e for Consum er Disputes:
In the U K , V odafone was instrum ental in establishing OTelO, and T-Mobile and Orange in
establishing the Com m unications and Internet Services Adjudication Schem e (CISAS), an
om budsm an for ISPs and phone com panies. OTelO charges all m em bers a fee, CISAS is free –
an interesting exam ple of regulatory com petition372.
Prem ium Service Self-Regulator:
ICSTIS is a m em ber of the International Audiotex Regulators N etwork (IARN ), the European
(and Australian) self-regulatory network for prem ium services. Operating since 1995, the 15th
m eeting in N ovem ber 2002 has been followed by a long period of inactivity373. Mem ber self-
regulators have indicated that continuing support for IARN depends on resum ed activity. If
IARN were to fail, that would signal decreased pan-European coordination in prem ium services,
at a critical tim e when the m obile industry needs such a co-ordinatory network.
European broadcasters, fixed and m obile operators increasingly use prem ium services to fund
interactive television ‘reality’ and quiz program m es, such as G ross Bruder/Big Brother/G ran
Herm ano. In the U K , m obile prem ium texting is doubling in value each year, and was worth
£200m illion/€300m illion annually in 2003374. Total prem ium voice calls were worth €1billion in
2003, with m obile about 10% of that total. Directory enquiry calls were worth €400m illion. The
entire U K prem ium rate industry – the largest in Europe – was worth €1.7billion.
372 See http://www.arbitrators.org/cisas/ for ISP and mobile ombudsman; http://www.out-

law.com/php/page.php?page_id=ispsandtelcosmust1069270107&area=news for news of CISAS’ establishment and parentage, and
http://www.otelo.org.uk/content.php?menuID=2&pageID=23 for details of the OtelO membership board. For legal measures, see Article 34 of the Universal
Service Directive: 2002/22/EC of 7 March 2002:
http://europa.eu.int/information_society/topics/telecoms/regulatory/new_rf/documents/l_10820020424en00510077.pdf
373 See http://www.iarn.org/ last visited 3 March 2004, and interviews with national self-regulators February-March 2004.
374 Source: IAPCODE estimate based on ICSTIS 2003 figures.
126
Tables: ICSTIS 2002 Annual Report Breakdown of Com plaint Types and Services375
The biggest increase in com plaints were for adult, text m essage and dating services. Interviewees
have explained that m any com plaints and especially telephone enquiries relate to adult services,
which are in fact spurious based on fam ily m em bers discovering breaches of callers’ anonym ity.
Other Regulatory Requirem ents on Telephony Content
• Intercept, Integrity and Surveillance: N etworks m ust com ply with network integrity and
security m easures to ensure surveillance is possible, that the European em ergency
num ber 112 is accessible.
• Mobile Handset Theft: Mobile networks also have system s to deactivate the SIM card of
phones reported as stolen.
• N um ber Portability: Further m easures to m onitor phone use include a Hom e Location
Register (HLR) in each Mem ber State, to perm it m obile num bers to be ported by
subscribers from one network to a new subscription on a different network.
• Spam Blocking: In several Mem ber States, unsolicited com m ercial m essages (spam ) are
regulated by, for instance, the UK Telephone Preference Service
(www.tpsonline.org.uk/) and E-Mail Preference Service, again in com m on with fixed
telephony. Mobile networks also undertake unilateral action.
6.4. Analysis of Adult and Illegal Content Regulation
In the position paper presented in N ovem ber 2003, Ahlert, Alexander and Tam biniexplain that:
375 ICSTIS (2003) Annual Activity Report 2002 at p7. Note that complaints increased by 60%from 2001to 11,552 :
http://www.icstis.org/icstis2002/pdf/ACTIVITY_2002.PDF
127
Major concerns for the self-regulatory fram ework include adult content (porn),
interactive services, unsolicited m essages, com m ercial transactions, location based dating,
gam bling, and peer-to-peer. The m ajor necessary strategies for dealing with these
concerns are content rating, filtering and blocking, notice-and-take-down procedures,
public awareness, cooperation with the governm ent.376
They identify the scale econom ies which m ake effective and sustainable regulation of the m obile
Internet possible:
The em erging m arket for 3G services in the European U nion will be dom inated by a few
m ajor m obile network operators, which in theory should m ake self-regulation a realistic
and viable alternative to state regulation. U ncertainty of actual consum er uptake despite
projected high popularity of 3G , m edia convergence, and the evolving EU regulatory
fram ework all offer incentives for 3G m obile operators to invest in self-regulation.377
N ote in the value chain below several innovations com pared with the fam iliar fixed line Internet
value chain. First, for services in the m obile portal, there is a strong contractual sanction for
MCPs failing to fulfil their self-regulatory duties, which in the fixed environm ent is true only for
the largest portals, such as MSN , Yahoo! and AOL. Second, the pre-pay user has no regulatory
sanction from the MSP, with no contract and no billing relationship, though the MSP could
discover the identity and block service to the SIM card of users if unacceptable use is discovered.
Third, the type of network control at the institution of work/research/education that the public
access layer establishes, is not relevant in the m obile environm ent except in the case of group
contracts for m obiles given to em ployees.
Even before the start of the value chain there is a fourth critical difference: the MSP owns the
network and can control the content flow onto networks in a m anner unfam iliar to narrowband
fixed ISPs. Therefore the lack of control over end-users is replaced by a control over the network.
This is a critical change from end-to-end where control m ust be exercised close to or at the end-
device, in that m obile networks can institute control in the network itself, are required so to do
for law enforcem ent purposes, and choose so to do to stop spam overwhelm ing the network.
That is not to suggest that as a policy choice such a radical departure from fixed Internet
regulation is to be recom m ended, not least on speech grounds, but it does represent a different
architecture of the ISP-network provider relationship.
Mobile CP Mobile Public Device

Program m e codes Mobile SP Pre Pay U ser access
Control
376Ahlert, Alexander and Tambini (2003) European 3G Mobile Industry Self-Regulation: IAPCODE Background Paper for World Telemedia Conference at
filters,
Code of
2: see http://www.selfregulation.info/iapcoda/031106-mobiles-revised-bckgrd.pdf
Editorial & Ethical Term s of N ot awareness,
377 Ibid at 2. Conduct
G uidelines zoning
Privacy service Relevant –
Contract Effect U nknown
Policy
128
The m obile industry is identified as learning lessons from a decade of Internet regulation, and
from earlier experience in Japan and South K orea, where 3G networks have operated since
2001/2. The early experience of spam , child prostitution and peer-to-peer breaches of copyright
were noted prior to the drafting of the first European Code of Conduct, that in the U K 378. It is
predated by the broad content controls instituted across all electronic m edia distribution
networks in G erm any379. N ote that sm aller m arkets in the sam e language norm ally follow the
larger: hence Austria and Switzerland follow G erm any, and Ireland follows the U K 380.
6.5. The U K Case Study: Mobile Content Code of Conduct
The U K Code was drafted by a com m ittee including all six U K network operators and virtual
operators (3, V odafone, Orange, T-Mobile, V irgin Mobile, O2). Inform al consultation with
content providers, infrastructure and handset suppliers and governm ent at national and European
Com m ission levels took place and with IAPCODE381. The Code itself was written in the
‘regulatory vacuum ’ of 2003 as the new super-regulator OfCom was being established, against a
background of discrete coordinated lobbying by m obile networks, and pressure for self-regulation
from Parliam entary debate during the Com m unications Bill 2002-3. There was therefore a
com bination of regulatory com m itm ent (fostered by cooperation in 2002-3), resource freed by
the handover period from existing regulators to OfCom (second half 2003), and political pressure
to establish a workable regim e prior to the broad 3G launch in 2004.
The six operators include all four of the largest pan-European operators382. A draft was presented
for public consultation prior to the full publication of the Code in January 2004. Details of the
Code’s im plem entation (see below) were announced on 7 February 2005 with the launch of the
Independent Mobile Classification Body (IMCB)383. The Code itself is unrem arkable, but its ex
378 Mobile Network Operators (January 2004) UK code of practice for the self-regulation of new forms of content on mobiles available at www.orange.co.uk
379 See Ahlert et al at p20 on the KJM interstate regulatory commission, implementing the 2003 Interstate Treaty on the Protection of Minors and Human
Dignity in the Media
380 Interviewee from Austrian 3G network licensee, London 30 January 2004.
381
Dialogue continued after the launch of the Code: European networks, content providers, consumer
groups and regulators met in London 29-30 January 2004 for the ‘Delivering Adult Mobile Content
Responsibly’ conference which IAPCODE chaired for Total Telecom. Network operators were also
present at the Safer Internet Action Plan’s Safer Internet Day event on 6 February.
382
56% of the 2000 European subscriber market was O2, Vodafone, T-Mobile and Orange – TIM and
Telefonica Moviles, with less significant interest outside their domestic markets, are small in pan-
European terms. See Ahlert et al (2003) at p4. The proposed 2006 merger of Telefonica Moviles with
O2 creates an even larger standards consortium for mobile content.
383 See Classification Framework at http://www.imcb.org.uk/assets/documents/ClassificationFramework.pdf
129
ante adoption, prior to m any adult services being known to the general public, is exceptional and
reflects high awareness in the sector both of potential harm s and of the value of self-regulation. In
part, this can be attributed to the m arket size and regulatory resources of the four giant
com panies behind the drafting.
The m ain points of the code are:
• All com m ercial content unsuitable for under-18s will be classified as “18”, and will only
be m ade available to custom ers when the networks are satisfied that the custom er is 18 or
over.
• The classification fram ework will be com parable to those applied to other m edia, and
will be created by a body independent of the m obile operators.
• Chat room s available to under-18s will be m oderated.
• Parents and carers will be able to apply filters to network operators’ Internet access
service to restrict the content available via a particular phone.
• Mobile operators will work to com bat bulk and nuisance com m unications.
In addition, the Code observes the sam e ‘notice and take-down’ requirem ents with regard to
illegal m aterial as those applying to fixed-line ISPs. Thus Section 3 of the Code states:
“Mobile operators will work with law enforcem ent agencies to deal with the reporting of
content that m ay break the crim inal law. W here a m obile operator is hosting content,
including web or m essaging content, it will put in place notify and take-down
provisions.”
There are, however, several lim itations on what the Code covers. The U K Code explains that:
“The Code covers new types of content, including visual content, online gam bling,
m obile gam ing, chat room s and Internet access. It does not cover traditional prem ium
rate voice or prem ium rate SMS (texting) services, which will continue to be regulated
under the ICSTIS Code of Practice.”384
N or does it cover wider Internet content not directly supplied by third parties to the m obile
operator. Responsibilities here m irror those of fixed-line ISPs. However, the Mobile
Entertainm ent Forum (MEF), a trans-Atlantic grouping of over 70 content providers, has issued
its own Mobile Code of Conduct, dealing with prem ium content. This m ay prove a precedent
for a Code dealing with adult content385. The code also fails to cover issues which have already
stim ulated m edia concern such as the use of cam era phones and Bluetooth technologies for
384 UK Code at p2.

385 The Code was launched in the UK on 19 January 2005, and worldwide on 15 March 2005: see http://www.m-e-f.org/news032005.html
130
content creation and distribution that does not require downloading from a website, or other
form s of P2P file-sharing.
The extent of likely Internet filtering by m obile operators is som ewhat unclear. U nder Section 4,
operators pledge to ‘continue’ to take action against spam – they already have prevented m uch
content arriving on-net. To at least this extent, then, Internet content is to be filtered. Further,
although the Code com m itted each operator to introducing an adult content filter only V odafone
has so far fulfilled its com m itm ent, with the U K ’s other operators m issing their agreed end of
2004 deadline.
The U K Code for Content does not have a central arbitrator for disputes: “Each m obile operator
m ay choose or need to use different organisational and technical solutions to enable it to m eet
aspects of the Code.” The content schem e is an opt-in self-classificatory schem e overseen by an
independent classification body386. Content is classified as ‘18’, adult content, or not – with
optional interim ratings for younger children (in Section 7). Enforcem ent of the Code is form ally
dependent on individual operators and it is unclear how this will be recorded and publicized:
“Each m obile operator will enforce the term s of the Code through its agreem ents with
com m ercial content providers.” However, in practice the IMCB is likely to set precedents for all
operators: the IMCB has to date issued no rulings.
The six operators have consulted widely with stakeholders, and intend to convey their Code to
custom ers, already having the Code on their websites, though navigation to the Code requires a
specific search387. Encouragingly, the need to ensure brand awareness is not tarnished by adverse
press and publicity is likely to m ake this a central elem ent not only of a m inor com pliance budget
(as with m obile phones and child exposure to radiation) but part of the m assive m arketing budget
of the operators. The involvem ent of consum er groups in im plem enting redrafts and
im plem entation of the Code is not known.
This approach covers m ost of the potential areas of concern for parents, and dem onstrates to
governm ent that the industry has taken its corporate responsibility seriously, but does still leave
unanswered som e im portant questions. Issues include how to build a relationship of cooperation
between m obile operators and com m ercial content providers, particularly sm aller non-MEF
m em bers, raising awareness of the code and the role of retailers. Also, the code is heavily
dependent on age verification procedures, which are still far from fool-proof and are open to
fraud.
386 See http://www.imcb.org.uk/

387
See for instance http://www.orange.co.uk/about/regulatory_affairs.html - the Code is the first
download item in the middle of the page.
131
It is also worth noting that age verification procedures to be applied by the m obile industry are
clearly only of use where a phone is being purchased or a new contract established; in so far as
m any children m ay just inherit or borrow phones from other fam ily m em bers or friends, it is vital
that the code and the child protection m easures available are publicized as widely as sim ilar
m easures for PC-accessed Internet. In other words som e responsibility m ust lie with parents and
carers, and not just governm ent and industry.
It also still rem ains to be seen just how successful content rating system s will be. The U K body
responsible for setting content rating guidelines, the Independent Mobile Classification Body was
only launched in February 2005 and whilst it is initially up to the content production industry to
label their services correctly, it will be m obile operators who are ultim ately responsible for
ensuring the appropriate rating of the content they carry. It rem ains to be seen how this
relationship will pan out but such a contractually required labelling system for “com m ercial”
content com ing from third parties should work, because inappropriate content is to be filtered
out at the “gateway” between the network operator and the provider. Content on the open
Internet will pose larger problem s. Therefore, it is likely that the m ost effective approach will be
to com bine the utilization of filtering software, content labels, and U RL block lists388.
In Japan, evidence of public concern leading to legislation to outlaw spam subject to opt-in, and
to outlaw children’s access to dating sites, has em erged. ICSTIS, the U K prem ium regulator and
V odafone com m issioned research in 2003 to identify consum ers’ attitude to prem ium rate
m obile services. It dem onstrates overwhelm ing approval for child access controls, and own-access
controls389.
Table: Consum er Attitudes to Mobile Content 2003 (ICSTIS/V odafone)
388
Zittrain, J. and Edelman, B. 2004. '
Documentation of Internet Filtering Worldwide', pp137-148 in
Hardy, C. and Moller, C. eds Spreading the Word on the Internet : 16 Answers to 4 Questions, Vienna,
OSCE.
389 Beaufort International (2003) Premium SMS Services Research, at p5, at
http://www.icstis.org.uk/icstis2002/pdf/SMS_RESEARCH_REPORT_MAY03.PDF
132
90%
Percentage of Mobile U sers

