The FTTH Prism

Vol.5 No.1 January 2008

GPON Concerns, This Issue

David Chaffee, FTTH Prism.............................................................................................2
Issue 1---PON vs. Point-to-Point

PON-The Technology of Choice

Mario Rossi, Consultant
David Russell, Calix..........................................................................................................5

The Case for Active Ethernet

Irit Gillath, Telco Systems.................................................................................................15
Issue 2---FCC MDU Dominance Ruling and FTTH Implications

Weighing in on Exclusive Contracts for MDUs

Diane Kruse, Zoomy Communications........................................................................….20