80% Yes No
70%
60%
50%
40%
30%
20%
10%
0%
Prem ium All Consum ers U pper Price Aware of Approve Child Approve
Content U sers on Child Lim it for Technical Access Controls on
on Child Access Children Controls Controls Own Access
Access
Since the U K Code was introduced in 2004, two other European countries have introduced their
own variant (Ireland and Italy), whilst dialogue is ongoing in at least one other state (France) as
to the desirability and feasibility of such a code. Australia has been holding a public consultation
around proposals for changes to its rather m ore heavily regulatory fram ework for dealing with
content delivered via m obile devices, whilst, in the U S, Federal Com m unications Com m ission
has seem ingly asked for an industry-led m obile and wireless industry education cam paign for
parents, but also suggesting that carriers review the adequacy of their existing code of conduct.
Internationally, policym akers are sizing up to the opportunities and risks presented by the growth
of 3G services and the increasing ubiquity of Internet-enabled m obile phones and other devices.
In Japan, evidence of public concern leading to legislation to outlaw spam subject to opt-in, and
to outlaw children’s access to dating sites, has em erged390.
6.5.1. Six Differences Between Fixed and W ireless Internet Regulation
The Code itself is unrem arkable, but its early adoption, prior to m any adult services being known
to the general public, is rem arkable. In part, this can be attributed to the m arket size and
regulatory resource of the four giant com panies behind the drafting. However, there are also vital
concerns that m ake the adoption of content controls by the m obile industry both different to
narrowband ISPs and potentially a forewarning of broadband fixed ISPs’ role: this Code is
actually the first BSP Code.
390 See ITU (2004) Shaping the Future Mobile Internet Society: The Case of Japan, Document SMIS/06, authored by Srivastava and Kodate, at
http://www.itu.int/osg/spu/ni/futuremobile/general/casestudies/JapancaseLS.pdf at p50: (2002, July) Law on Regulation of Transmission of Specified
Electronic Mail – spam law, and (2003, September) Law of regulating the act that attracts children using the Internet opposite-sex introduction sites – anti-
paedophile law.
133
There are several features of m obile phones which m ake such concerns either m ore and less
pressing. For exam ple, risks m ight be decreased in the sense that it is easier for network operators
to adopt content controls, such as filters, as they are the only gatekeeper to the Internet for
individual users of their services. In addition, network operators also have influence over which
online services will be available to consum ers as they currently have a degree of control over the
user interface on their handset due to their role in providing the software and operating system s
in conjunction with the handset m anufacturers.391 An increased risk is that children’s use of
m obile phones is m uch less open to supervision by parents and educators and m ight therefore
pose a greater risk than PC-based access to the Internet392. These nuances can be sum m arized as
follows:
1. U biquity: given the increasing pervasiveness of colour screen technology in even standard
m obile phone m odels, m any secondary and even prim ary school students are likely to
have phones with colour screens at the birth of the wireless Internet, whilst m ost children
have only gained access to the Internet via PCs at a later stage in that technology’s
developm ent.393 This m eans the need for protective m easures is pressing.
2. Supervision: unlike PC-based access to the Internet, m obile use is m ore likely to be
private and by its nature is m ainly unsupervised. This m ay be tem pered to som e extent
by the possibility of parental m onitoring of item ized phone bills, although in the U K as
elsewhere in Europe, large num bers of m obile users have Pay-as-you-go accounts, with
under-16s especially likely to have such accounts;
3. Control: with PCs, access to the Internet is provided and controlled by an Internet
Service Provider (ISP) and users can choose which ISP they contract with after buying
their PC. Such choice and com petition am ongst ISPs m eans that individuals could easily
opt in or out of various filtering options by shopping around am ongst ISPs. In the case
of m obile-accessed Internet, however, a handset is usually bought as part of a contract
with a particular operator. Even if it is practically possible to change network, this is not
som ething which m any custom ers would do on a regular basis. So long as this rem ains
the case it is easier for network operators to adopt content controls, such as filters, as at
391 This is not the case with all newer devices, and indeed may change in the future if we see increasing convergence between palmtop PC and mobile phone
functions. For example, some new PDAs and some smart phones, run on Windows CE with voice communication as an additional feature incorporated into
this Microsoft operating system.
392 Categories adapted and extended from Marsden (2004)
EDITORIAL... Illegal and Offnet: Is Peer to peer the next
Mobile Regulation Challenge? at http://www.selfregulation.info/iapcoda/0402xx-selfregulation-
review.txt.
393 Figures from the UK Department for Education and Skills show that in 2002 41% of children between 5 & 18 owning a mobile phone, a figure which is
likely to have increased still further in the past two years (DfES 2002). A recent consumer survey suggested that mobile phones are owned by over 5 million
under-16s in the UK. (MobileYouth 2005).
134
any point in tim e they are the only gatekeeper to the Internet for individual users of their
services. This feature, com bined with the different m odels for content delivery described
above m eans that m obile network operators can and do provide so-called ‘walled
gardens’, which effectively lim its Internet access to content approved by (and financially
benefiting) the network operator.
4. Filtering Defaults: it was widely expected that on m obile phones, filtering defaults when
available would largely be opt-in, unlike opt-out Internet Explorer, AOL and G oogle,
m eaning that 2.5/3G m obile phone users would by default have access to adult content.
In the U K , this position has been reversed, in large part due to lobbying by child
protection groups, m eaning that those purchasing new m obile phones will now usually
need to opt out of filtering applications, which will only be possible after age verification.
5. Convergence of Capture and Distribution in One Device: m ost devices now offer digital
im age capture capabilities and also enable distribution of these im ages - picture
m essaging is an exam ple of this. This m eans that in principle, the distribution of
inappropriate pictures, or even pornography is only ‘One Click Away’ from digital im age
capturing; but in a way that cannot be controlled by filters at the network level. This is a
potential loophole in the widely acclaim ed filtering strategies currently used by network
operators, although it rem ains to be seen whether this will be a significant concern in
practice.
6. Peer-to-Peer File Sharing: G iven that 3G bandwidth is still m uch slower than standard
broadband connections, p2p file-sharing of photos, m ovies or m usic is still unlikely as it
is tim e consum ing and costly However, as operators are now starting to offer seam less
roam ing packages whereby m obile phones can be used at hom e with standard wireless
broadband connections and on the m ove with wireless hotspots, P2P m ay further drive
usage of m obile-accessed Internet by children.
Technically, the m obile Internet can be m uch m ore restrictive than fixed Internet use. 3G
services do not allow wider access to the Internet, lim iting access to a ‘walled garden’ of online
services. This m ay restrict uptake but does effectively elim inate networked P2P problem s of illegal
file-sharing via the Internet (though not by Multim edia Messaging Services [MMS], such as the
picture m essaging used in an infam ous Irish schoolgirl pornography case of 2003/4394) even
though this is perhaps an over-reaction to existing and em erging problem s.
394 See http://www.out-law.com/php/page.php?page_id=irishpoliceinvesti1074855914&area=news
135
Sim ilarly, filters m ay prove overly effective. Filtering technologies in use are m uch m ore effective
in m obile phones – if the m easurem ent for effectiveness is that som e adult sites are sim ply fully
blocked – than in the traditional fixed line environm ent. However, early anecdotal evidence also
suggests that som etim es over-blocking occurs and perfectly legitim ate services such as Hotm ail or
other Internet sites are not reachable. At the sam e tim e it rem ains to be seen how easy it is to get
around the filters, as this will be one of the m ost crucial points in evaluating self- and co-
regulatory approaches to m obile content.
Another difference, in term s of policy concerns, is history: the 3G debate profits from m any
lessons learned over a decade of Internet availability. Therefore broadcast video on 3G , whether
tim e-delayed or live-stream ed, presents fewer problem s than the early video-over-broadband
debates circa 1995 because of principles which were laid down then: don’t regulate directly but
expect networks to observe watersheds and adult content rules wherever possible. Revisions to the
EU ‘Television W ithout Frontiers’ debate are being discussed just as video over m obile appears,
but m obile platform s are currently excluded from direct regulation.
6.5.2. Free Speech Only W ithin a W alled Proprietary G arden?
Technically, m obile Internet on-net can exclude off-net and really ‘wall in’ m obile Internet users.
That would elim inate P2P including pornographic im ages by banishing m obile users from the
wider Internet, which appears an over-reaction to existing and em erging problem s, which creates
restrictions on speech freedom s. It is the approach which was initially taken by all U K operators,
which in Septem ber 2005 (T-Mobile, 3) finally perm itted 3G broadband m obile users to access
the open Internet.
Broadcast video on 3G , whether tim e-delayed or live-stream ed, presents less problem s than the
early video-over-broadband debates circa 1995. That’s because of principles then laid down:
don’t regulate directly but expect networks to observe watersheds and adult content rules
wherever possible. Revisions to the EU ‘Television W ithout Frontiers’ debate are being discussed
just as video over m obile appears, but m obile platform s are currently excluded from regulation.
Legitim ate adult content will be a m ajor driver for wireless Internet profits: adult content
filtering (V odafone’s Content Control bar launched this spring, bango.net’s filters) will be opt-
in, not opt-out. Legitim ate networks and content owners need to protect ‘on-net’ brand and
lim it liability from the porn and P2P piracy that ‘off-net’ users and cowboy site operators will
create. Com m ercial adult content on the Internet is driven by the free referral site m odel: free
porn in sm all doses leads to paid porn on ‘official’ sites.
136
Illegal and harm ful content can be entirely user-generated and distributed. Legitim ate 3G
content owners m ake no m oney from off-net P2P adult content; because the receiver pays
(unlike in fixed telephony and ISP access) and networks do profit per bit, m obile operators are
perceived by consum ers to have a higher duty of care. I am not suggesting that peer-to-peer
should be regulated by networks on behalf of governm ents driven by m oral outrage at well-
publicised child porn cases. However, evidence from Japan suggests that the m obile Internet has
played a part in paedophilia and child prostitution and worse. G iven the personalisation of
Internet technology am ongst children, that is to be expected.
Protecting children from profitable adult content in walled gardens is the first, and very least,
that networks can do. They have learnt from the fixed Internet in taking such action early. But
m ore will have to be done, especially with Com m unication: educating consum ers and
introducing effective filters early. The opt-in filtering of m obile networks will help (though
bango.net reports that 1% of its 5m illion credit-carding users have opted to avoid adult content).
N etworks still have the dilem m a of acting on the distribution of P2P illegal and harm ful content,
the distribution of pirated and adult content. N etworks have engaged in the debate early to m ake
sure that regulatory action is reasoned, not tabloid-generated.
Co--Regulation of Internet Content
6.6. Conclusions: A Long Future for Co
There m ay be still be significant hurdles for workable Codes of Conduct to overcom e: it rem ains
to be seen whether there is any preferable alternative. Certainly it is hard to im agine the m obile
or content industries welcom ing a m ore directly regulatory solution. Self or co-regulation is also
likely to be the m ost appropriate response in the context of rapid technological advance; it will
alm ost inevitably be easier for industry groups to assess the im plications of such change and to
revise their Codes of Practice accordingly. Self-regulation would have the benefit of being a m ore
m oderate response to the problem , but co-regulation would provide m ore transparency,
accountability and room for public and governm ental engagem ent. At the governm ental level,
the EU Com m issioner V iviane Reding has again m ade clear her com m itm ent to co-regulation in
the m obile com m unications m arket395, although this is not necessarily supported at the level of
all European states: Spain, for exam ple, has so far shown little appetite for controlling children’s
access via m obile phones to inappropriate m aterial on the Internet. The latter point is im portant
to note. There will inevitably be disagreem ent between states as to the extent of the risk posed to
under-18s by m obile phones and Internet access. W hat is deem ed to be inappropriate or even
harm ful in one country m ay be regarded as com pletely unproblem atic in other European states.
395 V. Reding ‘Mobile Communications : a key driver to make Lisbon succeed’ delivered to the 3GSM World Congress, Cannes, 14th February 2005.
137
To this extent, the co-regulatory approach is an appropriate one, allowing public and
governm ental input to ensure a degree of variation between the Codes applied in different
countries396.
To be workable, Codes m ust first and forem ost be the result of genuine dialogue between
governm ent and industry with room for m eaningful (rather than m erely trivial) periods of
consultation with non-governm ental groups such as child protection or civil rights groups and
the general public. Secondly, the establishm ent of such Codes m ust be clearly understood both by
those who are lim ited by its principles and those who seek its protection. Finally, it is essential
that such Codes be backed up by the creation of clear lines of accountability and m onitoring.
G iven the im portance of the rights and privileges that are protected and lim ited by these
principles, it is essential that these are treated with respect.
Co--Regulation in the Media
6.6.1. Analysis of Successful Co
The vital lessons from co-regulatory studies are:
Co-regulation is a m oving target – the national and sectoral tem plates for co-regulation have to be
m odified following each survey in order to encom pass the different and dynam ic practices of co-
regulation in each geography and sector exam ined. This m akes continued experience of
designing and im plem enting co-regulatory surveys essential – law in books is of little assistance
in so inform al and dynam ic a field;
Codes of conduct, in order to be legitim ate, credible, transparent and effective need to include
clear and workable procedures for review and am endm ent of the code. Ideally this should include
som e input from the adjudication body. The m ost effective and skilled code operators take the
following issues into account when revising their codes:
• the convergence of national, regulatory and corporate cultures;
• the changing nature of the relationship between governm ent and industry;
• the evolving technological architecture that underwrites self-regulation;
• the further developm ent of standards, Codes, and rules; and
• the growth and change of cultural norm s and of public understanding surrounding self-
regulation.
• third party consultation or audit.
There are general recom m endations which specifically can help the effective developm ent of
m edia Codes of Conduct, and co-regulation of Internet content. Technological progress brings
396
See further Ahlert, Marsden and Nash (2005) Protecting Minors from Exposure to Harmful Content
on Mobile Phones, for European Internet Coregulation Network, at
http://network.foruminternet.org/article.php3?id_article=24
138
about change and co-regulation can respond m ore rapidly and efficiently than state regulation.
There is no universally acceptable recipe for successful co-regulation, as regim es m ust be adjusted
to the needs of each sector and other circum stances (technological change, changes in policy to
respond to changes in technology, a country’s legal system , case law of European courts, and so
on).
The trend is towards continued delegation (with regulatory authority audit of the resources,
procedures, transparency, stakeholder participation and m arket effect of the self-regulatory
schem e adopted).
Adequate resourcing is the key to successful co-regulation. Policy on co-regulation m ust take into
account a broader view of the sustainability, effectiveness and im pact on free speech of self-
regulatory codes and institutions. Auditing procedures establish sustainable co-regulatory
institutions and codes. N othwithstanding the centrality of speech freedom s in constitutions, this
regulatory audit burden is a m inim al price to pay for effective co-regulation in the public interest.
Significant econom ies of scale are likely to be realized through functional integration of certain
key aspects of the content regulation value chain horizontally across sectors and across EU
Mem ber States. Com puter gam es rating by IFSE (PEG I) has illustrated the potential for
developing a com m on pan-European ratings structure. G erm any (K JM) and the N etherlands
(N ICAM) operate a cross-m edia rating and labeling schem e. In a situation of increasing cross
border trade within the EU , this trend is set to continue. Although the legislative role of the
European institutions in the m edia sectors is currently lim ited (with non-ratification of the EU
Constitution), several recom m endations can be m ade. And it is likely that in a single m arket
context, there will be a significant self-interest on the part of industry in co-regulation. More
research and developm ent, benchm arking and technical assistance in dissem inating best practice
between Mem ber States is clearly essential to assist industry bodies in the exploitation of
econom ies of scale and scope in co-regulation across the various converging m edia sectors in the
single m arket, and to ensure greater effectiveness of co-regulation.
The general trend is towards an expansion of scope of co-regulation, often at the expense of
statutory regulation. IRAs such as Ofcom in the U K are exploring the possibility of ‘sunsetting’
particular regulations in the event that co-regulatory alternatives can be found. W here there is a
clear industry interest in co-regulation to im prove m arket penetration, or to head off threats of
statutory regulation, there are adequate m arket incentives for resources to be allocated to co-
regulatory activities. However the calculation of enlightened self interest required is vulnerable to
changing personnel and m arket structures such that co-regulatory institutions, where they do not
139
have access to com pulsory funding, will not enjoy the funding necessary to m eet standard
requirem ents of transparency, accountability and due process. A wide variety of m odels of co-
regulatory tools exist. Som e of these are based on adequate standards of transparency, inclusion,
due process, resources, and som e clearly are not. As a result there is som e concern with the
developm ent of codes that insufficient standards apply to both law enforcem ent/child protection
and protection of freedom of expression rights. If these m echanism s are im properly structured we
can expect public harm to result in the m edium term .
140
Chapter 7: Standards and the Internet
“If code is law then standards bodies are governm ents. This flawed but powerful
m etaphor suggests the need to exam ine m ore closely those standards bodies that are
defining standards for the Internet.”
L. Jean Cam p (2004)397
7.1. Is the Internet broken?

Is the Internet in crisis? On 1 Febuary 2006, it is accessed by 1,000,000,000 people, has alm ost
20% (200m illion) of those on broadband connections, and has transform ed m any industries,
notably com m unications and finance. Superficially it appears the m ost rem arkable new
technology ever invented – even m ore so than the m ore ubiquitous m obile phone. However, the
current version of the dom ain system , Ipv4, has not been upgraded to Ipv6 in the decade or m ore
that it has been com m ercially available. The Internet is only suitable for Rom an character types,
creating frustration for Asian, Arabic and other non-Rom an languages, which have introduced
Ipv6 and are currently introducing dom ain nam e registries and other logical infrastructures for
‘rival’ Internets. As Crowcroft and other Internet pioneers explain, even without the security
threats from spam , viruses and Denial of Service (DOS) attacks, the Internet is fragm enting
fast398. The threats issued by other governm ents to the U nited States over control of the root
server system in the Tunis m eeting (see Chapter 6) were not entirely em pty399.
The techno-econom ic argum ents for faster progress in Internet developm ent relies on relatively
unim peded developm ent of the com m ercial access (telecom s) physical layer, com bined with
governm ent-funded developm ent of the logical layer – the Internet protocols. The need for a halt
to the levels of viruses, Trojans and particularly hijacked networks of Personal Com puters (PCs)
397
Camp and Vincent (2004) Setting Standards: Looking to the Internet for Models of Governance, at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=615201
398
This is despite earlier predictions that open source software and more global standards could
maintain progress in interoperability. See Crowcroft, J. and Philips I. (2001) TCP/IP & Linux Protocol
Implementation: Systems Code for the Linux Internet, Wiley & Co.; see also Eltzroth, Carter (1999) E-
Commerce: Open for Business, Study on the Classification of Standardisation Requirements in
Electronic Commerce for World Bank and the European Commission DG MARKT
399
See Wall Street Journal front page article 19 January 2006.
141
is im m ediately apparent400. The Cam bridge-MIT Instititute’s Com m unications Research

N etwork (CRN ) Denial of Service W orking G roup (DOS W G )401 has discovered that the
strength of distributed DOS networks has reached the m illions402, in the case of a 2005
N etherlands-based hacker who was prosecuted. The estim ates for the percentage of em ail that is
spam are in the 55-75% range, roughly that of postal m ail in Anglo-Saxon countries403. However,
this is generally pre-filtered by Internet Service Providers (ISPs), with 80% of all em ail sent to
AOL m em bers rem oved as spam in 2005. In Anglo-Saxon countries, with highly efficient spam
filtering system s in place, spam concerns peaked in late 2003 and have since subsided404. Of
continued and rising concern are the m ore sophisticated types of Internet crim e, especially those
involving distributed attacks using ‘hijacked’ PCs.
Clark describes a four-stage developm ent required:
1. “G ive the m edium a basic security architecture -- the ability to authenticate whom you
are com m unicating with and prevent things like spam and viruses from ever reaching
your PC.
2. “Make the new architecture practical by devising protocols that allow ISPs to better route
traffic and collaborate to offer advanced services without com prom ising their businesses.
3. allow future com puting devices of any size to connect to the Internet -- not just PCs but
sensors and em bedded processors.
4. add technology that m akes the network easier to m anage and m ore resilient.”405
Clark is therefore proposing the introduction of a m ore filtered Internet, with the potential for a
change from a ‘best effort’ end-to-end design that has no quality of service and total anonym ity,
to a m ore graduated design with authentication. This profoundly alters the nature of the
network, and creates the probability of widespread packet discrim ination based on security,
speed, cost and other considerations.
400
For definitions, see Edwards, Lilian (2005) Chapter 2 in Edwards, L. (ed) (2006) The new Legal
Framework for E-Commerce in Europe, Hart Publishing, Oxford, at
http://www.hart.oxi.net/pdf/1841134511.pdf'
401
See http://www.communicationsresearch.net/dos-resistant/
402
Estimates vary from 1.5-3million PCs infected and linked in the largest DDOS network yet
discovered. Source: emails within CRN DOS WG on file with the author.
403
Australia Post reports that over half of all postal mail in the US and UK is ‘promotional’ mail,
compared to only 15% in Australia, reflecting very wide differences based on national direct marketing
maturity and consumer acceptance: Australia Post (2005) Annual Report 2004 at
http://www.auspost.com.au/annualreport2004/lett_direct.asp
404
See AOL Time Warner (2005) Press Release
http://media.timewarner.com/media/newmedia/cb_press_view.cfm?release_num=55254499 at 28
December.
405
Clark (2005): “For example, a new design should allow all pieces of the network to detect and
report emerging problems -- whether technical breakdowns, traffic jams, or replicating worms -- to
network administrators”.
142
This change is actually already taking place in broadband ISPs, with ‘packet sniffing’ via new
form s of servers, including blade servers which have far higher ability to determ ine file types
being transferred. The ability to detect packet types allows providers to set different rules for
executable program m e files (.exe) which are often blocked, and m edia files, which can either be
blocked or prioritised. It also enables advanced filtering of unsolicited em ail types, though as with
all form s of filtering or censoring, the technology and design will enable only as m uch accuracy as
the prior rules.
V inton Cerf argues that the problem lies in rewriting bad program m ing, not better Internet
networking:
"I'm not happy with the current state of affairs. I'm not happy with spam ; I'm not happy
with the am ount of vulnerability to various form s of attack. I do want to distinguish that
the prim ary vectors causing a lot of trouble are penetrating holes in operating system s.
It's m ore like the operating system s don't protect them selves very well. An argum ent
could be m ade, 'W hy does the network have to do that?' It's really hard to have a
network-level thing do this stuff, which m eans you have to assem ble the packets into
som ething bigger and thus violate all the protocols. If Dave Clark...sees som e notions
and ideas that would be dram atically better than what we have, I think that's im portant
and healthy. I sort of wonder about som ething, though. The collapse of the N et, or a
m ajor security disaster, has been predicted for a decade now."406
Cerf is then – as befits a form er Senior V ice President of the world’s largest Internet com pany
and largest corporate fraud, W orldCom – sanguine about predictions of disasters ahead.
However, the argum ent that spam , crim e m ore generally and cyber-attacks are prevalent is an
im portant elem ent in the legal and political debate about the Internet’s future. If politicians and
lawyers are struggling to understand the Internet in order to m ake it work within the law, how
can they assist in stopping its very successes – global seam less networking connectivity and
anonym ity – overwhelm ing its twenty-year old infrastructure? U nfortunately the
com m ercialisation and global spread of the Internet m akes it harder to change than in its
research-led past. However, a solution that ‘patches’ the security holes, or adopts a series of
increm ental fixes, is m ore likely than an entirely new structure built on the existing Internet. This
chapter considers why and how the Internet has been regulated for standards in the past, and why
these public goods problem s arise: why fixing problem s becom es m ore difficult and even
im possible as the num ber of actors involved in the problem rises.
406
Ibid at Part 3: http://www.technologyreview.com/InfoTech/wtr_16056,258,p2.html
143
7.2. Applying Standards Theory to Internet Regulation

As stated in the opening chapter, Internet standards have created an ‘end-to-end’ design407, where
the network does not discrim inate between packets and the innovation is placed at the edge of
the network, in PC applications typically408. By contrast, telecom s networks provide intelligent
routers that allow dum b term inal devices, with the intelligence in the core. However, these
distinctions are breaking down as telecom s networks becom e all-IP using the sam e routers
(typically Cisco) as the Internet backbone, and pressure grows from com m ercial providers to
perm it quality of service distinctions between, say, em ail (routed at standard speed) and video
(given superior treatm ent)409. W hile this debate is very current and very loud, this chapter does
not attem pt to enter an opinion, preferring to explain what standards are and why they m atter. It
is for others to claim m ore or less norm ative roles for network design. For instance, Lem ley and
Lessig state:
W e do not yet know enough about the relationship between these architectural
principles and the innovation of the Internet. But we should know enough to be
skeptical of changes in its design. The strong presum ption should be in favor of
preserving the architectural features that have produced this extraordinary innovation.410
They believe this is is of critical im portance to the growth and innovation that the Internet m ight
produce. Lem ley and Lessig state:
By keeping the network sim ple, and its interaction general, the Internet has facilitated
the design of applications that could not have originally been envisioned. And by
keeping the cost of innovation low in the future — especially in the context of
broadband m edia — the Internet should continue to facilitate innovation411.
407
See Saltzer, Jerome W., David P. Reed, and David D. Clark (1984) End-to-end arguments in system
design, ACM
Transactions in Computer Systems 2(4) pp277-288
408
See Gould, Mark (2000) ‘Locating Internet Governance: lessons from the standards process’,
Chapter 10 in Marsden, C. (ed) Regulating the Global Information Society, Routledge, New York;
Kahin, Brian and Janet Abbate (1995) Standards Policy for Information Infrastructure; Lemley, M.
(1999) Standardizing Government Standard-Setting Policy for Electronic Commerce, Berkeley
Technology Law Journal, Vol. 14, P. 745; Lessig, L. (1999) The Limits in Open Code: Regulatory
Standards and the Future of the Net, 14 Berkeley Tech. LJ 2 at 759-770.
409
See Marsden, C. (2001) The Start of End-to-End: Internet Protocol Television 29 Intermedia 3
410
Lemley, M. and Lessig, L. (1999) Ex Parte Declaration Of Professor Mark A. Lemley And
Professor Lawrence Lessig In The Matter Of: Application For Consent To The Transfer Of Control Of
Licenses Mediaone Group, Inc. To At&T Corp CS Docket No. 99-251 Before The Federal
Communications Commission, Washington, D.C. 20554at Paragraph 4. See further Faulhaber, G.
(2002) LBS paper.
411
Supra at Paragraph 22. Though note that end-to-end has prevented many Internet architectural
innovations including Quality of Service. Dave Clark, one of the originators of the end-to-end
principle, has pleaded for less zealotry on the part of lawyers interpreting his work as though it were a
144
However, telecom s com panies as well as Internet broadcasters can raise costs to innovators by the
existence of their relational power to governm ent:
End-to-end design does not only prom ote innovation by creating the opportunity for
innovators to offer services to the network. In our view, the effect com es as well from the
expectation that innovation will not be countered by strategic actors who m ight control
the flow of com m erce. The potential of an actor in the distributional network to act
strategically is a cost to innovation.412
This raising of barriers is neatly sum m arised in AT& T’s reaction to Paul Baran’s original concept
of the Internet:
[AT& T’s] views were once m em orably sum m arised in an exasperated outburst from
AT& T’s Jack Osterm an after a long discussion with Baran. ‘First,’ he said, ‘it can’t
possibly work, and if it did, dam ned if we are going to allow the creation of a com petitor
to ourselves.’413
The end-to-end design is not an accident, but a result of a standards process which has changed
radically in the past twenty years. The chapter that follows considers the institutions that were
changed in this evolution.
7.3. Standards Organisations and Their Analysis
Internet standards policy underpins m uch of the functionality of the Internet, including its
‘layers’ m odel considered in Chapter 2, and its ‘end-to-end’ design. The dynam ic and flexible
developm ent of standards bodies in form al governm ental and inform al consortia and other fora
presents a series of policy challenges. Standards policy m aking is highly com plex in process,
organisational and substantive term s, encom passing m arket and legal as m uch as form al technical
discussions. Analysis of standard setting is interdisciplinary in nature, com prising lawyers,
econom ists, technical experts and political scientists to be fully cognizant of the challenges facing
standard setting. I recognize fundam ental trade-offs in standards policy between interoperability
and security, between stakeholder inclusiveness and dynam ic innovative standard setting and
between interoperability and IPR policies. The com plexity of the challenge of analyzing best
practice and m apping future developm ents is acknowledged.
‘law’: see Clark and Blumenthal (2001) in a very thoughtful restatement of the ideas:
http://itc.mit.edu/itel/docs/jun00/TPRC-Clark-Blumenthal.pdf
Clark more recently expressed his views that end-to-end anonymity was actually a bad mistake,
particularly in the spam context. See: http://www.cambridge-mit.org/cgi-
bin/default.pl?SID=5&CALEVID=192
412
Supra n.23.
413
Naughton John (1999) A Brief History of the Future 107. See further Hafner, K. and Matthew Lyon
(1996) Where Wizards Stay Up Late pp52-66. See Lemley supra para 27.
145
Analysis of Internet standards requires assessm ent of a range of venues, from the m inutae of
process and strategic interactions within and am ong bodies as diverse as Digital V ideo
Broadcasting G roup and the Internet Architecture Board, to m acro-level assessm ent of the place
of research and developm ent in the future developm ent of the Internet. All these analyses m ust be
placed within the context of legislative m andates, and the globalisation of standardisation
processes and the Internet itself that will accelerate in the next decade. The analysis m ust be built
on an analytic understanding of the past and present, but m ust equally look to the future in ways
that draw out the knowledge and judgem ent of key stakeholders.
First I list the standards bodies m entioned below as a list of acronym s – acronym s are inevitable
with such a diversity of technical bodies, and we use them extensively. In the exam ples, I look for
lessons from the counter-intuitive. It is very straightforward to assess procedures or m em bership
levels in order to assess the prior success of a standards body, and these qualitative and
quantitative m easures will be em ployed. However, it is also im portant to explore the lim its of the
form al processes, and the inform al constraints and pressures that prevent successful outcom es414.
The criteria for success are typically internally generated in the prior outcom es of standard setting
for that body, in contrast to potentially rival fora.
Table of Acronym s for Standards Bodies
Acronym Standards Body
DV B Digital V ideo Broadcasting Project
EBU European Broadcasting U nion
ETSI European Telecom s Standards Institute
IAB Internet Architecture Board
IEEE Institute of Electrical and Electronic Engineers
IETF Internet Engineering Task Force
ISO International Standards Organisation
MPEG Motion Picture Experts G roup
Other acronym s used are ICTs (Inform ation and Com m unications Technologies), IT
(Inform ation Technology), IPRs (Intellectual Property Rights), G SM (G lobal System for Mobile
Com m unications), API (Applications Program m e Interface), and CDMA (Code Division
Multiple Access).
414
Here, fieldwork can supply much-needed contextualization of the desk-based empirical work. Given
access to key policymakers and stakeholders in these organisations, I examine that limit on the
organisational and procedural capacity that results in an unsuccessful policy.
146
ent--Diktat
7.4. From G overnm ent Corporate--Led: The N ew Approach to Standards
ktat to Corporate
The rapid increase in innovation driven by ICTs has led to m ore flexible standardisation
processes and organisations since the 1970s, and so has the role of states in standardisation
processes415. States perceive com patibility with international standards as a prerequisite for
national econom ic and social developm ent. European ICT standard setting has been driven by
industry following the form ation of ETSI in 1988416.
I start with a brief discussion of the special features of the Internet in relation to standards:
1. The im portance of interoperating system s (m eaning both that a separate standards layer
covering linkages, data exchange and APIs is needed) and that standards for system
com ponents m ust have som e overall coherence (in term s of governance as well as
content).
2. Relatively low entry barriers in m any parts of ICT m ean that there are a diversity of
possibly-diverse players, so the im plicit de facto standardisation flowing from key patents
or dom inance by single firm s m ay not suffice to keep the sector as a whole ‘in line’.
3. Rapid developm ent m eans that standards m ust be updated, relatively perm issive,
backwards com patible and devised with an eye to future innovation.
4. ICT is infrastructural to the econom y/Inform ation Society. It is thus im portant to
protect the public good of system perform ance in the face of com plexity, divergent
dom ain perspectives (e.g. IT vs telecom s, hardware/software/applications, telecom layers,
com petition ‘on the m erits vs. com petition for the attention of end-users, etc.), possibly
obsolete IPR regim es and free-riding.
5. Because uptake and use depend on trust and beliefs as m uch as ‘real’ perform ance, the
im pression of reliable perform ance provided by good standards is critical.
6. The need to control the tipping tendency. N etwork m arkets tend to ‘tip’ in favour of a
single dom inant standard due to positive externalities: an addition to the network
increases its utility for all existing users417. Inform ation m arkets are network m arkets in
which ‘winner takes all’ strategies often succeed, in which m arket failure is endem ic, and
415
Neil, Suzanne, Lee McKnight and Joseph Bailey (1995) ' The Government' s Role in the HDTV
Standards Process: Model or Aberration', in Brian Kahin and Janet Abbate, Standards Policy for
Information Infrastructure ,pp.276-289
416
Levy, D.AL. (1997) The regulation of digital conditional access systems. A case study in European
policy making, Telecommunications Policy 21 (7)at 661 – 676
417
Lemley, M.A. & David McGowan (1998) Legal Implications of Network Economic Effects, 86
Calif. L. Rev. 479
147
Chart: The Dependents of the New Approach to Standards
InternetNew Approach01/02/2006
Regulation 1988
Confirmed 1998
Review 2004/5
Globalisation
Industry-led
Fast-paced
Innovation-Led
Organisations Co-regulatory process External Impacts

CEN Legal background R&D policy
CENELEC ETSI Finance Regional promotion:
ITU Industry-led Commission/Member State
IP standards International lobbying DVB-NTSC/HDTV Interoperability: MHP

IEEE Multinational support GSM-CDMA Linux
IETF Interoperability WCDMA-Qualcomm DVD-R
ISO - Java Quasi-de facto standards Global standards WiMAX HiperLAN2
in which predatory pricing, bundling of dom inant products with others (tied sales), and
refusal to deal in the IPRs which form s a bottleneck, are com m on strategies418.
Before addressing these substantive issues, I address the context, and the lessons to be learnt from
im plem entation of the European N ew Approach prom ulgated in 1988.419
Discussion of Internet standards generally begins from the form ation of the Internet Engineering
Task Force (IETF) in 1984, overlooking the fact that standard setting was ‘privatized’ not by the
then alm ost irrelevant Internet com m unity, but the dom inant telecom s and ICT stakeholders in
the m id-1980s. The evolution of the N ew Approach since 1988, and its reconfirm ation in 1998,
has not rem oved the process and contextual problem s of standardisation processes, but has added
greater flexibility and com plexity.
In the Chart above, I illustrate organisational procedural, legislative and broad policy im pacts,
with the various challenges represented to new processes. The organisations such as ETSI have
been challenged by even newer and m ore global form s of standard setting fora, such as the IETF
and som e evolving ISO processes420. The legislative field has been com plicated by de facto
standard setting, com petition law challenges and the increasing financial consequences of
standards enforcem ent (consider MS-DOS). W e note the co-evolution of form al and inform al
418
See Shapiro, C. and Varian, H.R.(1998) Information Rules: A Strategic Guide to the Network
Economy, Harvard Business School Press
419
The relevant policy statements are: Council Directive 83/189/EEC on general standardisation
principles; Council Decision 87/95/EEC, defining the overall objective of “end-to-end data and systems
interoperability”; COM (96) 359 final on “Standardisation and the Global Information Society: the
European Approach”; Council Directive 98/34/EC on European standardisation policy, including the
ICT domain; COM (2004) 674 final on “The role of European standardisation in the framework of
European policies and legislation”; SEC (2004) 1251 on “The challenges for European
standardisation”.
420
See the excellent illustration, McGowan, David (2000) ‘The Problems of the Third Way: A Java
Case Study’, Chapter 13 in Marsden, C. (ed) Regulating the Global Information Society, Routledge,
London and New York
148
(and overlapping) standards, both open and closed – and sim ultaneous changes in m arkets,
technologies and their uses. The overall ICT environm ent has itself becom e m uch m ore
globalised, with interoperability and open standard concerns param ount for end-users and sm all
businesses attem pting to take advantage of the increased scale econom ies and vertical
disintegration of ICT value chains. Security and interoperability are perhaps the greatest
challenge facing new types of flexible standard setting in the near future, with device
proliferation, m obility, and system ic insecurity the latest challenges. Standards m ay serve as an
integrator of these potentially conflicting developm ents.
There is a broad literature on the failure of the 1980s industrial policy which governm ents drove
through the EU , and in particular the “fifth generation” com puting challenge of Japan, random
access m em ory chip technology and the developm ent of high definition television (HDTV ). In
addition to the utility of m arket/evolutionary forces for ‘selecting’ and m otivating useful
standards, there are lessons for any policy regim e that attem pts to control or pre-em pt global
m arket forces, to recognise that there are several players and co-ordinate accordingly. There is
power in coordination, but also risk (‘disadvantageous syndication’). The history of these
program m es broadly agrees that the successful enterprises relied m ore on m arket m echanism s to
develop their technologies. The result is sum m arised by Martin Cave: “The [EU ] record in this
regard had been dism al … [following his appointm ent in Decem ber 1992, Com m issioner]
Bangem ann quickly killed off the HD-MAC strategy”421.
The retreat from HD-MAC m arked the end of “essentially a closed – corporatist – partnership
between Com m ission actors and m ajor firm s”422, itself based on strategic responses to Japanese
and U .S. technological advances. Such was the political response to the failure of industrial policy
in HD-MAC that, as W atson-Brown notes: ”Earlier catastrophes have played a strong role in
encouraging collaboration … DV B itself was born from the ashes of HDTV in 1992”423. The
new m odel of standardisation developed in response by broad industry-led coalitions, has been
characterised as follows:
A research or regulator-driven approach to standardisation no longer operates effectively
in industries with com plicated value chains with m any different types of m arket player. If
421
Cave, M. (1997) Regulating digital television in a convergent world, Telecommunications Policy
21 (7) Special Issue: The Economics Regulation of Pay Broadcasting at 575 – 596, at 593.
422
Cawson, A. (1994) Business interests and the regulation of new technology in Europe: the case of
television, 4 Business and the Contemporary World at 73, quoted in Levy (1997) supra n.3 at 664. See
further Borrus, M. and Zysman, J. (1997) Globalisation with Borders, 4 Industry and Innovation 2.
423
Watson Brown, Adam (1999) Industry Consortia and the Changing Roles of Standards Bodies and
Regulators, 325 IPTS Report, Institute of Prospective Technology Studies, Seville, Spain, also through
DGXIII, at http://www.jrc.es/pages/f-report.en.html
149
the regulator places itself at the centre of such a process, m arket actors will concentrate
on trying to m anipulate the regulator.424
ICT standards deliver benefits via system perform ance and both horizontal and vertical
integration. In the following sections we discuss the prim ary policy concerns, beginning with
interoperability and the associated topic of open or closed standards, which both affect the ‘end
to end’ principle.
7.5. ICT Interoperability and Open Standards
Interoperability can be prom oted and standards applied at the technical, sem antic and regulatory
levels. Standardisation is not necessary for interoperability (a series of bespoke interfaces will do it
as well), but it is essential if a level playing field is to be m aintained, and to secure a range of
diverse developm ents, collective approaches to security and reliability, and so on. Standardisation
is an essential tool to foster interoperability but it is insufficient to achieve network
interconnection and interoperability of services at international level. Furtherm ore,
interoperability is a com plex m atter involving various types of actions and various types of actors.
Should standardisation address other “functionalities” such as reliability, usability, efficiency,
m aintainability and portability, subject to policy considerations? W atson Brown explains:
the speed of technological developm ent and the m ove towards software-based
functionality m ean that proprietary solutions always com e first, with openness and
interoperability as a secondary consideration … The flexibility of software solutions now
m eans that industries are less prisoners of their installed base than in the hardware-
dom inated analogue era … regulators therefore need to adopt a m ore proportional
approach to standardisation, m andating only what is absolutely necessary425.
Entry barriers are lower, so custom ers are less im prisoned by proprietary offering and m ore
accessible to alternatives. At the sam e tim e, the incentives to lock them in via incom patibilities
rem ain. This suggests that regulators m ust balance openness with overall coherence – developing
and setting relatively technology-neutral, ‘m inim um ’ com m on standards and periodically
‘cleaning up’ based on m arket experience.
Standard setting can ensure the interoperability and interconnection of networks and services426.
In form erly vertically integrated IT, telecom s and broadcasting, the com plexity of the network
424
de Cockborne, Clements and Watson Brown 1999
425
Watson Brown, Adam (1999) Industry Consortia and the Changing Roles of Standards Bodies and
Regulators, 325 IPTS Report, Institute of Prospective Technology Studies, Seville, Spain, also through
DGXIII, at http://www.jrc.es/pages/f-report.en.html
426
An excellent review of the literature is contained in Besen, S.M. and Saloner, G. (1989) and Noam
Eli M. (1989) in Crandall, Robert W. and Flamm, Kenneth (eds). Note the title which reflects the
150
was hidden from the consum er, and the standards process was an engineering issue internalised
within organisations427, both on national and international levels, where the ITU and EBU set
standards.
Borrus et al describe traditional governm ent-assured standardisation an extrem ely open but
inflexible system : “a form al, public process with governm ent im prim atur”428. They contrast this
with “a Balkanised N II (N ational Inform ation Infrastructure) [which] would have lim ited
interconnection and interoperability”. In both absolutes, open and closed, com plexity is hidden
in the system . Only where hybrids, such as com puting or com petitive telecom m odels develop,
does the extrem e open/closed scenario becom e m ore com plex, and with that, the choice of system
to consum ers becom e possible. N ote that “Interconnection is binary – you are either connected
or not – but interoperability com es in degrees [and] presupposes a higher level of logical
com patibility”429: the higher the com patibility, the greater the interoperability.
Table: Telecom Interconnection vs. Com puter Interoperability
Telecom Interconnection Com puter Interoperability
Standards Appliance/service to network; network to Service to service; service to
interface network appliance
Term inal to phone network; cable W ord processor to operating
telephony to backbone network system program m e; W indows
program m e to Apple com puter
Econom ic G oal Prom ote network effects Reward Innovation
Any-to-any connectivity ‘W intel’ dynam ism
Reduce barriers to entry
Market Bottleneck Regulation of Dom inant Actor Com petition and IPR
Dynam ics and Essential Facilities Doctrine protected; Behavioural
Policy controls on abuse
MS W indows tying to
Explorer
collection’s analysis of dynamic technology-led market entry and regulatory reform in US and
international telecommunications. See Crandall, Robert W. (1989) The Role of the US Operating
Companies, at p115 in Crandall and Flamm (1989) eds. He notes that regulatory reform was a pre-
requisite for the convergence of computing and communications in the late 1980s, forerunner of the
more profound convergence process observable by the period 1993-7. See also the Editors’ Overview
at 1-5.
427
David, Paul and Shurmer, M. (1996) Formal standards setting for global communications and
information services: towards an institutional regime transformation? Telecommunications Policy
20[10] 789-816
428
Borrus, Bar and Steinberg (1996) Islands in the Bitstream: Charting the NII Interoperability Debate,
BRIE Working Paper No.79 at http://www.stanford.edu/¬fbar/islands.pdf
429
Ibid
151
7.5.1. Interoperability: V ideo Exam ple

Explaining interoperability and the associated technical and m arket forces driving towards m ore
closed proprietary standards m ay best be illustrated by an exam ple, from broadcasting technical
standards. In broadcasting, interoperability between devices (e.g. TV receivers) and services is
param ount especially because broadcast standards are adopted for decades. In the late 1990s
DV B took up standardisation work for an MHP, a com m on open application program m ing
interface430. Early success in this area would have allowed the speedy developm ent of a broad
horizontal m arket in Europe of interoperable consum er equipm ent. Such a technology would
have appeal also outside of Europe thanks to the take-up of transm ission and other DV B
standards and the DV B’s ties to other standards bodies through ETSI. The DV B standardisation
process created a standardised environm ent for the developm ent of broadcast digital delivery
system s but those who understood these specifications could erect technical barriers to entry. The
result was a delay, notably by owners of legacy system s. The com paratively slow launch of MHP
was due to the need to consider m igration strategies; develop arrangem ents to diffuse control
perceived to be in the hands of Sun Microsystem s; address, technically, accom m odation of legacy
“plug-ins”; and ultim ately the down-turn in the ICT sector and consum er m arkets. The
continual but unfulfilled prom ise of voluntary m igration to interoperable platform s is seen in the
developm ent of the MHP by DV B:
One of the key objectives is to enable gradual m igration from current proprietary APIs
towards a com m on, open API. Present work favours the use of the Java language
developed by Sun Microsystem s Inc. MHP is intended as an open architecture which
will enable the m arket to m ature. A key objective is to enable the m arket players to
m igrate gradually away from decoders containing significant proprietary technologies
which lim it interoperability.431
U ltim ately MHP m ay have greater success in greenfield m arkets, such as Italy, and in
environm ents where strong com petitive reasons (and com paratively few m arket participants)
drive the m arket, such as U S cable (based on MHP).
In com puting software, open system s would approxim ate to:
• open code, such as U nix;
430
See Cockborne, Jean-Eric de, Bernard Clements and Adam Watson Brown (1999) EU policy on
Multimedia regulation, paper presented on 14 June 1999 at the Montreux Symposium '
99 at
http://www.ispo.cec.be/infosoc/telecompolicy/en/montreux.html
431
Ibid.
152
• open-but-closed would be system s which were proprietary but offered interoperability,

such as Java,
• closed-but-open m ight approxim ate to Microsoft W indows operating system s, in which
only a partial publication of source code is offered to other program m ers. The effects of
patents on interoperability are im portant here, and the approach taken by for exam ple
the Trusted Com puting G roup where standards groups m aintain patent pools where all
IPR essential for interoperable im plem entations is cross-licensed to group m em bers.
• largely closed system s without interoperability would be the initial choice of Apple
Com puters.
The dynam ism and flexibility of the form of self-regulation chosen creates incentives to reach
consensual agreem ent within the standards body rapidly, to prevent pre-em ption by non-
m em bers, especially where barriers to m arket entry for non-m em bers are relatively low.
- In the case of DV B, the m assive installed base of proprietary decoder boxes acted to
encourage first-m overs to prevent agreem ent on a com m on standard, given that m arket
entry was relatively difficult.
The ISO MPEG adoption of the MPEG 4 Internet Protocol video standard was m ore rapid,
given that com peting proprietary solutions risked being adopted by the m arket, engineered by the
two dom inant form ats, Real N etworks’ and Microsoft’s W indows Media 9. Barriers to adoption
are lower as the download of free software onto com puter hard disks over the Internet is an
alm ost zero-cost transaction, in contrast to the hardware installation in the DV B MHP
standard432.
7.5.2. The Role of Consortia and Fora
Standards organisations are responding to the increasing role that consortia and fora are playing.
ETSI launched its High Level review with the objective to re-design the organisation to respond
to the new needs, a process under discussion to be com pleted by N ovem ber 2005433. G overnance
alternatives and the ‘problem ’ of overlapping fora could com plicate the consensual for a ideal
inter-regional standards body and an end to ‘standards wars’ between rival non-interoperable
standards434. Businesses have attem pted to rectify the slow procedures of standardisation processes
by creating 'consortia' defined as 'a collection of like-m inded interests that participate in the
432
Flynn, Barry (2001) 4 Play: In turning its attention to the Web, DVB has placed MPEG4 as a
possible destination on its roadmap, 4 Inside Digital TV 5, March 12 at pp6-7
433
http://portal.etsi.org/ga/GA_ad-hoc_HLRG-IG.asp
434
Grindley, P., Salant, D.J. and Waverman, L. (1999) Standards Wars: the use of standard setting as a
means of facilitating cartels in third generation wireless telecommunications, International Journal of
Communications Law and Policy 3(2) at http://www.ijclp.org/3_1999/ijclp_webdoc_2_3_19999.html
153
developm ent of what m ay be a m arket accepted solution'435. Their m ain difference from
organisations like the ISO is their restricted m em bership, as well as independent rules and
procedures for the developm ent of their own standards. There have been calls for states to
provide technical and financial support to these institutions, and dem ands that the analysis be
developed.
Literature about consensus standardisation processes in the inform ation technology arena has
stated that standards should originate out of negotiations or 'consensus' between potential users
and developers of inform ation technology solutions, since de facto and de jure standards cannot
m atch the principles of open architecture and open, flexible and universal access to the
Internet436. Standard developm ent processes are 'com petitive developm ent procedures' since
businesses participate with well-defined individual objectives and goals that need to be
harm onised through negotiations437. Their participation is often aim ed at m onitoring
developm ents am ongst their com petitors that m ay affect profitability or, in extrem e cases,
com m ercial survival. There is also scope for ‘rules for setting rules’ and for interesting alternative
m onitoring and enforcem ent m echanism s. Standards can be viewed as endogenous norm s, or as
boundaries for a set of interacting agents/applications. On the other hand, they could protect
fitness for purpose and thus overcom e the problem of asym m etric inform ation in the m arket
(where users don’t always know what program m es do – and also where suppliers don’t always
know how their products will be used). This even extends to networks in which players all
contribute and the standards allocate credit and blam e438.
Scholars have focused on trying to reform standardisation processes m anaged by organisations
such as the ISO or AN SI439. Can these institutions am end their procedures to m atch the
dynam ism of the Internet and inform ation technologies in general? N o adm inistrative process can
435
Cargill, Carl and Martin Weiss (1992) ' Consortia in the Standards Development Process'Journal of
the American Society for Information Science, vol. 43 no.8 (September 1992), pp.559
436
Molka, Judith (1992) ' Surrounded by Standards, There is a Simpler View' , Journal of the American
Society for Information Science, vol. 43 no. 8 (Spring 1992), p. 526
437
Weiss, Martin and Marvin Sirbu (1990) ' Technological Choices in Voluntary Standards Committee:
An Empirical Analysis' , Economics of Innovation and New Technology, vol. 4 no.1 (April 1990),
pp.111-133
438
The case of Open Office is instructive here – especially since people volunteer code and naturally
use whatever technology is easiest for them to use – for many, this means writing in Java. The
standard for source code does give the possibility of compiling the application for a very wide range of
operating systems, however. The standard protects the common functionality, and points the way to
porting more advanced features – this is an open example of the ‘first proprietary, then standard’
dynamic. But an example drawn from biometrics shows that acts of generosity can turn out to be
Trojan horses bent on capturing the commons.
439
Rankine, John (1995) ' Do We Need a New Standards System?'in Brian Kahin and Janet Abbate,
Standards Policy for Information Infrastructure, pp. 564-578.
154
guide the developm ent of standards unless it is cognizant of new technical developm ents440 but
technological developm ents are not always exogenous, and a foresighted adm inistrative process
could do better.
7.5.3. Open standards
Open standards ensure m axim um com patibility between software, as long as technical integrity is
m aintained. Increasing com plexity and the enorm ous profits available to private standards have
encouraged attem pts by first-m overs to establish their standard as de facto industry standard, by
prom oting the m anufacture and distribution of the technology, and licensing to downstream
sectors. It is characteristic of ICT industries that early leads are difficult to overcom e, so capture
of the de facto standards am ounts to long-term m arket control. The result of this ‘capture’ is that
long-run com petitive and evolutionary forces driving innovation and im provem ent are distorted
or blunted. W here the m arket structure is not vertically integrated, and is therefore com posed of
separate m arket actors, standardisation involves the co-operation of actors, which m ay be
problem atic where they com pete in horizontal m arkets, or where one or m ore actors have m arket
power, form ing a ‘bottleneck’ at a stage of the value chain. A classic case of this is the Microsoft
W indows operating system duopoly, where the API enables the browser to com m unicate between
the network, the W W W , and the other com puter program m es. Java is an attem pt by Sun
Microsystem s to displace the Microsoft platform in favour of its own, with the strategy of
attem pting to licence the specifications for its Java source code through the ISO (though note its
ultim ate failure to secure a favourable outcom e)441.
7.5.4. ‘Multistakeholderism ’: U ser Focus and Parti

Participation in Standards Consortia
At the core of consensus standards is the need for the accurate reading of the requirem ents of
their potential users, as well as the environm ent where they are expected to be used. Effective
industry-led standards should jointly reflect the interests of state and businesses, which have been
specified as 'beneficiaries', as well as the environm ent in which these actors operate. U ser
participation – whether SME or consum er groups directly – can assist in ensuring the user-focus.
Interoperability and open standards can also address user concerns.
440
Greenstein, Shane (1992) ' Invisible Hands and Visible Advisors: An Economic Interpretation of
Standardisation' , Journal of the American Society for Information Science, vol. 43 no.8 (September
1992), p. 543.
441
McGowan (2000) supra.
155
One of the objectives of the eEurope Standardisation Action Plan was to increase participation of
all stakeholders442. W hich action m ay be taken to further im prove SME’s participation in
standardisation and subsequently their com m itm ent to im plem entation of standards? A
suggestion can be taken from best practice processes in ETSI and IETF 443. In ETSI, there is a
specific requirem ent to include user and SME representatives. By contrast, IETF has form al
equality between participants, and user concerns are only indirectly represented through the
Internet Society representatives. Sim ilarly, DV B has a low threshold for m em bership and
welcom es a wide range of SMEs. However DV B counts on its broad m em bership and their access
to consum er m arkets to identify user interests. W hether direct representation (ETSI) or less
form al flexiblility (IETF) is m ore effective is analysed. Stakeholders’ groups are vocal and active in
certain standards fora, for instance open source standards.
W hile the m arket responsiveness of standards is a well-established principle, the responsiveness of
standards to societal issues, such as environm ent, increase of em ploym ent, security, privacy, is less
evident (although the IETF now for exam ple requires that all standards docum ents contain a
“Security Considerations” section). Functional security of system s, products and services should
be dealt with. How can the participation of societal groups in the standardisation system be
increased? Increased participation by N G Os, as happens with the Centre for Dem ocracy and
Technology (CDT) in the IETF and Electronic Frontier Foundation (EFF) in DV B?
Participation by governm ent bodies such as the Com m ission’s Article 29 W orking Party on data
protection?444 Concerns are aired that dem ocratic accountability is threatened by the increasing
resort to private standard setting bodies with no form al connection to governm ent due process.
W here there is little direct interest in standards fora, participation can be supplem ented by
indirect soundings.
7.5.5. IPR Policies: Consensus related to processes and deliverables
IPR policies applicable within usually closed fora and consortia seem to be m ore attractive for
industry than the policy applied by the ESOs. It should be noted that various international
standards bodies vary widely in the process to ensure the availability of licences for essential IPRs
442
The latest of the series of European statements on standards: COM (2004) 674 final “The role of
European standardisation in the framework of European policies and legislation”; COM (96) 359 final
on “Standardisation and the Global Information Society: the European Approach”; Decision
87/95/EEC defining the overall objective of “end-to-end data and systems interoperability”; Directive
83/189/EEC on general standardisation principles; Directive 98/34/EC on European standardisation
policy, including the ICT domain; SEC (2004) 1251 on “The challenges for European standardisation”.
443
http://portal.etsi.org/Portal_Common/home.asp
444
See Reidenberg, J. (1999) Restoring Americans’ Privacy in Electronic Commerce, 14 Berkeley
Technology Law Journal 2 at 771-792
156
in a fair reasonable and non-discrim inatory m anner, a critical aspect in adoption of standards445.
It is clear that the granting of a state m onopoly for a lim ited period to an IPR holder is a
suspension of com petition in itself446. This raises com petition issues, in that the IPR granted m ay
be the basis for a global ICT system . W here IPR in a standard form s a ‘bottleneck’ facility, it m ay
be that the consum er interest is best served by requiring the owner to perm it access on reasonable
and non-discrim inatory term s, or to bring the consortium into a m andated standards forum to
ensure due process447.
In a converged environm ent, devices m ay use technologies across several industries. The relevant
standard m ay encom pass, through norm ative reference, the specifications developed in a num ber
of fora. As a result, there m ay be a num ber of varying policies applicable for access by
im plem enters to essential IPRs. One solution – advanced for exam ple by DV B – is to foster the
developm ent of joint licensing arrangem ents (“patent pools”), assem bling in a one-stop shop the
IPRs essential for a technology. This approach reduces adm inistrative costs for the im plem enter
and arguably results in lower overall royalties.
V iewed as fences around an area of application, technology or practice, the separate IPR route
m ay involve exclusive ownership, a ‘m em bership-fee’ club (negotiated licensing) or an easem ent
m odel (com pulsory licensing). The pure standards approach starts from a com m ons m odel, and
allows various degrees of enclosure or rights to subsequent developm ent. There are two hybrid
m odels as well – one extends the power of IPRs by prom ulgating standards for which specific
patented technologies are essential; the other strengthens the incentive to develop or im prove the
standard by allowing a tem porary period of proprietary ownership (via closed standards or
patents) before incorporating the IPR into an updated (and m ore open) standard. Paym ent can
be set adm inistratively (by collective bodies that ‘own the standard) or negotiated. There can also
be ‘pay-or-play m echanism s whereby access to the standard (or certification of com pliance) is
provided in exchange for contributions by m em bers to the collective health of the sector.
445
European standards deliverables produced by CEN and CENELEC are subject to copyright and sold
by the National standards bodies. ETSI standards, on the contrary, are freely available for download on
the ETSI web site, and subject to copyright meaning that there is no right to modify them. Taking into
account the public interest in standards, especially in the ICT domain, free availability of standards can
contribute to raise the awareness of standards and the subsequent commitment to the implementation.
446
Barton, John H. (1997) The Balance Between Intellectual Property Rights and Competition:
Paradigms in the Information Sector, European Competition Law Review Issue 7 at 440-445
447
See variously Cave, M. and Cowie, C. (1996) Regulating Conditional Access in European Pay
Broadcasting Communications and Strategies, No.23, 3rd quarter at 119, IDATE: Montpelier; Cave, M.
and Cowie, C. (1998) Not Only Conditional Access. Towards a Better Regulatory Approach to Digital
TV, Communications and Strategies, No 30, 2nd Quarter, 1998 at pp 77-101; Cowie, C and Marsden, C.
(1999) Convergence: Navigating Through Digital Pay-TV Bottlenecks 1 Info 1 at 53-66, Cambridge:
Camford Publishing.
157
7.6. G lobal standards

Inform ation Society standardisation has by definition a global dim ension. The global character of
the Internet will, in order to foster im plem entation of new services and applications, require
consensus on standards. How can consensus on, preferably, global standards and on specific
needs be achieved? How to leverage EU consensus at international level? W hen com petition
between standards occurs, the opportunities for strategic behaviour – and hence m arket failure or
at least coordination problem s - increase. The specific standards-setting exam ple largely focuses
our attentions on regional rivalry in substantive issues. Taking the exam ples of 3G wireless
standard-setting, Coen and G rant claim that “national regim es risk becom ing rule-takers not
rule-m akers”448. This can apply in a variety of settings, and the EU faces a significant risk that –
as R& D is increasingly offshored by European com panies – standards are form ed in em erging
m arkets such as East Asian m obile IP, rather than in Europe. China’s 2003 failure to enforce its
separate security solution for the IEEE 802.11 fam ily of wireless IP standards m ay be seen as the
start of a m ore assertive approach by growing consum er m arkets that m ay transform from
‘standards takers’ to ‘standards m akers’ assisted by European, U S and hom e-grown com panies.
The U nited States Trade Representative’s negotiations with China over IEEE802.11 m ay signal a
growing trend that European Com m ission officials m onitor and em ploy, over and above existing
bilateral U S-EU negotiations449. There is clear evidence that standards processes in these m arkets
contain far greater governm ent intervention than in Europe or N orth Am erica.
The m aintenance of open interoperable standards m ay ensure that the innovation and dynam ism
of the Internet continues. However, the geographical and sectoral fragm entation of the Internet,
driven by m obile network architectures as m uch as Ipv6 and actions taken unilaterally by
governm ents in East Asia and elsewhere, is rapidly proceeding. The future of Internet governance
has been shown in this chapter to depend m ore on these features of code than any U nited nations
inspired international governance forum . The future of the Internet is intim ately connected to
the trans-disciplinary analysis of standards setting.
448
Coen, David and Wyn Grant (2000) Corporate Political Strategy and Global Policy: A case study of
the Transatlantic Business Dialogue Working Paper 42, London Business School Regulation Initiative,
p5, at http://www.lbs.ac.uk/ri/paper42.pdf
449
See press coverage and Marsden interview with Adam Golodner, Cisco Systems, Information
Security Counsel, Washington 1 November 2005.
158
Chapter 8; Conclusion: Towards an Interdisciplinary Analysis of the Internet?
“The Internet is just beginning, its really only taking off now with broadband, which has
a few tens of m illions, one hundred or two hundred m illion at m ost, able to get it. In
twenty years’ tim e, everyone in the world is going to have broadband, and be able to
express them selves on it.”
Rupert Murdoch, Chair and CEO, N ews Corporation450
8.1. Regulating the Internet: Medium Law?

Can we now talk about m edium law? Is content delivery converging such that only one m edium ’s
regulation m atters? That m edium is the broadband Internet, which for consum ers has em erged in
the period since the ‘dotcom ’ bubble burst in March 2000. In advanced developed countries, the
m ajority of Internet users now use a connection of 512K b/s or higher. This m eans that the
Internet’s W W W interface, popularized in 1993 with the launch of the Mosaic browser, no
longer is confined to text and im ages. The growth of audio and video over this m edium m eans
that the Internet m ay no longer be treated as an extension of the print m edium that was its roots.
The consequence is that the non-regulation of the Internet m ay be replaced by som ething m uch
closer to a m erger of printed press and audiovisual regulation. W e m ay be seeing the
Com m issioner’s “regulation of the Internet”451.
Internet regulation has seen the increasingly determ ined application of offline m edium -specific
content and crim inal rules to the Internet, at the sam e tim e as the broadband Internet is
becom ing overwhelm ed by increasingly sophisticated attacks against increasingly powerful
broadband connections owned by increasingly ignorant end-consum ers; 300 m illion broadband
consum ers by end-2005. That m eans the corpus of rules applying to the audiovisual m edium and
to public safety increasingly being applied to the outm oded and threatened Internet. It was a fate
that the m uch younger and less powerful Internet escaped in 1996-7 when in both U nited States
450
Murdoch, Rupert (2006) Interview with Jeff Randall, BBC, at
http://news.bbc.co.uk/1/hi/help/3681938.stm
451
See Marsden (2000) (ed) Regulating the Global Information Society, Routledge, New York and
Goldsmith, Jack and Tim Wu (2006, forthcoming) Who Controls the Internet? Illusions of a Borderless
World, Oxford University Press.
159
and Europe regulators were persuaded not to intervene in the em ergent m edium 452 (a judgm ent
proved correct by the later adoption of broadband than previously predicted). In the Suprem e
Court’s quashing of the Com m unications D ecency Act in ACLU vs. Reno (1996) and the decision
of the European Parliam ent narrowly to avoid extension of the Television W ithout Frontiers
(TVW F) D irective of 1989 to the Internet in its 1997 revision453, the regulators gave the new
m edium a breathing space to self-regulate and otherwise dem onstrate its m aturity and disprove
the need for regulation454. That breathing space is now over.
In autum n 2005, the legislators signalled their intent to regulate the Internet via the Audiovisual
Content Services D irective (ACSD) com bined with the new Recom m endation on the Protection of
M inors455 (EU ) and the new proposed Com m unications Act456 (U S). I concentrate here on the
form er, as a European constitutional and com petition lawyer. I note in Chapter 4 that the
regulation of m ore oligopolistic industries than Internet content457 proved conducive to that self-
regulation that the Internet has not apparently achieved. Exam ples are the m obile Internet and
interactive gam ing environm ent458, to which one m ay be able to add the V ideo on Dem and
industry. However, for the 100 billion pages459 and perhaps 100 m illion sites on the Internet,
such regulation has proved im possible to coordinate. The European Com m ission is determ ined
to step into the gap as the Internet becom es m ore powerful and pervasive. It also intends to fulfil
its cultural and econom ic goals in ensuring that the unregulated Internet is not able without
im pedim ent to substitute in advertising revenues for the broadcast audiovisual environm ent. As
the Internet grows in power, it also grows predictably in scrutiny, as I point out in the concluding
chapter.
452 See: European Commission (1996) Green Paper on the protection of minors and human dignity in audiovisual and information services on 16 October
1996; Council resolution on illegal and harmful content on the Internet of 17 February 1997 OJ C 70, 6. 3. 1997; Economic and Social Committee Opinion OJ
C 214, 10. 7. 1998; European Parliament Opinion OJ C 339, 10. 11. 1997; Economic and Social Committee Opinion OJ C 287, 22. 9. 1997; Committee of the
Regions Opinion OJ C 215, 16. 7. 1997.
453 See Tongue, C. (1999) Chapter 5 in Marsden, C. and Verhulst, S. (eds) (1999) Convergence in European Digital TV Regulation.
454 Notably via the Recommendation on the Protection of Minors (1998) in the EU, and the PICS system of self-rating and filtering adopted by the W3C.
455 See COM(2004)0341 European Parliament legislative resolution on the proposal for a recommendation of the European Parliament and of the Council on
the protection of minors and human dignity and the right of reply in relation to the competitiveness of the European audiovisual and information services
industry– C6-0029/2004 – 2004/0117(COD) at
This will soon replace and update the Green Paper on the protection of minors and human dignity in audiovisual and information services, COM (96) 483,
16.10.97; Communication on Illegal and Harmful content on the Internet, COM(97) 487, 16.10.97; Council Recommendation 98/560/EC on the development
of the competitiveness of the European audiovisual and information services industry by promoting national frameworks aimed at achieving a comparable and
effective level of protection of minors and human dignity OJ L 270, 7.10.1998.
456 See draft at http://energycommerce.house.gov/108/News/09152005_staff_disc.pdf and commentary at
http://blog.pff.org/archives/2005/09/long_live_publi.html
457 Noam 2004
458 See Leonardi, Marsden and Tambini 2006 forthcoming.
459 Estimate by University of Southampton, though obviously the number is a moving target. An approximation is that, given the firewalls which cut off much
of the public Internet from search engines, there are perhaps 10 firewalled pages for every one that search engines can access.
160
Regulators do not universally share this view. At the Liverpool Audiovisual Conference held by
the Presidency of the Council of Ministers, the Chair of OfCom , U K regulator, stated460:
Ofcom ’s concern is that the practical benefits to producers of a harm onised regim e m ay
be finely balanced in com parison with the costs of additional regulation … as the
Com m ission itself has repeatedly acknowledged, m any, possibly all, of these benefits can
be achieved through the em powerm ent of consum ers to protect them selves through
m echanism s such as rating, filtering and parental controls, coupled with effective
industry initiatives to block access to som e form s of m aterial. W e need to avoid
regulatory double-banking.
So Ofcom ’s view, boiled down to its essential, is one of scepticism about the case m ade
for the extension of scope, in so far as it seeks to extend regulation to services currently in
their infancy, and concern about the practicalities involved.
Jam es Murdoch, Chairm an of Sky Television, the largest broadcaster in Europe, responded very
negatively, as had his father at the previous Audiovisual Assizes in Birm ingham in 1998461. Other
politicians in the room rem arked that, even for elected politicians rather than regulatory
specialists, the Com m issioner’s Cybercrim e Convention462 com m ents (therefore irrelevant for
econom ic legislation fora) were at least five years’ out of date for European rhetoric, and pre-
ACLU v. Reno for U S politicians. The European Com m issioner com m ented:
The debate gave rise to m any passionate statem ents, som etim es largely inspired by fantasy. I
have listened, always with interest, som etim es with am usem ent.
There is thus a pre-Millenial air of regulatory overbearance in European revision of Internet
regulation. The European Parliam ent passed a Resolution expressing that it:
22. Considers that the revision of the Directive should ensure the developm ent of new
technologies and new services, in order to secure the growth of the European econom y in
accordance with the Lisbon strategy;
24. Fears that, on such an im portant subject, the debate and consultation will give
preference to econom ic considerations and inter-governm ental relations; is aware that the
m arket alone will not resolve the problem s and that the institutions m ust respond to the
concerns of Europeans about the cultural content of television;
460 Lord Currie, Speech 21 September 2005, available at: http://www.ofcom.org.uk/media/speeches/2005/09/liverpool_conf

461 http://www.advanced-television.com/2005/news_archive_2005/Sep19_Sept23.htm
462 Convention on Cybercrime CETS No.185 entered into force on 1 July 2004, see
http://conventions.coe.int/Treaty/Commun/QueVoulezVous.asp?NT=185&CM=1&DF=&CL=ENG
161
26. Is concerned about the pressure to reduce regulation in this sector and recalls that the
Directive establishes m inim um standards which have not succeeded in preventing a
deterioration in the quality of program m es.463
This regulatory approach is however, firm ly rooted in co-regulation, in new form s of self-
regulation underpinned by legislative foundation and the possibility of intervention by the
regulator464. Co-regulation is therefore the approach proposed by the ACSD, in sim ilar term s to
the extrem ely detailed piece of ‘soft law’ that is the 2005 updated Recom m endation on the
Protection of Minors. This piece of ‘political signalling’ (a Recom m endation has no binding
force) is intended to address the online industry by reference to its self-regulatory m echanism s.
There is also the Electronic Com m erce D irective (ECD) 2000/31/EC to consider. Its standing in
Com m unity law will be decided in the next twelve m onths, as it is due to be reviewed in late
2006, and no doubt the greatly expanded jurisdiction of the ACSD will affect greatly the
application of the ECD. Currently, it rem ains the legal instrum ent m ost suited to the online
environm ent. Indeed, m uch of the ACSD is intended to harm onize broadcasting law online by
reference to the legal definitions used in the ECD. The inter-service negotiation between the
arbiters of the ECD and those of the ACSD will be an extrem ely im portant elem ent in the
internal European Com m ission discussion of the ACSD, prior to its introduction to the Council
of Ministers (and then European Parliam ent) in Decem ber 2005.
The ECD leaves m uch detailed regulation to the m arket actors them selves, in m uch the sam e way
as the ACSD in draft. It is therefore in assessing regulation of the Internet – and m edium law as I
have characterized it – that we look m ore closely at this novel form of ‘code’ which joins software
code and legal code in the policing of the online world. W e will see all the pitfalls as well as
advantages of self-regulation that are highlighted by Pitofsky:
From a public policy perspective, self-regulation can offer several advantages over
governm ent regulation or legislation. It often is m ore prom pt, flexible, and effective than
governm ent regulation. Self-regulation can bring the accum ulated judgm ent and
experience of an industry to bear on issues that are som etim es difficult for the
463
P6_TA-PROV(2005)0322 European Parliament resolution on the application of Articles 4 and 5 of
Directive 89/552/EEC ("Television without Frontiers"), as amended by Directive 97/36/EC, for the
period 2001-2002 (2004/2236(INI)) at
464 Examples include broadcast advertising, where a Charter – a Code of Conduct - exists for the European Association of Advertising, and computer games,
where a ratings system – PEGI - has been in place for 2 years. The latter is fascinating in its use of comparative regulatory techniques, using rating by NICAM
in Holland, the pan-sectoral sef-regulatory system, and dispute resolution using UK adjudication, judged as the most time- and cost-efficient as well as
procedurally sound.
162
governm ent to define with bright line rules. Finally, governm ent resources are lim ited
and unlikely to grow in the future. Thus, m any governm ent agencies, like the FTC, have
sought to leverage their lim ited resources by prom oting and encouraging self-regulation.
Of course, self-regulation can be anticom petitive. Com petitors m ay use the self-
regulatory process to disadvantage new rivals or new form s of com petition.465
I suggest in this concluding Chapter that this is exactly the intention of the actors lobbying the
European Com m ission over the new ACSD, even when the official pronouncem ents of the actors
are in favour of self-regulation. The Com m ission’s relative lack of interest in ensuring continued
low barriers in this dom ain in favour of well-resourced and socially responsive actors is an act of
suprem e political courage. Raising the drawbridge on the European broadband Internet to new
entry is surely one of the m ost profound changes in European public policy towards the
Inform ation Society.
I do not attem pt an analysis in detail of the definitions of content in European law, which I and
others have covered elsewhere466, but clearly I acknowledge that the ‘political signalling’ of a
Directive which regulates Internet content does not signify any real com pliance or enforcem ent
m otivation on the part of m ost Mem ber States – m ost especially the U K . N evertheless the
political winds are blowing towards a gesture that will regulate Internet video without regard for
its definition. As the com petitive telecom s operators group states, there are m ultiple form s of
m ovem ent possible on websites: re-transm ission of traditional broadcasting; video-on-dem and
services; interactive com puter gam ing; anim ated G IFs; online anim ation (e.g. Flash); dynam ic
HTML to switch between im ages. They state:
There is a serious risk that, whatever boundaries the focus groups, or indeed the
Com m ission, envisage, the reality will be considerable legal uncertainty and huge
differences in the scope of regulation at national level.467
U nited States observers have by now realised that what they are observing is a broadband video
Yahoo!v. France. That is correct and the tide is com ing in.
Consider the threats listed by Lem ley and Lessig of increasing entry barriers:
55. The first is the cost of losing ISP com petition. As we have argued, one should
not think of ISPs as providing a fixed and im m utable set of services. Right now ISPs
465 ‘Self Regulation And Antitrust’, Prepared Remarks of Robert Pitofsky Chairman, Federal Trade Commission, D. C. Bar Association Symposium,
February 18, 1998, Washington, D.C. at http://www.ftc.gov/speeches/pitofsky/N_2_#N_2_
466 See Woods, L. and Scheuer, A. (2004) Advertising Frequency and the Television Without Frontiers Directive, 29(3) European Law Review at
366-384,.
467 ECTA Comments on the DG Information Society’s Issues Papers for the Liverpool Audiovisual Conference at
http://www.europa.eu.int/comm/avpolicy/revision-tvwf2005/docs/ip1-3-4-5-ecta.pdf
163
typically provide custom er support, as well as an IP address that channels the custom er’s
data. Com petition am ong ISPs focuses on access speed, as well as som e com petition for
content...
56. In the future, however, ISPs are potential vertical com petitors to access providers
who could provide com petitive packages of content, or differently optim ized caching
servers, or different m ixes of custom er support, or advanced Internet services. This ISP
com petition would provide a constant pressure on access providers to optim ize access.
57. The benefits from this com petition in the history of the Internet so far should
not be underestim ated. The ISP m arket is extraordinarily com petitive. This com petition
has driven providers to expand capacity and lower prices. It has also driven providers to
give highly effective custom er support. This extraordinary build-out of capacity has not
been incented through the prom ise of m onopoly protection. The com petitive m arket has
provided a sufficient incentive, and the m arket has responded.
58. The second cost is the risk that legacy business m odels will im properly affect the
architecture of the net. Broadband is a potential com petitor to traditional cable video
services. Traditional cable providers m ight well view this com petition as a long term
threat to its business m odel, and they m ay not want to change to face that com petitive
threat. By gaining control over the network architecture, however, cable providers are in
a position to affect the developm ent of the architecture so as to m inim ize the threat of
broadband to their own video m arket. For exam ple, a broadband cable provider that has
control over the ISPs its custom ers use m ight be expected to restrict custom ers’ access to
stream ing video from com petitive content sources, in order to preserve its m arket of
traditional cable video.
59. The third cost of such control by a strategic actor is the threat to innovation.
Innovators are less likely to invest in a m arket where a powerful actor has the power to
behave strategically against it. Innovation in stream ing technologies, for exam ple, is less
likely when a strategic actor can affect the selection of stream ing technologies, against
new, and com petitive system s.
To which we can add the regulatory costs instituted by a paternalistic industrial policy applied to
‘level playing fields’ when what is actually being provided is a new playing field. Broadcast
regulators on the Internet are like elephants playing ice hockey.
164
It is the perfect storm of anti-com petitive pressures which now threatens European Internet
innovation.
• First, the shattering of the com petitive ISP m arket by the com bination of the financial
crisis of 2000-3 and the deregulation of incum bents (m ore especially in the U S);
• then the m assive pressures to regulate content m ore tightly in the wake of the security
threats of hackers, spam m ers and terrorists in 2001-4;
• now the broadcasters’ triple wham m y in the guise of ‘levelling the playing field’.
W e are entering the era of broadband incum bents dom inating. They are in m obile, in fixed
broadband, and in video-over-IP.
8.2. ISP Actors: Self-

Self-Regulatory Resources and Internet Policy
The broadband Internet is a m ore powerful and m ore pervasive m edium than its narrowband
predecessor. It offers new ‘regulability’ in the form of the BSPs, m ore profitable and well-
resourced actors than their narrowband forebears. It also requires well-funded BSPs, given the
range of access and content control costs, notably increased by peer-to-peer file sharing and
copyright enforcem ent.
As ISP incom es from narrowband Internet access, which used the national public telecom s
network as ‘free riders’, were replaced by less favourable broadband access charges, ISPs
increasingly find their financial resources for self-regulation eroded. The ‘triple wham m y’ of
broadband access charges, advertising revenue falls, and the absence of profitable content services
on either narrow- or broadband, m eant that the period 2001-3 was one of enorm ous econom ic
failure for ISPs.
Table: Incum bent Share of European Broadband Retail Markets Decem ber 2002468
468 Source: IDATE Digiworld (2003) The World Internet Access Market, at Figure 41.
165
In the years 2001-3, users of the Internet reached a m ajority of citizens in m ost OECD countries.
Broadband take-up appears to be the fastest growing com m unications service ever adopted,
according to OECD analysis for June 2003469.
Table: OECD Broadband, Mobile, N arrow-band, ISDN Subscriber G rowth
Though broadband via wired telecom (DSL or cable m odem ) service is not available to all,
coverage is growing, reducing the Digital Divide in access.
These large or even dom inant, well-resourced and carrier-backed BSPs are in a position to
provide the effective self-regulation that narrowband ISPs are increasingly unable to effectively
resource470. BSPs face m ajor legal challenges. The copyright bottleneck caused by record/m ovie
com panies and their contractors (typically collecting societies for publishers/authors,
m usicians/actors) failure to release m aterial online has been overcom e by illegal file-sharing
between users: peer-to-peer networks. W hereas governm ents, conscious of exposing the
convenient fiction of ISP liability online, have not prosecuted breaches, copyright owners are
pursuing BSP custom er records vigorously, the Recording Industry Association of Am erica
leading471. In January and February 2004, the RIAA brought actions against over 1000 file-
sharers known only by their IP addresses, in ‘John Doe’ litigation472. The broad issues of ISP
469 See OECD ICCP Broadband Update October 2003, accessed 28 February 2004 at http://www.oecd.org/dataoecd/18/9/18464850.pdf
470 Table from Executive Summary, ITU (2003) Birth of Broadband, accessed 28 February 2004 at
http://www.itu.int/osg/spu/publications/sales/birthofbroadband/BoBexecsumm.pdf Note independent cable operators, retail broadband ISPs as well as third
party broadband providers erode market share: thus KPN in Netherlands has strong retail competition on its own network, some local loop unbundling
competition and strong cable network competition. In Germany, competition is far more limited.
471 See http://www.riaa.com/news/newsletter/021304.asp accessed 28 February for the broad alliance International Intellectual Property Alliance.
472 See http://www.riaa.com/news/newsletter/pdf/sampleJohnDoeLawsuit.pdf for an example of the RIAA proceedings.
166
response to copyright infringem ent is considered in other works, notably ‘Losing Liberty’ and the
chillingeffects.org website and program m e473.
Econom ies of scale in Internet industries do not only apply to provision of services: they also
apply to self-regulatory capacity. A com pany with operations in six sectors is able to use lawyers
and policy-m akers across the different sectors because the basic skills needed to identify copyright
and privacy policies, interconnection and com petition law issues, child protection and content
filtering, are closely related. Further a com pany with international subsidiaries can not only
deploy its browser software across international m arkets but also its privacy policy. Policies and
regulatory strategies are inform ation goods just like browser software: once created (at high fixed
cost) they can be redeployed according to national m arket characteristics (at low variable cost).
Com panies such as AOL, Microsoft, W orldCom and Yahoo! have a decade of experience in
Internet regulatory policy, which they have redeployed in several dozen territories internationally.
It is not unreasonable in the absence of strong governm ent rejection of U .S. self-regulatory policy
towards the Internet to expect that the policies espoused by these ‘first m overs’ in Internet
regulation to achieve a regulatory convergence between the various m arkets in which they
operate: a convergence of Internet policy to accom pany m arket convergence.
Regulatory com pliance can form part of a m arketing strategy, in content as in carriage. Q uality of
service is defined in ISP service largely by reliable speed of connection, but com panies such as
AOL, Yahoo! and Microsoft N etwork also m arket their superior spam filtering and parental
control filters. W here an industry self-regulates, it can set standards at levels that m ake
com pliance relatively low or high cost – depending on the participants’ views of best com m ercial
outcom e. In a rapidly growing m arket with little or no governm ent social regulatory intervention,
low costs m ight guarantee expansion, but in a m ature m arket with active social regulation, higher
costs m ight protect actors from new com petition. Regulatory initiatives such as Rightswatch 474,
the copyright regulatory body, are expensive and voluntary participation by sm aller ISPs was
prohibitively expensive. In a m ore consolidated m arket with higher com pliance and other
473
Ahlert, Marsden and Yung 2004
474
Stated to be “a research project aimed at developing consensus and promoting awareness of self-
regulatory notice and takedown (NTD) procedures for Europe, as a tool to achieve prompt removal of
copyright-infringing material from the Internet. The impetus behind the project specification was
provided within the text of the EU’s E-Commerce Directive which encourages a self-regulatory
approach through voluntary agreements for ‘development of rapid and reliable procedures for
removing and disabling access to illegal information.’ Activities and research taking place throughout
the two year duration of the project (2000 – 2002) have been funded under the European Commission' s
Information Society Technology ("IST") programme.” See http://www.rightswatch.com/
167
regulatory costs, ISPs can free m ore regulatory resource because they are guaranteed that other
ISPs also take their regulatory responsibilities seriously.
8.3. Medium Law and Its Regulatory Cost

Cost-
ost-Benefits
W e saw in Chapter 6 how m obile Internet access providers have elim inated m any of the end-to-
end options in Internet design by closing off a walled garden for public safety and their private
profit. In the discussions in the new ACSD, the large well-resourced telecom s com panies saw the
opportunity to m ake com m on cause with m obile operators, public service broadcasters and
com m ercial television com panies in an unholy alliance to prevent the open Internet video m odel
em erging. The regulation of the Internet that is rapidly taking place is being driven –
unquestionably – in Europe by politicians for public safety reasons. They are erecting entry
barriers with the connivance of the incum bent players, with potentially enorm ous consequences
for free speech, for free com petition and for individual expression. This m ay be the correct policy
option: claim s by the European Com m issioner tat regulating the Internet is not the intention do
not flatter the intelligence of the audience. That m ay be the intention of the paternalistic and
oligopolistic interests represented in the new regulated Internet.
Regulators should beware killing the goose that laid the golden egg. Academ ics and other policy
entrepreneurs can help to steer the debate by dem onstrating the com plex trans-disciplinary nature
of Internet regulatory analysis. I hope in this book to have contributed to better understanding
the com plexity of the questions, and the subtlety required in attem pting to provide fixed answers
at any point in tim e.
168
References
Abram son, Bruce (2005) D igitalPhoenix:W hy the Inform ation Econom y Collapsed and H ow It W illRise
Again, MIT Press
A CLU v. Reno Suprem e Court Case N o. 96-511, 1997
Ahlert, Alexander and Tam bini(2003) European 3G M obile Industry Self-Regulation: IAPCODE
Background Paper for W orld Telem edia Conference at 2: see
http://www.selfregulation.info/iapcoda/031106-m obiles-revised-bckgrd.pdf
Ahlert, Marsden and N ash (2005) Protecting Minors from Exposure to Harm ful Content on Mobile
Phones, for European Internet Coregulation N etwork, at
http://network.forum internet.org/article.php3?id_article=24
Ahlert, Marsden and Yung (2004) H ow ‘Liberty’D isappeared from Cyberspace:The M ystery Shopper Tests
Internet ContentSelf-Regulation at http://pcm lp.socleg.ox.ac.uk/text/liberty.pdf
Am erican Psychological Association (2005) Resolution on Violence in Video G am es and Interactive M edia at
http://www.apa.org/releases/resolutiononvideoviolence.pdf released August 17 at
http://www.apa.org/releases/videoviolence05.htm l
Andersen, J. (2005) Regulatory Challenges in a Converging and LiberalTelecom s W orld, at

http://www.ero.dk/42ACC9DD-B775-4948-8B63-CA7EA18B0329?fram es=no&
AOL Tim e W arner (2005) Press Release at 28 Decem ber

http://m edia.tim ewarner.com /m edia/newm edia/cb_press_view.cfm ?release_num =55254499
Arrow. K . (1995) Barriers to ConflictResolution.
Auerbach, K . (2005) http://www.cavebear.com /cbblog-archives/000188.htm l
Australia Post (2005) A nnualReport2004 at http://www.auspost.com .au/annualreport2004/lett_direct.asp
Ayres, Ian and John Braithwaite (1992) Responsive Regulation:Transcending the D eregulation D ebate Oxford
U niversity Press
Baird Z oe (2002) G overning the Internet:Engaging G overnm ent,Business,and N onprofits, Foreign Affairs,
N ovem ber/Decem ber 2002
Baldwin, R. and C. McCrudden (1987). Regulation and public law W eidenfeld and N icholson
Baldwin, R. et al (1995) Regulation in Q uestion: The G row ing Agenda Merck, Sharp and Dohm e
Baldwin, R. et al (1998) Socio-LegalReader on Regulation, Oxford U niversity Press
Balkovich, E., Baer, W alter S., V ollard, B. (2003) ‘Research Topics for Inform ing Broadband Internet Policy’
for Office of Science and Technology, Rand Corporation Science and Technology Issue Paper at
169
http://www.google.co.uk/search?sourceid=navclient& ie=U TF-8& rls=G G LD,G G LD:2003-

48,G G LD:en& q=rand+broadband+policy RAN D Corporation, Santa Monica, Cal.
Ball, K . and W ebster, F. (2003) The Intensification ofSurveillance:Crim e,Terrorism and W arfare in the
Inform ation Age, Pluto, London.
Bar, Francois, Cohen S., Cowhey P., De Long B., K leem an M., Z ysm an J. (2000) Access and Innovation
Policy for the Third-G eneration Internet, 24 Telecom s Policy at 489-518
Baran. P. (1962) O n D istributed Com m unications N etw orks RAN D P-2626 RAN D Corporation, Santa
Monica, Cal.
Barlow John Perry (1996) D eclaration ofthe Independence ofCyberspace,

http://www.eff.org/~barlow/Declaration-Final.htm l
Barnes, Fod (2000) ‘Com m entary:w hen to regulate in the G IS? A public policy perspective’, Chapter 7 in
Marsden, C. (ed) Regulating the G lobalInform ation Society, Routledge
Barroso, Jose Manuel (2005) Case IE/2004/0121:A ccess and callorigination on public m obile telephone
netw orks in Ireland:Com m ents pursuantto Article 7(3) ofD irective 2002/21/EC Letter of 20 January to
Com Reg
Barry M. Leiner, V inton G . Cerf, David D. Clark, Robert E. K ahn, Leonard K leinrock, Daniel C. Lynch,
Jon Postel, Larry G . Roberts, Stephen W olff (undated, current) A BriefH istory ofthe Internet
www.isoc.org/internet/history/brief.htm l
Barton, John H. (1997) The Balance Betw een IntellectualProperty Rights and Com petition:Paradigm s in the
Inform ation Sector, European Com petition Law Review Issue 7 at 440-445
Bauer Johannes M., Junghyun K im and Steven S. W ildm an (2005) A n Integrated Fram ew ork For Assessing
Broadband Policy O ptions Michigan State Law Review 21
Beaufort International (2003) Prem ium SM S Services Research, at p5, at

http://www.icstis.org.uk/icstis2002/pdf/SMS_RESEARCH_REPORT_MAY03.PDF
Bell, D. (1977) The Future W orld D isorders
Bell, D. (1999) The W inding Passage
Bell, Daniel (1973) The Com ing ofPost-IndustrialSociety, revised paperback edition 1999
Benkler, Yochai(1998) Com m unications infrastructure regulation and the distribution ofcontrolover content
Telecom m unications Policy, vol.22, no. 3, pp. 183-196, at: http://www.benkler.org/PolTech.pdf
Benkler, Yochai(1998) O vercom ing agoraphobia:building the com m ons ofthe digitally netw orked
environm ent Harvard Journal of Law and Technology, vol. 11, pp. 287-400, at:
http://www.law.nyu.edu/benklery/agoraphobia.pdf
170
Bentham , J. (1995) The Panopticon W ritings
Besen, S.M. and Saloner, G . (1989) The Econom ics ofTelecom m unications Standards Setting, pp177-220 in
Crandall, Robert W . and Flam m , K enneth (eds). Changing the Rules:technologicalchange,international
com petition,and regulation in com m unications, Brookings Institute, W ashington D.C.
Bick, Jonathan D. (1998) W hy Should The Internet Be A ny D ifferent?,19 PACE L. REV . 41, 63
Bishop, K ay and Mayer (1995) The Regulatory Challenge Oxford U niversity Press
Blackm an, C., M. Cave, P.A David (1996) The new internationaltelecom m unications environm ent.
Com petition,regulation,trade and standards 20 Telecom m unications Policy 10 at 721 – 724.
Blackm an, Colin (1998) ‘Convergence betw een telecom m unications and other m edia. H ow should regulation
adapt?’, Telecom m unications Policy, vol. 22, no. 3, pp. 163-170
Blom -Hansen, J. (1997) A ‘N ew Institutionalist’Perspective on Policy N etw orks, 75 Public Adm inistration
669-693, at 674-5
Blum enthal, M. and Clark, D. (2001) http://itc.m it.edu/itel/docs/jun00/TPRC-Clark-Blum enthal.pdf
Boddewyn J. J. (1988) A dvertising Self-regulation and O utside Participation.
Bohlin Erik et al (eds.)(2000) Convergence in Com m unications and Beyond, Elsevier Science
Bollinger, Lee C. (1976) Freedom ofthe Press and Public Access:Tow ard a Theory ofPartialRegulation 75
Michigan Law Review 1
Borrus, Bar and Steinberg (1996) Islands in the Bitstream :Charting the N II Interoperability D ebate, BRIE
W orking Paper N o.79 at http://www.stanford.edu/¬fbar/islands.pdf
Borrus, M. and Z ysm an, J. (1997) G lobalisation w ith Borders, 4 Industry and Innovation 2.
Braithwaite, J. and Drahos, P. (2000) G lobalBusiness Regulation, Cam bridge U niversity Press
Brennan T. (2003) The FCC and policy federalism :broadband Internet access regulation pp173-199 in G .
Madden (ed.) InternationalH andbook ofTelecom m unications Econom ics V . III, Edward Elgar
British Institute of International and Com parative Law (2004) Effective A ccess and Procedure in
Telecom m unications D isputes in Europe BIICL, London at
http://www.biicl.org/adm in/files/BIICLTelecom s20Report202004.pdf
Broadband Trends (2005) http://broadbandtrends.com /Fast% 20Facts.htm
Brown Allan and Picard, Robert eds. (2004) D igitalTerrestrialTelevision in Europe Lawrence Erlbaum ,
Mahwah N ew Jersey
Brunel for Departm ent of Trade and Industry (2002) Investigating Broadband Technology D eploym ent in
South K orea, Brunel U niversity U xbridge, at http://www.broadbanduk.org/reports/SK orea_report.pdf
171
Brunner and Bewer (1972) Policy and the Study ofthe Future:G iven Com plexity,Trends or Processes? P-4912
http://www.rand.org/publications/bib/SB1019.pdf RAN D Corporation, Santa Monica, Cal.
Buiges, P. and P. Rey eds. (2004) The Econom ics ofA ntitrust and Regulation in Telecom m unications, Edward
Elgar.
Burk, Dan L. (1999) VirtualExit in the G lobalInform ation Econom y, 73 Chicago K ent Law Review 4, at
943-995.
Burstein, Dave (2006) D SL Prim e N ew sletter, January
Cadm an, R. and Dineen, C. (2004) Broadband Access M arkets in Europe:W hy Regulation should Prom ote
Com petition at http://www.spcnetwork.co.uk/uploads/20040315_investm ent_paper.pdf
Calvert, Justin (2002) N ew age rating system for gam es in Europe, ‘G am espot’ 25 October at
http://news.zdnet.co.uk/hardware/em ergingtech/0,39020357,2124493,00.htm
Cam p, L. and V incent (2004) Setting Standards:Looking to the Internet for M odels ofG overnance, at
http://papers.ssrn.com /sol3/papers.cfm ?abstract_id=615201
Cargill, Carl and Martin W eiss (1992) 'Consortia in the Standards D evelopm entProcess'Journal of the
Am erican Society for Inform ation Science, vol. 43 no.8 (Septem ber 1992), pp.559
Carnoy, Martin, Manuel Castells, Stephen S. Cohen and Fernando Henrique Cardoso (1993) The N ew G lobal
Econom y in the Inform ation Age;Reflections on O ur Changing W orld Macm illan
Case C-245/01 23 RTL Television G m bH v. N iedersächsische Landesm edienanstalt für privaten Rundfunk
Case IE/2004/0121: A ccess and callorigination on public m obile telephone netw orks in Ireland
Cave M. (2004) Econom ic A spects ofthe N ew Regulatory Regim e for Electronic Com m unication Services in P.
Buiges and P. Rey (eds) The Econom ics ofA ntitrust and Regulation in Telecom m unications, E. Elgar, 27-41
Cave, M. (1997) Regulating digitaltelevision in a convergent w orld, Telecom m unications Policy 21 (7)
Special Issue: The Econom ic Regulation ofPay Broadcasting at 575 – 596, at 593.
Cave, M. and Cowie, C. (1996) Regulating ConditionalAccess in European Pay Broadcasting Com m unications
and Strategies, N o.23, 3rd quarter at 119, IDATE: Montpelier
Cave, M. and Cowie, C. (1998) N otO nly ConditionalA ccess. Tow ards a Better Regulatory A pproach to
D igitalTV, Com m unications et Strategies, N o 30, 2nd Q uarter, 1998 at pp 77-101, IDATE: Montpelier
Cave, Martin (2005) Com petition and the exercise ofm arket pow er in broadcasting:a review ofrecentU K
experience 7 info 5 at pp. 20-28
Cawson, A. (1994) Business interests and the regulation ofnew technology in Europe:the case oftelevision, 4
Business and the Contem porary W orld at 73
172
CEC COM (2002) 695 Eighth report on the im plem entation of the telecom m unications regulatory
package, Brussels, Decem ber, at:
http://europa.eu.int/inform ation_society/topics/telecom s/im plem entation/annual_report/8threport/finalre
port/annex2.pdf
CEC COM (2003) 776 http://europa.eu.int/com m /avpolicy/legis/reports/com 2003_776final_en.pdf
CEC COM (2004) 341 European Parliam ent legislative resolution on the proposal for a recom m endation
of the European Parliam ent and of the Council on the protection of m inors and hum an dignity and the
right of reply in relation to the com petitiveness of the European audiovisual and inform ation services
industry– C6-0029/2004 – 2004/0117(COD) at http://www.europarl.eu.int/om k/sipade3?PU BREF=-
//EP//TEXT+TA+P6-TA-2005-
0330+0+DOC+XML+V 0//EN & L=EN & LEV EL=0& N AV =S& LSTDOC=Y& LSTDOC=N
CEC COM (2004) 674 on the role of European standardisation in the fram ework of European policies
and legislation
CEC COM (96) 359 Standardisation and the G lobal Inform ation Society: the European Approach
CEC COM (96) 483 Draft Report on the Com m ission G reen Paper on The Protection of Minors and
Hum an Dignity in Audiovisual and Inform ation Services - C4-0621/96) PE 221.804 of 24 April 1997
CEC COM (97) 487 Com m unication on Illegal and Harm ful content on the Internet 16.10.97; .
CEC COM (99) 02M anagem ent ofthe Internet
CEC Decision 676/2002/EC of the European Parliam ent and of the Council of 7 March 2002 on a
regulatory fram ework for radio spectrum policy in the European Com m unity (“Radio Spectrum
Decision”); OJ L 108,
http://europa.eu.int/inform ation_society/topics/telecom s/regulatory/m aindocs/index_en.htm
CEC Decision 87/95/EEC defining the overall objective of end-to-end data and system s interoperability
CEC Decision N o 1151/2003/EC of the European Parliam ent and of the Council of 16 June 2003 OJ L
162, 1.7.2003
CEC Directive 2002/19/EC of the European Parliam ent and of the Council on access to, and
interconnection of, electronic com m unications networks and associated facilities (“Access Directive”), OJ L
108, at: http://europa.eu.int/inform ation_society/topics/telecom s/regulatory/m aindocs/index_en.htm
CEC Directive 2002/20/EC of the European Parliam ent and of the Council on the authorisation of
electronic com m unications networks and services, (“Authorisation Directive”), OJ L 108, at:
173
CEC Directive 2002/21/EC of the European Parliam ent and of the Council on a com m on regulatory
fram ework for electronic com m unications networks and services, (“Fram ework Directive”), OJ L 108, at:
CEC Directive 2002/22/EC of 7 March 2002 of the European Parliam ent and of the Council on universal
service and users’ rights relating to electronic com m unications networks and services (“U niversal Service
Directive”), OJ L 108,
at:http://europa.eu.int/inform ation_society/topics/telecom s/regulatory/m aindocs/index_en.htm
CEC Directive 2002/58/EC of the European Parliam ent and of the Council of 12 July 2002 concerning
the processing of personal data and the protection of privacy in the electronic com m unications sector
(“Directive on privacy and electronic com m unications”), OJ L 108, at:
CEC Directive 83/189/EEC on general standardisation principles
CEC Directive 92/59/EEC on G eneral Product Safety
CEC Directive 95/46/EC on Data Protection
CEC Directive 97/7 on Distance Selling
CEC Directive 98/34/EC on European standardisation policy, including the ICT dom ain
CEC Recom m endation 98/257/EC on out-of-court settlem ents
CEC Recom m endation 98/560/EC on the developm ent of the com petitiveness of the European
audiovisual and inform ation services industry by prom oting national fram eworks aim ed at achieving a
com parable and effective level of protection of m inors and hum an dignity OJ L 270, 7.10.1998.
CEC Resolution on illegal and harm ful content on the Internet of 17 February 1997 OJ C 70, 6. 3. 1997;
CEC SEC (2004) 1251 on the challenges for European standardisation
Cherry, B. and W ildm an, S. (2000) Preventing Flaw ed Com m unications Policies by A ddressing Constitutional
Principles, 2000 L.REV . M.S.U .-D.C.L. 1 pp55-107, at 96-99,
Cleevely David (2000) Standards:Laying the Foundations, London: U K Departm ent of Culture Media and
Sport
Coase, R. (1998) Com m enton Thom as W . H azlett, J. Law and Econom ics XLI pp 577-580 at 580
Coase, R.H. (1937) The N ature ofthe Firm Econom ica 4 386-405
Cockborne, Jean-Eric de, Bernard Clem ents and Adam W atson Brown (1999) EU policy on Multim edia
regulation, paper presented on 14 June 1999 at the Montreux Sym posium '99 at
http://www.ispo.cec.be/infosoc/telecom policy/en/m ontreux.htm l
174
Coen, David and W yn G rant (2000) Corporate Political Strategy and G lobal Policy: A case study of the
Transatlantic Business Dialogue W orking Paper 42, London Business School Regulation Initiative, p5, at
http://www.lbs.ac.uk/ri/paper42.pdf
Com petition Com m ission (2002) V odafone, O2, Orange and T-Mobile: Reports on references under
section 13 of the Telecom m unications Act 1984 on the charges m ade by V odafone, O2, Orange and T-
Mobile for term inating calls from fixed and m obile networks; V olum e 2: Chapters 3 to 15 report at
http://www.ofcom .org.uk/static/archive/oftel/publications/m obile/ctm _2003/ctm 2.pdf
Convention for the Protection of Hum an Rights and Fundam ental Freedom s CETS N o.: 005 entered into
force 3 Septem ber 1953, at
http://conventions.coe.int/Treaty/Com m un/Q ueV oulezV ous.asp?N T=005& CM=8& DF=04/10/2005& C
L=EN G
Council of Europe Convention on Cybercrim e, CETS N o.185, opened for signature Budapest 23
N ovem ber 2001, entered into force on 1 July 2004 see http://press.coe.int/cp/2004/135a(2004).htm
Cowie, C and Marsden, C. (1999) Convergence: N avigating Through Digital Pay-TV Bottlenecks 1 Info 1
at 53-66
Crandall, Robert W . (1989) The Role ofthe U S O perating Com panies, at p115 in Crandall and Flam m
(1989) eds.
Crandall, Robert W . and Flam m , K enneth eds (1989) Changing the Rules:technologicalchange,
internationalcom petition,and regulation in com m unications, Brookings Institute, W ashington D.C.
Crandall, Robert W . and H.J. Singer (2005) Are Vertically Integrated D SL Providers Squeezing U naffiliated
ISPs (and Should W e Care)? at http://ssrn.com /abstract=710601
Crowcroft, J. and Philips I. (2001) TCP/IP & Linux ProtocolIm plem entation:System s Code for the Linux
Internet, W iley & Co.
Croxford, I. and Marsden, C. (2001) I W antm y W iFi! The O pportunity for Public A ccess W ireless LocalArea
N etw orks in Europe, from Re: Think!, London, at www.croxford.org
Cukier K enn (2004) M ultilateralControlofInternetInfrastructure and its Im pact on U .S. Sovereignty 5

October www.tprc.org.
Currie, Lord, of Marylebone (2005) Speech 21 Septem ber 2005, available at:
http://www.ofcom .org.uk/m edia/speeches/2005/09/liverpool_conf
Dalkey (1967) Delphi, RAN D P-3704 RAN D Corporation, Santa Monica, Cal.
Dalkey (1969) Predicting the Future, P-3948 RAN D Corporation, Santa Monica, Cal.
175
David, P. and Shurm er, M. (1996) Form alstandards setting for globalcom m unications and inform ation
services:tow ards an institutionalregim e transform ation? 20 Telecom m unications Policy 10 at 789-816
De Jouvenel B. (1965) Futuribles, January P-3045 RAN D Corporation, Santa Monica, Cal.
Deakin, S. (1999) Law versus econom ics? Reflections on the N orm ative Foundations ofEconom ic Activity,
pp30-49 in Richardson, M. and Hadfield, G . (1999) The Second W ave ofLaw and Econom ics, Federation
Press, Sydney
Decision N o 1151/2003/EC of the European Parliam ent and of the Council of 16 June 2003 OJ L 162,
1.7.2003 am ending Decision N o 276/1999/EC of the European Parliam ent and of the Council of 25
January 1999 adopting a Multiannual Com m unity Action Plan on prom oting safer use of the Internet and
new online technologies by com bating illegal and harm ful content prim arily in the area of the protection of
children and m inors OJ L 33, 6.2.1999, p.1
Decision N o 276/1999/EC of the European Parliam ent and of the Council of 25 January 1999 adopting a
Multiannual Com m unity Action Plan on prom oting safer use of the Internet and new online technologies
by com bating illegal and harm ful content prim arily in the area of the protection of children and m inors OJ
L 33, 6.2.1999, p.1 as am ended by Decision N o 1151/2003/EC of the European Parliam ent and of the
Council of 16 June 2003 OJ L 162, 1.7.2003
Devetak, Richard (2002) H yperpow er Am erica rew rites the globalrules, at

http://www.theage.com .au/articles/2002/12/08/1038950270370.htm l
Dorem us, P. N ., W illiam W . K eller, Louis W . Pauly, and Sim on Reich (1999) The M yth O fThe G lobal
Corporation Princeton U niversity Press
Drake, W illiam J. (2000) ‘The rise and decline of the international telecom m unications regim e’, Chapter 4
in Marsden C. (ed) Regulating the G lobalInform ation Society, Routledge
Dror (1968) Futures in G overnm ent, RAN D P-3909, RAN D Corporation, Santa Monica, Cal.
Dutton W .H. ed (1997) Inform ation and Com m unication Technologies: V isions and Realities, Oxford
U niversity Press
Dutton W .H. ed (1998) Society on the Line: The Inform ation and Com m unication Revolution, Oxford
U niversity Press
ECTA (2005) Com m ents on the DG Inform ation Society’s Issues Papers for the Liverpool Audiovisual
Conference at http://www.europa.eu.int/com m /avpolicy/revision-tvwf2005/docs/ip1-3-4-5-ecta.pdf
ECTA (2005) http://www.ectaportal.com /uploads/BroadbandScorecardQ 42004.pdf
Edwards, L. (ed) (2006) The N ew LegalFram ew ork for E-Com m erce in Europe, Hart Publishing, Oxford, at
http://www.hart.oxi.net/pdf/1841134511.pdf'
Einhorn, Michael (2004) M edia,Technology,and Copyright:Integrating Law and Econom ics, Edward Elgar
176
Elkin-K oren & N .W . N etanel eds. (1999)The Com m odification ofInform ation, Inform ation Law Series,
V ol. 11, K luwer Law International
Elkin-K oren, N iva (1995) CopyrightLaw and SocialD ialogue on the Inform ation Superhighw ay:The Case
Against Copyright Liability ofBulletin Board O perators, 13 CARDOZ O ARTS & EN T. L.J. 345, 399-410
Ellickson, R.C. (1998) Law and Econom ics D iscovers SocialN orm s, XXV II Journal of Legal Studies Part II
at 537-552, at 549
Eltzroth, Carter (1999) E-Com m erce: O pen for Business,Study on the Classification ofStandardisation
Requirem ents in Electronic Com m erce for W orld Bank and the European Com m ission DG MARK T
ETP 2006 (01) O n the technology,business m odels and regulatory aspects ofN G N at http://www.etp-
online.org/downloads/06_01_ETP% 20public% 20N G N % 20final.pdf
European Parliam ent (2005) Legislative resolution on the proposal for a recom m endation of the European
Parliam ent and of the Council on the protection of m inors and hum an dignity and the right of reply in
relation to the com petitiveness of the European audiovisual and inform ation services industry, 7 Septem ber
(COM2004-0341 – C6-0029/2004 – 2004/0117(COD)) at
http://www.europarl.eu.int/om k/sipade3?PU BREF=-//EP//TEXT+TA+P6-TA-2005-
European Parliam ent(2004) resolution on the application of Articles 4 and 5 of Directive 89/552/EEC
("Television without Frontiers"), as am ended by Directive 97/36/EC, for the period 2001-2002
(2004/2236(IN I)) at P6_TA-PROV (2005)0322 at http://www.europarl.eu.int/om k/sipade3?PU BREF=-
//EP//TEXT+TA+P6-TA-2005-
Fallenböck, Markus (2003) ‘On the technical protection of copyright: the Digital Millennium Copyright
Act, the European Com m unity Copyright Directive and their anti-circum vention provisions’ International
Journal of Com m unications Law and Policy 7 at
http://www.ijclp.org/7_2003/ijclp_webdoc_4_7_2003.htm
Farrell Joseph & Philip J. W esier (2003) M odularity,VerticalIntegration,and O pen Access Policies:Tow ard a
Convergence ofAntitrustand Regulation in the InternetAge, Departm ent Of Econom ics U CB Paper E02-
325
Federal Trade Com m ission (2005) Effectiveness and Enforcem entofthe CAN -SPA M Act:A FederalTrade
Com m ission Reportto Congress Decem ber at http://www.ftc.gov/reports/canspam 05/051220canspam rpt.pdf
Figueiredo, R.J.P. and Spiller, P. (2000) Strategy,Structure and Regulation:Telecom m unications on the N ew
Econom y, 2000 L.REV . M.S.U .-D.C.L. 1 at pp253-285
Flynn, Barry (2001) 4 Play:In turning its attention to the W eb,D VB has placed M PEG 4 as a possible
destination on its roadm ap, 4 Inside Digital TV 5, March 12 at pp6-7
177
Foster, C.D. (1992) Privatization,Public O w nership and the Regulation ofN aturalM onopoly, Blackwell,
Oxford
Foucault, M. (1979) D iscipline and Punishm ent:The Birth ofthe Prison
Freedm an, J.O. (1978) Crisis and Legitim acy in the Adm inistrative Process:A H istoricalPerspective,
Cam bridge U niversity Press
Froom kin, Michael A. (2000) ‘Sem i-private internationalrule-m aking:lessons learned from the W IPO dom ain
nam e process’, Chapter 11 in Marsden, C. (ed)
Fukuyam a Francis (1992) The End ofH istory and the Last M an, Penguin.
G . Madden (ed.)(2004) InternationalH andbook ofTelecom m unications Econom ics V . III, Edward Elgar
G aines, Sanford E. and Cliona K im ber (2001) Redirecting Self-Regulation, Env. Law 13(157)
G arzaniti, L. J. H. F. (2000) Telecom m unications,broadcasting,and the Internet:E.U . com petition law and
regulation. Sweet and Maxwell
G eist, M. (2006) N ew Yahoo D ecision Raises O ld Q uestions

http://news.bbc.co.uk/2/hi/technology/4641244.stm and
http://www.m ichaelgeist.ca/index.php?option=com _content& task=view& id=1083
G eist, Michael (2000) iCraveTV and the N ew Rules ofInternet Broadcasting, 23 U niversity of Arkansas at
Little Rock Law Review 123
G eorghiou, L. (2003) Foresight:Conceptand Practice as a Toolfor D ecision M aking, paper presented to the
Technology Foresight Sum m it, Budapest, March 27-29.
G eradin, D. and O’Donoghue, R. (2005) The Concurrent Application ofCom petition Law and Regulation:
The Case ofM argin Squeeze A buses in the Telecom m unications Sector, G CLC W orking Paper 04/05, College
of Europe, Bruges
G ibbons (2005) Com petition policy and regulatory style – issues for O fCom , 7 info 5 at 42-51
G ibbons, T. (2006 forthcom ing) Regulating the M edia 3rd Edition, Sweet and Maxwell
G ifford, Daniel J. & David McG owan (1999) A M icrosoft D ialog, 44 AN TITRU ST BU LL 619
G ilder, G . (1993) M etcalfe’s Law and Legacy, Forbes ASAP 13 Septem ber at
http://www.seas.upenn.edu/~gaj1/m etgg.htm l
G oldberg, D. and V erhulst, S. (1997) LegalResponses to Regulating the Changing M edia in the U nited
K ingdom 8 U til L.R.
G oldring John (1997) N etting the Cybershark: Consum er Protection,Cyberspace,the N ation-State,and

D em ocracy in K ahin and N esson eds. (1997) Borders in Cyberspace 322
178
G oldsm ith Jack and Tim W u (2006) D igitalBorders:N ationalboundaries have survived in the virtualw orld-
and allow ed nationallaw s to exert controlover the Internetat http://www.legalaffairs.org/issues/January-
February-2006/feature_goldsm ith_janfeb06.m sp
G oldsm ith, Jack (1998) W hatInternetG am bling Legislation Teaches A boutInternetRegulation, 32 IN T’L
LAW Y. 1115
G oldsm ith, Jack and Tim W u (2006, forthcom ing) W ho Controls the Internet? Illusions ofa Borderless
W orld, Oxford U niversity Press.
G ould, Mark (2000) ‘Locating Internet G overnance:lessons from the standards process’, Chapter 10 in
Marsden, C. (ed) Regulating the G lobal Inform ation Society, Routledge, N ew York.
G reenstein, Shane (1992) 'Invisible H ands and Visible A dvisors:An Econom ic Interpretation of
Standardisation', Journal of the Am erican Society for Inform ation Science, vol. 43 no.8 (Septem ber 1992),
p. 543.
G rindley, P., Salant, D.J. and W averm an, L. (1999) Standards W ars:the use ofstandard setting as a m eans of
facilitating cartels in third generation w ireless telecom m unications, International Journal of Com m unications
Law and Policy 3(2) at http://www.ijclp.org/3_1999/ijclp_webdoc_2_3_19999.htm l
G SM Association (2004) http://www.gsm world.com /news/statistics/index.shtm l
Haberm as, J. (1970) Tow ards a RationalSociety, Boston, Beacon Press
Hafner, K . and Matthew Lyon (1996) W here W izards Stay U p Late.
Harcourt, A. (2004) European Institutions and Regulation ofthe M edia Industry.
Hardy, C. and Moller, C. eds (2004) Spreading the W ord on the Internet :16 A nsw ers to 4 Q uestions,
V ienna, OSCE
Hawkins, R. R. Mansell, W . E. Steinm ueller (1999) Tow ards D igitalInterm ediation in the Inform ation
Society, Journal of Econom ic Issues, V ol. XXXIII (2): 383-391.
Hayek, F. (1979) Law ,Legislation and Liberty. A N ew Statem ent ofthe Principles ofJustice and Political
Econom y. Routledge
Held and Scheuer (2006) D raft Report:Study on Co-Regulation M easures in the M edia Sector; Study for the
European Com m ission, Directorate Inform ation Society and Media U nit A1 Audiovisual and Media
Policies, Hans Bredow Institute
Henderson, Hazel (1998) View ing "The N ew Econom y" From D iverse Forecasting Perspectives, 30 Futures 4,
PP. 267-275
Higgott R., U nderhill, G . and Bieler, A. eds (2000) N on-State Actors and A uthority in the G lobalSystem ,
Routledge, London
179
High Tech 141

Hirschm an A. (1970) Exit,Voice and Loyalty – Responses to D ecline in Firm s,O rganizations and States,
Harvard U niversity Press
Hodgson, G eoffrey M. (1988) Econom ics and Institutions:A M anifesto for a M odern InstitutionalEconom ics,
Cam bridge, U K : Polity Press
Hoeren, Thom as (2004) The first-ever ruling on the legalvalidity ofthe G PL - A Critique ofthe Case at
http://www.oii.ox.ac.uk/resources/feedback/OIIFB_G PL3_20040903.pdf
Hoffm an-Riem , W olfgang (1996) Regulating M edia:The Licensing and Supervision ofBroadcasting in Six
Countries G uildford Press
Hugenholtz P.B. (2000) ‘The G reatCopyright Robbery. Rights allocation in a digitalenvironm ent’, paper
presented at N ew York U niversity School of Law, 2 April, at
http://www.ivir.nl/publications/hugenholtz/PBH-Ecology.doc
Hugenholtz, Bernt (2002) ‘Copyright and Freedom ofExpression in Europe’, in N . Elkin-K oren & N .W .
N etanel (eds.), The Com m odification ofInform ation, Inform ation Law Series, V ol. 11, K luwer Law
International, pp. 239-263.
Hum phreys, Peter J. (1996) M ass m edia and m edia policy in W estern Europe
Huntington Sam uel P. (1996) The Clash ofCivilizations and the Rem aking ofW orld O rder, N ew York:
Touchstone
ICSTIS (2003) A nnualActivity Report2002 at

http://www.icstis.org/icstis2002/pdf/ACTIV ITY_2002.PDF
IDATE Digiworld (2003) The W orld Internet Access M arket, Montpelier, France.
Independent Mobile Classification Body (2005) IM CB G uide and Classification Fram ew ork for U K M obile
O perator Com m ercialContentServices at p6 at
http://www.im cb.org.uk/assets/docum ents/ClassificationFram ework.pdf
Internet G overnance Consortium (2005) The Future U S Role in Internet G overnance:7 Points in Response to
the U S Com m erce D epartm ent’s Statem ent on U S Principles
Internet Society (2005) brief introduction to the history of the Internet at:
http://www.isoc.org/Internet/history/brief.shtm l
IRIS (2003) Special:Co-Regulation ofthe M edia in Europe, Strasbourg, Council of Europe at

http://www.obs.coe.int/oea_publ/iris_special/2003.htm l.en
ISFE (2004) Press Release 14 D ecem ber:PEG I Celebrates Its FirstTw o Years O fO perations at
http://www.isfe-
180
eu.org/index.php?PHPSESSID=ea3fccbb3951ccab88f9ca9a568e1772& tem plate[0]=m atrice.htm l& tem pla

te[1]=rubrique.htm l& oidit=T001:96dec7f314175b346499b34f5ad64fda
ISFE (2005) Appendix to ISFE’s com m ents on Issue 6:Protection ofM inors and H um an D ignity, Septem ber at
http://www.europa.eu.int/com m /avpolicy/revision-tvwf2005/docs/ip6-isfe-encl.pdf
ISFE (2005) Position paper on the Com m unication from the Com m ission “Challenges for the European
Inform ation Society beyond 2005” at http://www.isfe-
eu.org/index.php?PHPSESSID=804e6e3cfdc7506303904aa81afe850e& tem plate[0]=m atrice.htm l& tem pla
te[1]=rubrique.htm l& oidit=T001:dfb8bc4a6dcf3052747cea0839f32f75
ITU (2003) Birth ofBroadband at

http://www.itu.int/osg/spu/publications/sales/birthofbroadband/BoBexecsum m .pdf
ITU (2004) Shaping the Future M obile Internet Society:The Case ofJapan, Docum ent SMIS/06, authored
by Srivastava and K odate, at
http://www.itu.int/osg/spu/ni/futurem obile/general/casestudies/JapancaseLS.pdf
ITU (2005) Initialcom m ents by the European U nion and the acceding countries Rom ania and Bulgaria,on the
reportofthe W orking G roup on InternetG overnance 1 August, W SIS-II/PC-3/CON TR/19-E,.
ITU (2005) The InternetofThings, Appendix at

http://www.itu.int/osg/spu/newslog/Broadband+Bargains+2005.aspx
ITU (2006) N G N W orking G roup. www.itu.int
Jaszi(2000) Video O n The Internet:Icravetv.Com A nd O ther RecentD evelopm ents In W ebcasting Hearing
Before The Subcom m ittee On Telecom m unications, Trade, And Consum er Protection Of The
Com m ittee On Com m erce House Of Representatives - One Hundred Sixth Congress; Second Session:
February 16, 2000 Serial N o. 106-94. Testim ony of Peter Jaszi, at:
http://com notes.house.gov/cchear/hearings106.nsf/a317d879d32c08c2852567d300539946/8d45454ad29
3f0db85256965006e67c1/$FILE/94.pdf
Jessup, P. (1956) TransnationalLaw .
Jowell and Oliver eds. (1994) The Changing Constitution (3rd edition) Clarendon Press
Jum pTV (2001) Copyright Board D enies Application to Suspend Jum pTV Proceedings June 4 at
http://www.jum ptv.com /m ediaroom /pr2001-06-04.1.htm l
K ahin, B. and N esson C. eds.(1997) Borders in Cyberspace:Inform ation Policy and the G lobalInform ation
Infrastructure, MIT Press
K ahin, Brian and Abbate, Janet eds. (1995) Standards Policy for Inform ation Infrastructure, MIT Press
K ahin, Brian and K eller, Jam es H. (eds)(1997) Coordinating the Internet, MIT Press
181
K ahn (1976) The N ext200 Years
K ahn Robert E. and V inton G . Cerf (Decem ber, 1999) W hat Is The Internet(And W hat M akes It
W ork)Internet Policy Institute W ashington D.C. at http://www.internetpolicy.org/briefing/12_99.htm l
K aplan, A. , A. L. Skogstad, M. A. G irshick (1949) The Prediction ofSocialand TechnologicalEvents. P-93

RAN D Corporation, Santa Monica, Cal.
K aul, Inge, Isabelle G runberg and Marc Stern (eds) (1999) G lobal Public G oods: International
Cooperation in the 21st Century, Oxford U niversity Press,
K oh, Harold A. (1996) TransnationalLegalProcess 75 N eb, L. Rev 181
K oh, Harold A. (1997) W hy do nations obey internationallaw ? 106 Yale L.J. 2599
Laffont J.J. and Tirole J. (2000) Com petition in Telecom m unications, MIT Press
Larouche P. (2000) Com petition Law and Regulation in European Telecom m unications, Hart
Larouche, Pierre (2001) ‘Com m unications convergence and public service broadcasting’, at
http://infolab.kub.nl/uvtweb/bin.php3?id=00011353& m im e=application/pdf& file=/tilec/publications/laro
uche2.pdf
Lem ley, M.A. & David McG owan (1998) LegalIm plications ofN etw ork Econom ic Effects, 86 Calif. L. Rev.
479
Lem ley, M.A. (1999) Standardizing G overnm ent Standard-Setting Policy for Electronic Com m erce, Berkeley
Technology Law Journal, V ol. 14, P. 745;
Lem ley, M.A. (2000) ‘W illthe Internetrem ake antitrust law ?’Chapter 12 in Marsden, C. (ed) Regulating
the G lobal Inform ation Society, Routledge, N ew York;
Lem ley, M.A. and Lessig, L. (1999) Ex Parte D eclaration O fProfessor M ark A. Lem ley And Professor
Law rence Lessig In The M atter O f:Application For Consent To The Transfer O fControlO fLicenses M ediaone
G roup,Inc. To At& T Corp CS Docket N o. 99-251 Before The Federal Com m unications Com m ission,
W ashington, D.C. 20554at Paragraph 4.
Lessig, L. (1996) Reading the Constitution in Cyberspace 45 Em ory Law Journal 869
Lessig, L. (1998) The N ew Chicago SchoolXXV II Journal of Legal Studies Part II 661-691
Lessig, L. (1999) Briefas Am icus Curiae U .S. v. M icrosoft, 65 F.Supp.2d 1 (D.D.C. 1999) (N o. Civ. 98-
1232 (TPJ), Civ. 98-1233 (TPJ) at < cyber.law.harvard.edu/works/lessig/AB/abd9.doc.htm l>
Lessig, L. (1999) Code,and O ther Law s ofCyberspace, Free Press. N ew York.
Lessig, L. (1999) The Lim its in O pen Code:Regulatory Standards and the Future ofthe N et, 14 Berkeley
Tech. LJ 2 at 759-770.
182
Lessig, L. (2001) The Future ofIdeas:The Fate ofthe Com m ons in a Connected W orld, Random House;
Lessig, L. (2004) Free Culture:H ow Big M edia U ses Technology and the Law to Lock D ow n Culture and
ControlCreativity, Penguin Books
Levy, B. and Spiller P. (1994) The InstitutionalFoundations ofRegulatory Com m itm ent:A Com parative
Analysis ofTelecom m unications Regulation 10 J. Law, Econom ics and Organization 2 at 201-246
Levy, D.A.L. (1997) The regulation ofdigitalconditionalaccess system s. A case study in European policy
m aking, Telecom m unications Policy 21(7) at 661 – 676
Levy, D.A.L. (1999) Europe’s digitalrevolution:Broadcasting regulation,the EU and the nation state N ew
York: Routledge
Lewis, Paul (2005) Courtverdict hailed as spam stopper, 28 Decem ber, The G uardian, at
http://business.guardian.co.uk/story/0,16781,1674429,00.htm l
Liebowitz S.J. and Stephen E. Margolis (1999) W inners,Losers & M icrosoft:Com petition A nd AntitrustIn
H igh Technology
Lindblom , C. (1977) Politics and M arkets:The W orld’s Political-Econom ic System s Basic Books: N Y
Mansell, R. and Roger Silverstone, R. eds (1996) Com m unication by D esign:The Politics ofInform ation and
Com m unication Technologies, Oxford U niversity Press
MarconiFoundation (2006) MarconiPrize: see

http://www.m arconifoundation.org/pages/news_room /index.htm
Marcus, Scott (2005) Is the U S dancing to a differentdrum m er? 60 Com m unications et Strategies pp39-58.
Marsden (2000) N otSo Special2 info 1
Marsden (2005) Free,O pen or Closed – A pproaches to the Inform ation Ecology 7 info 5 at 6-19
Marsden C. (2000) ‘The European digitalconvergence paradigm :from structuralregulation to behavioural

com petition law ?’chapter 20in Erik Bohlin et al (eds.) Convergence in Com m unications and Beyond, Elsevier
Science
Marsden, C. (1999) Pluralism in M ulti-ChannelD igitalCom m unications:Suggestions for Regulatory Scrutiny,

MM-S-PL [99] 12 def 2, available at IJCLP W eb-Doc 5-4-2000 at
http://www.ijclp.org/4_2000/pdf/ijclp_webdoc_5_4_2000.pdf
Marsden, C. (2001) Cyberlaw and InternationalPoliticalEconom y:Tow ards Regulation ofthe G lobal
Inform ation Society, 2001 L.REV . M.S.U .-D.C.L. 1 at pp253-285
Marsden, C. (2001) The StartofEnd-to-End:InternetProtocolTelevision 29 Interm edia 3
183
Marsden, C. (2003) K eynesian-K orean Strategies for Recovering from the G reatTM T D epression 31 Interm edia 2
at 38-42
Marsden, C. (2003) Video over IP,the Challenges ofStandardisation:Tow ards the N ext-G eneration Internet
Chapter 8 in EliN oam et al (eds) Television over the Internet:Im plications for N etw orks,Infrastructures,
Strategy,Policy and Content, Lawrence Erlbaum , Mahwah, N ew Jersey
Marsden, C. (2004) Editorial... Illegaland O ffnet:Is Peer to peer the next M obile Regulation Challenge? at
http://www.selfregulation.info/iapcoda/0402xx-selfregulation-review.txt.
Marsden, C. (2004) H yperglobalized Individuals:the Internet,globalization,freedom and terrorism 6

Foresight 3 at 128-140.
Marsden, C. (2005) Co-Regulation in European M edia and InternetSectors Multim edia und Recht 1 at 3-7
Marsden, C. ed. (2000) Regulating the G lobalInform ation Society, Routledge
Marsden, Christopher T., and V erhulst, Stefaan G . (1999) Convergence in European D igitalTV Regulation,
London, Blackstone Press.
Mathiason and Mueller (2005) Q uo Vadis? At http://dcc.syr.edu/m iscarticles/IG P-quovadis.pdf
Mayer-Schoenberger, V . (2003) Crouching Tiger,H idden D ragon:Proxy Battles over P2P File Sharing,
m im eo
Mayer-Schonberger V iktor and Teree E. Foster (1997) A Regulatory W eb:Free Speech and the G lobal
Inform ation Infrastructure in K ahin and N esson (eds) Borders in Cyberspace;
McCahery, J., Bratton, W .W ., Picciotto, S., and Scott, C., eds (1996) InternationalRegulatory Com petition
and Coordination Oxford U niversity Press
McChesney Fred (1999) Econom ics versus Politics in A ntitrust, 23 Harvard J. L. and Public Policy 1
McCullagh (2005) U S Reasserts Controlover the Internet, 1 July – Asst. Com Sec Michael G allagher,
“m aintain its historic role in authorizing changes or m odifications to the authoritative root zone file”
http://news.zdnet.co.uk/Internet/0,39020369,39206653,00.htm
McG owan, David (2000) ‘The Problem s ofthe Third W ay:A Java Case Study’, Chapter 13 in Marsden, C.
(ed) Regulating the G lobal Inform ation Society, Routledge, London and N ew York
Meadows (1972) Lim its to G row th
Meesen K arl M. (1989) An InternationalAntitrust Challenge:Com petition ofCom petition Law s, 10 Int. L.
and Bus. 17.
Minges, M. (2003) at http://www.itu.int/ITU -D/ict/publications/wtdr_03/m aterial/W TDR03.pdf
184
Mobile N etwork Operators (January 2004) U K code ofpractice for the self-regulation ofnew form s ofcontent
on m obiles available at www.orange.co.uk
Moe Terry M. (1997) The Positive Theory ofPublic Bureaucracy in Dennis C. Mueller (ed) Perspectives on
Public Choice:A H andbook, Cam bridge U niversity Press, Cam bridge, at pp456-9.
Molka, Judith (1992) 'Surrounded by Standards,There is a Sim pler View ', Journal of the Am erican Society
for Inform ation Science, vol. 43 no. 8 (Spring 1992), p. 526
Morth, U . ed (2004) SoftLaw in G overnance and Regulation:A n Interdisciplinary Analysis, Edward Elgar.
Mueller Dennis C. (ed 1997) Perspectives on Public Choice:A H andbook, Cam bridge U niversity Press,
Cam bridge
Mueller Dennis C. (ed)(1997) Perspectives on Public Choice:A H andbook, Cam bridge U niversity Press
Mueller, Milton (2000) Technology and Institutional Innovation: Internet Dom ain N am es, 5
International Journal of Com m unications law and Policy at
http://www.ijclp.org/5_2000/ijclp_webdoc_1_5_2000.htm l
Mueller, Milton (2002) Ruling the Root:Internet G overnance and the Tam ing ofCyberspace, MIT Press
Murdoch, Rupert (2006) Interview w ith JeffRandall,BBC, at

http://news.bbc.co.uk/1/hi/help/3681938.stm
Murray and K lang (eds) (2005) H um an Rights in the D igitalAge, G lasshouse Press, London.
Murray, A. (2006 forthcom ing) Regulatory W ebs and W ebs ofRegulation:Regulating the D igital
Environm ent, G lasshouse/U CL Press.
N aftel, Mark and Spiwak, Lawrence (2001) The Telecom s Trade W ar, Hart Publishing,
http://www.phoenix-center.org/telindex.htm l
N ational Consum er Council (2001) Soft Law in the European U nion, at

http://www.ncc.org.uk/europe/softlaw.pdf
N aughton John (1999) A BriefH istory ofthe Future
N eil, Suzanne, Lee McK night and Joseph Bailey (1995) 'The G overnm ent's Role in the H D TV Standards
Process:M odelor A berration', in Brian K ahin and Janet Abbate, Standards Policy for Inform ation
Infrastructure ,pp.276-289
N eiswender. R. (1966) The Exploration ofthe Future [Realites] N o.245, June 1966, Pp 50-58.
N eum an, R., McK night, L., & Solom on, R. (1997) The G ordian K not:Politicalgridlock in the inform ation
highw ay. Cam bridge, MA: MIT Press;
185
N oam , E. (1989) InternationalTelecom m unications in Transition, pp257-297 in Crandall, Robert W . and

Flam m , K enneth (eds) Changing the Rules:technologicalchange,internationalcom petition,and regulation in
com m unications, Brookings Institute, W ashington D.C.
N oam , E. (2000) The Com ing Cyber Trade W ars in Erik Bohlin et al (eds) Convergence in Com m unications
and Beyond, Elsevier Science.
N oam , E. (2003) The EffectofD eregulation on Industry Concentration:A n A nalysis ofthe Telecom s Actof
1996 and the Industry M eltdow n, Colum bia Science and Technology Law Review at
http://www.stlr.org/htm l/volum e4/noam .pdf
N oam , EliM. et al (ed.), 2003, Television over the Internet:Im plications for N etw orks,Infrastructures,
Strategy,Policy and Content, Lawrence Erlbaum
N orth, D. (1990) Institutions,InstitutionalChange and Econom ic Perform ance, Cam bridge U niversity Press
N ossal, K im Richard (1999) Lonely Superpow er or U napologetic H yperpow er? A nalyzing Am erican Pow er in
the Post-Cold W ar Era, South African Political Studies Association, Saldanha, W estern Cape at
http://post.queensu.ca/~nossalk/papers/hyperpower.htm
N TIA (205) U S D om ain N am e System Principles speech at W ireless Com m unications Association, 30 June
2005 http://www.ntia.doc.gov/ntiahom e/dom ainnam e/U SDN Sprinciples_06302005.htm
OECD (2003) ICCP Broadband U pdate October 2003 at

http://www.oecd.org/dataoecd/18/9/18464850.pdf
OECD (2005) D igitalBroadband Content:The O nline Com puter And Video G am e Industry
DSTI/ICCP/IE(2004)13/FIN AL at pp53-54, at http://www.oecd.org/dataoecd/19/5/34884414.pdf
OECD (2005) InternationalFutures Program m e

http://www.oecd.org/docum ent/59/0,2340,en_2649_34219_2410299_1_1_1_1,00.htm l
OfCom (2004) Consultation D ocum ent ‘Criteria for Transferring Functions to co-Regulatory
Bodies’http://www.ofcom .org.uk/consultations/past/co-reg/?a=87101
OfCom (2004) http://www.ofcom .org.uk/consultations/current/wbam /wbam .pdf
Office of the Telecom Adjudicator (2006) January 13LLU Forecastat http://www.offta.org.uk/charts.htm
Oftel (2001) Vodafone,O 2,O range and T-M obile:Reports on references under section 13 ofthe
Telecom m unications Act 1984 on the charges m ade by Vodafone,O 2,O range and T-M obile for term inating
calls from fixed and m obile netw orks; V olum e 2: Chapters 3 to 15
http://www.ofcom .org.uk/static/archive/oftel/publications/m obile/ctm _2002/docs_index.htm
Oftel (2003) Septem ber Broadband Briefat

http://www.ofcom .org.uk/static/archive/oftel/publications/internet/internet_brief/broad1003.pdf at p22.
186
Oftel’ (2000) FRIACO D irection ofM ay 2000 at

http://www.ofcom .org.uk/static/archive/oftel/publications/internet/fria0400.htm
Ogus, A. I. (1994) Regulation:LegalForm and Econom ic Theory
Ohm ae, K . (1991) The Borderless W orld:Pow er and Strategy in the Interlinked Econom y, Harper Business,
N Y.
Ohm ae, K enichi(1990) The Borderless W orld: Power and Strategy in the Interlinked Econom y, Collins,
London
Olson, M. (1971) The Logic ofCollective A ction, Harvard U niversity Press
Ovum (2005) U K Broadband in the G 7 at http://www.theregister.co.uk/2005/02/22/govt_boradband_g7/
Owen, B. (1999) The internet challenge to TV. Cam bridge, MA, Harvard U niversity Press;
Oxford Internet Survey (2005) at http://www.oii.ox.ac.uk/research/oxis/oxis2005_results.pdf
PEG I (2005) Info newsletter N o.7 at http://www.pegi.info/pegi/download.do?id=12
Pitofsky, Robert (1998) ‘SelfRegulation And A ntitrust’, Prepared Rem arks of Chairm an, Federal Trade
Com m ission, D. C. Bar Association Sym posium , February 18, 1998, W ashington, D.C. at
http://www.ftc.gov/speeches/pitofsky/N _2_#N _2_
Posner, R. A. (1984) Theories ofEconom ic Regulation 5 Bell Journal of Econom ics and Managem ent
Science 335
Posner, R.A. (1998) SocialN orm s,SocialM eaning and Econom ic A nalysis ofLaw :A Com m entXXV II
Journal of Legal Studies Part II at 553-565
Price W aterhouse Coopers (2005) G lobalEntertainm entand M edia O utlook 2005-9 from
http://www.pwcglobal.com /extweb/industry.nsf/docid/8CF0A9E084894A5A85256CE8006E19ED?opend
ocum ent& vendor=none#IAAS
Price, M. (1995) Television,The Public Sphere and N ationalIdentity Oxford: Oxford U niversity Press
Price, M., V erhulst, S. (2004) Self-Regulation and the Internet, N etherlands: K luwer
Price, Monroe and V erhulst, Stefaan (2000) ‘In search ofthe self:charting the course ofself-regulation on the
Internetin a globalenvironm ent’, Chapter 3 in Marsden, C. (ed) Regulating the G lobal Inform ation
Society;
Prosser, Tony (1994) “Regulation, Markets and Legitim acy” Chapter 9, 237-260 in Jowell and Oliver (eds)
The Changing Constitution (3rd edition) (1994) Clarendon Press
Q uade (1969) The System s Approach and Public Policy, RAN D P-4053 RAN D Corporation, Santa Monica,
Cal.
187
Rankine, John (1995) 'D o W e N eed a N ew Standards System ?'in Brian K ahin and Janet Abbate, Standards
Policy for Inform ation Infrastructure, pp. 564-578.
Reding V iviane (2005) ‘M obile Com m unications :a key driver to m ake Lisbon succeed’delivered to the
3G SM W orld Congress, Cannes, 14th February 2005.
Reding, V iviane (2005) Better regulation for Europe’s m edia industry:the Com m ission’s approach, Speech
05/532.
Reed, J. and Angel, I. (2003) Com puter Law , 5th Ed, OU P.
Reed, J. and Angel, I. (2003) Com puter Law , 5th Ed, Oxford U niversity Press
Reidenberg, J. (1993) Rules ofthe Road for G lobalElectronic H ighw ays:M erging the Trade and Technical
Paradigm s, 6 Harvard Journal of Law and Technology 287
Reidenberg, J. (1999) Restoring Am ericans’Privacy in Electronic Com m erce, 14 Berkeley Technology Law
Journal 2 at 771-792
Reidenberg, J. (2002) Yahoo and dem ocracy on the Internet, 42 Jurim etrics 261.
Reidenberg, J. (2004) States and Law Enforcem ent, 1 U ni.Ottawa L.& Tech.J.
Reidenberg, J. (2005) Technology and Internet Jurisdiction, 153 U N IV . OF PEN N . L. REV . 1951
http://ssrn.com /abstract=691501
Rescher (1967) The Future as an O bjectofResearch, RAN D P-3593 RAN D Corporation, Santa Monica,
Cal.
Resnick, Paul and Jam es Miller (1996) PICS:Internet A ccess Controls W ithout Censorship, Association for
Com puting Machinery vol. 39(10), pp. 87-93 at http://www.w3.org.PICS.iacwcv2.htm
Rhoads, Christopher (2006) Should the W orld M ake Room for Another W ide W eb? W all Street Journal 22
January page 1, at http://www.courant.com /business/hc-wsjinternet.artjan22,0,4344057.story?coll=hc-
headlines-business
Rice, C. (2005) 7 N ovem ber Letter to Jack Straw http://www.theregister.co.uk/2005/12/02/rice_eu_letter/

The Register
Richardson, M. and Hadfield, G . (1999) The Second W ave ofLaw and Econom ics, Federation Press, Sydney
Robins, K . and W ebster, F. (1999) Tim es of the Technoculture: Inform ation, Com m unication and the
Technological Order, Routledge, London.
Robinson C. ed.(2004) Successes and Failures in Regulating and D eregulating U tilities, Edward Elgar.
RTL Television G m bH v. N iedersächsische Landesm edienanstalt für privaten Rundfunk (2003) Case C-245/01
23 October 2003, nyr, (see http://curia.eu.int).
188
Saltzer, Jerom e W ., David P. Reed, and David D. Clark (1984) End-to-end argum ents in system design,
ACM Transactions in Com puter System s 2(4) pp277-288
Sam uelson P. and K urt Opsahl (1999) Licensing Inform ation in the G lobalInform ation M arket:Freedom of
Contract M eets Public Policy, 21 E.I.P.R. 386
Sam uelson, Paul A. (1954) The Pure Theory ofPublic Expenditure, 11 Review of Econom ics and Statistics
387-9.
Sandbach, Jonathan (2001) Levering open the localloop:shaping BT for broadband com petition, 3 info 1 at
195-202
Schiller, D. (1999) D eep Im pact:The W eb and the changing m edia econom y, 1 info 1 at 35-52.
Schneider, Andrea (1995) “Child Protection on G erm an Television – The Voluntary Television Review Body
(FSF)”, IRIS -3:7/13.
Schulz and Held (2001) Regulated Self-Regulation as a Form ofM odern G overnm ent. Hans Bredow Institut.
Schum peter (1942) Capitalism ,Socialism and D em ocracy
Scott, C. (1997)The Proceduralization ofTelecom m unications Law :A dapting to Convergence JILT [Journal of
Inform ation, Law and Technology] in: http://elj.warwick.ac.uk/jilt/wip/97_3scot/default.htm
Seeley Brown, John (2002) The SocialLife ofInnovation in the D igitalAge, 15 July presentation at
http://www.ruschlikon.net/IN TERN ET/rschwebp.nsf/(ID)/6C5A73B4FEBA95A9C1256C13002820A2/
$FILE/JSB-speech-.pdf
Select Com m ittee on European Scrutiny (2000) 21st Report, Annex 7, M anagem ent ofthe Internet,
Com m ission Com m unication COM (99) 02 at
http://www.pubications.parliam ent.uk/pa/cm 199900/cm select/cm euleg/23-xxi/2310.htm
Shapiro, C. and V arian, H.R.(1998) Inform ation Rules:A Strategic G uide to the N etw ork Econom y, Harvard
Business School Press
Shelanski, H. (1999) The Speed G ap:Broadband Infrastructure and Electronic Com m erce, 721-744 in 14
Berkeley Tech.LJ 2
Shinoda (2004) BB A ccess N etw ork D eploym ent in Japan Past Presentand Future:Reasons and Issues,
Presentation to IN ET 2004, Barcelona, May, available from this author
Sidak J.G . (2004) The failure ofgood intentions:the collapse ofA m erican telecom m unications after six years of
deregulation in Robinson C. (ed), Successes and Failures in Regulating and Deregulating U tilities, E. Elgar,
1-15
Slaughter, A-M. (1995) InternationalLaw in a W orld ofLiberalStates, 6 Eur. J. Int. Law 503
Slaughter, A-M., Tulum ello, A. S., W ood, S. (1998) 92 Am . J. of Int. Law 3 at 367-397
189
Snyder, F. (1995) “O ut on the W eekend”:Reflections on European U nion Law in Contextin W ilson, G .P.
Frontiers ofLegalScholarship, W iley & Sons
Soete, L. et al (1997) Building the European Inform ation society for us all, Final report of the High Level
Experts G roup, April 1997
Soros, G . (2004) The Bubble ofAm erican Suprem acy,
Spar, D. (2001) Pirates,Prophets and Pioneers:Business and Politics A long the TechnologicalFrontier,
Random House, N Y.
Standage, Thom as (1999) The Victorian Internet, Phoenix, N Y.
Stanfield, J. R. (1999) The Scope,M ethod and Significance ofO riginalInstitutionalEconom ics, XXXIII
Journal of Econom ic Issues 2, 231-256
Stiglitz, Joseph E. (1985) Inform ation and econom ic analysis:A perspective, Econom ic Journal 95
(supplem ent) pp21-41
Stiglitz, Joseph E. (1985) Inform ation and econom ic analysis:A perspective, Econom ic Journal 95
(supplem ent), pp21-41.
Strange (1998) W ho are EU ?Am biguities in the ConceptofCom petitiveness, 36 Journal of Com m on Market
Studies 1 at 101-114, at 105
Strange, S. (1970) InternationalRelations and InternationalEconom ics:A Case ofM utualN eglect, 46
International Affairs 2 at 304
Strange, S. (1998) W hat Theory? The Theory in M ad M oney [N ovem ber]

www.warwick.ac.uk/fac/soc/CSG R/wpapers/wp1898.PDF
Talbot, D. (2005) ‘The Internet is Broken’Technology Review, 19 Decem ber 2005,

http://www.technologyreview.com /InfoTech-N etworks/wtr_16051,258,p1.htm l
Tam bini, D. , Leonardi, D. and Marsden, C. (2006, April) Codifying Cyberspace:SelfRegulation of

Converging M edia, Routledge, London
Tem ple-Lang, J. (1998) The D uty ofN ationalAuthorities under Com m unity ConstitutionalLaw 23
European Law Review 109 at 119
Teubner (ed. 1986) D ilem m as ofLaw in the W elfare State, Berlin: W . de G ruyter
Teubner, G . (1986) 'The Transform ation ofLaw in the W elfare State'in G . Teubner (ed.), D ilem m as of
Law in the W elfare State, Berlin: W . de G ruyter
Thierer, A. ed. (2004) W ho Rules the N et? Cato Institute, W ashington DC.
Tiebout, C. (1956) A pure theory ofpublic expenditures, 64 Journal of Political Econom y 416-424.
190
Tongue, C. (1999) Chapter 5 in Marsden, C. and V erhulst, S. (eds) (1999) Convergence in European
D igitalTV Regulation.
Trachtm an, J. (1999) Book Review :the W TO as an InternationalO rganisation ed. A nne M . K rueger,London
U niversity ofChicago Press, 92 Am . J. of Int. Law 3 pp598-601 at 599.
Trachtm an, Joel P. (1998) Cyberspace,Sovereignty,Jurisdiction,and M odernism , Indiana Journal of G lobal

Studies V olum e 5 Issue 2 Spring 1998 at 561, Indiana U niversity School of Law – Bloom ington
http://www.law.indiana.edu/glsj/vol5/no2/10tract.htm l
Treaty on Principles G overning the Activities ofStates in the Exploration,D evelopm entand U se ofThe Internet
Com m ons (2004) at http://www.internationalunity.org/
Trebilock, M. (1999) The Value and Lim its ofLaw and Econom ics, pp12-29 at 29, in Richardson, M. and
Hadfield, G . (1999) The Second W ave ofLaw and Econom ics, Federation Press, Sydney
Trom penaars, A. and Ham pden Turner, C. (1993) The Seven Cultures ofCapitalism :Value System s for
Creating W ealth in the U nited States,Japan,G erm any,France,Britain,Sw eden,and the N etherlands,
Doubleday, London;
Trom penaars, A. and Ham pden Turner, C. (1997) Riding the W aves ofCulture:U nderstanding Cultural
D iversity in Business, N icholas Brealey, London
U .S. v. M icrosoftCorp., N o. 98-1233 (U .S. District Court for the District of Colum bia), Findings of Fact,
¶¶39-41. Available at www.usdoj.gov/atr/cases/f3800/m sjudgex.htm
U .S. v. M icrosoft, 65 F.Supp.2d 1 (D.D.C. 1999) (N o. Civ. 98-1232 (TPJ), Civ. 98-1233 (TPJ)
U N CTAD (1998) W orld Investm ent Report1998:Trends and D eterm inants, U N CTAD, G eneva.
U ngerer, H. (2002) ‘M edia in Europe:M edia and EU com petition law ’

http://europa.eu.int/com m /com petition/speeches/text/sp2002_006_en.pdf
U ngerer, Herbert (2005) Com petition in the m edia sector - how long can the future be delayed?
7 info 5 at pp. 52-60
U SG (2005) State D epartm ent Com m ents ofthe U nited States ofAm erica on InternetG overnance, 15 August
2005, at www.state.gov
V aisse, Justin (2001) L’hyper-Puissance A u D éfi D e L’hyperterrorism e at

http://www.brookings.edu/views/articles/fellows/20011023vaisse.pdf
V arian, Hal (2003) K eynote closing speech to ITS Europe 2003, HelsinkiFinland
V eblen (1967) A bsentee O w nership and Business Enterprise in Recent Tim es, Boston Beacon Press.
V ernon Raym ond (1971) Sovereignty atBay:The M ultinationalSpread ofU nited States Enterprises, Basic Books
191
V idal, G . (2002) PerpetualW ar for PerpetualPeace:H ow W e G otto Be So H ated,Causes ofConflict in the

LastEm pire, Clairview
V idal, G . (1969) Reflections U pon a Sinking Ship, Heinem ann, N Y.
V idal, G . (1994) U nited States:Essays 1952-1992, Abacus, N Y;
V idal, G . (2003) D ream ing W ar:Blood for O iland the Cheney-Bush Junta W eidenfeld and N icholson, N Y,
V olkm er Ingrid (1997) U niversalism and Particularism :The Problem ofCulturalSovereignty and G lobal
Inform ation Flow in K ahin and N esson (eds) Borders in Cyberspace
W akefield, Jane (2006) G oogle faces China challenges, 25 January BBC N ews
http://news.bbc.co.uk/1/hi/technology/4647468.stm
W are (1967) The Com puter in Your Future RAN D P-3626 RAN D Corporation, Santa Monica, Cal.
W arwick-OECD conference (June 1999) ‘Com m unications Regulation in the G lobalInform ation Society’:
http://www.digital-law.net/warwick/
W atson Brown, A. (1999) Industry Consortia and the Changing Roles ofStandards Bodies and Regulators, 35
Inst. Prospective Tech. Stud., June available at http://www.jrc.es/pages/f-report.en.htm l
W earden, G raham (2004) U K Police Chief:Shut D ow n A bhorrent W ebsites, at

http://news.zdnet.co.uk/internet/0,39020369,39147312,00.htm
W eiss, L. (1999) The M yth O fThe Pow erless State:G overning The Econom y In A G lobalEra Polity Press
W eiss, Martin and Marvin Sirbu (1990) 'TechnologicalChoices in Voluntary Standards Com m ittee:An
Em piricalA nalysis'4 Econom ics of Innovation and N ew Technology 1, April pp111-133
W erbach, K . (2005) FederalCom puter Com m ission N orth Carolina Law Review, at
http://papers.ssrn.com /sol3/papers.cfm ?abstract_id=763905
W hitehead, Phillip (1997) D raftReport on the Com m ission G reen Paper on The Protection ofM inors and
H um an D ignity in Audiovisualand Inform ation Services (COM[96]0483 - C4-0621/96) PE 221.804 of 24
April 1997.
W illiam son, O. (1975) M arkets and H ierarchies:A nalysis and A ntitrust Im plications, Free Press
W ilson, G .P. (1995) Frontiers ofLegalScholarship, W iley & Sons
W ood, Diane P. (1999) InternationalLaw and Federalism :W hat is the reach ofregulation? 23 Harvard J. L.
and Public Policy 1 at 97-110, at 110.
W oods, L. (2004) Free M ovem entofG oods and Services, Ashgate Publishing.
W oods, L. and Scheuer, A. (2004) Advertising Frequency and the Television W ithout Frontiers D irective,
29(3) European Law Review at 366-384
192
W orld Internet Project (2005) http://www.pewinternet.org/trends.asp led by the Pew Center
W oroch, G len A. (2002) O pen Access Rules and The Broadband Race, 3 L. REV . M.S.U .–D.C.L
W right, A. (2005) Coregulation ofFixed and M obile Internet Content,paper at‘Safety and Security in a
N etw orked W orld:Balancing Cyber-Rights and Responsibilities’conference, Oxford, Septem ber, at
http://www.oii.ox.ac.uk/research/cybersafety/?view=papers
W u, Tim (2003) N etw ork N eutrality & Broadband D iscrim ination, 2 J. Telecom . &
Yen. A. (2000) InternetService Provider Liability for Subscriber CopyrightInfringem ent,Enterprise Liability
and the FirstAm endm ent88 G eorgetown L.J. At
http://papers.ssrn.com /sol3/Delivery.cfm /SSRN _ID236478_code000726304.pdf?abstractid=236478
Yu Peter K . (2005) IntellectualProperty And The Inform ation Ecosystem Michigan State L.Rev. 1.
Yu, Christopher (2004) W ould M andating N etw ork Broadband N eutrality H elp O r H urt Com petition? A
Com m ent O n The End-To-End D ebate, J. Telecom . & High Tech;
Z ittrain, J. (2006 forthcom ing) The Regenerative Internet Harvard Law Review
Z ittrain, J. (2006) W ithouta N et- http://www.legalaffairs.org/issues/January-February-

2006/feature_zittrain_janfeb06.m sp
Z ittrain, J. and Edelm an, B. (2004) 'D ocum entation ofInternet Filtering W orldw ide', pp137-148 in Hardy,
C. and Moller, C. eds Spreading the W ord on the Internet:16 A nsw ers to 4 Q uestions, V ienna, OSCE.
193

Regulating The Broadband Internet: Towards An Interdisciplinary Analysis

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Regulating The Broadband Internet: Towards An Interdisciplinary Analysis

Transféré par

Droits d'auteur :

Formats disponibles

Internet Regulation 01/02/2006

Regulating the Broadband Internet:

Christopher T. Marsden LL.B. (Hons), LL.M. (Hons)

Researcher, Centre for Socio-Legal Studies, Oxford U niversity;

Teaching fellow, U niversity of Essex School of Law

Senior Analyst, RAN D Europe, Cam bridge

1st February 2006

5.2.2. Researching ICTs and W hy History Matters: Path Dependency........................................... 92

Chapter 1: Broadband Internet: The

1.1. Regulating the European Broadband Internet

1.2. Organisation of the Argum ent

Barlow’s declaration in 1996 in Davos and his institutionalisation of a libertarian perspective on

1.3. How to regulate the Internet

Figure: The Intersecting Elem ents in Lessig’s Model of Regulation

Market Incentives and

three Ellickson constraints, in addition to m arket constraints. He term s socialization

• governm ents have increasingly intervened in the Internet’s architecture to secure

Table: Adapting V erhulst’s Characterization of Lessig ‘Realm s of Social Order’

Architecture – Econom ics Social N orm s Law

"Internet" refers to the globalinform ation system that --

Econom ic Process Com ponent Cost-Efficiency Effect

1.6 Methodological N ote

Chapter 2: Broadband Internet Structure and Regulatory Challenges

Preface to International Cyberspace ‘Treaty’ 200465

2.1. Internet Regulation: Back to the Future?

2.2. W hom to Regulate on the Internet?

2.3. Jurisdictional Response to the Internet

W hereas in traditional broadcasting (analog or digital) the individual broadcaster is easily

Table: V ertical layer control of the m edia industries82

Control Analogue ‘free-to-air’ Printed Press Telecom s Digital pay-TV

2.2. Internet Industry and the Role of ISPs

Representing the value chain diagram m atically:

ICP ISP ISP U ser Public access Device Control

2.4.1. V ertical Integration of ISP s with ICPs

Access Content Access Peer to Peer Browser

Control Stage? Printed Press Telecom s Internet

2.4.2. Liability for Harm ful and Potentially Illegal Content

97 Section 30, 47 U.S.C. § 230(c)(1) (Supp. II 1996).

101 Ibid at 56.

2.5. Regulated Self

As in m ost European responses to technologically led change in the socio-econom ic landscape

2.6 From European to G lobal Internet Regulation Challenges

Chapter 3: The Institutional Perspective: Beyond N ational Regulation

3.1. W hy Doesn’t the Internet Change Everything?

3.2. G lobal Regulation and the Internet

3.3. Socioo--Econom ic Analysis of Inform ation N etworks as G lobal Public G oods

3.4. Public goods

3.4.1. Public G oods and Private Actors

3.4.2. Public Choice and ‘Races to the Bottom ’

3.4.3. G lobal Public G oods

Tem poral – Spatial Regional G lobal

Table: Societal Structure and Resource Inputs191

Bell explains that:

Chapter 4: Telecom m unications Regulation – Towards Behavioural Regulation

“An antitrust price squeeze test…yields no inform ation about consum er

4.1. Introduction: Institutional Failures in Com petition Policy

provisions’ International Journal of Communications Law and Policy, issue 7 at

Table: Developm ent of European Consum er Internet Market

Date Developm ent Com panies Result Legacy

240 See Oftel’s Direction of May 2000 at http://www.ofcom.org.uk/static/archive/oftel/publications/internet/fria0400.htm

4.2. Broadband Carriage: Infrastructure Deploym ent

Retailed by indep. ISP Retailed by incumbent LLU

3. lessons from 1970s-1980s cable deploym ent in the U S;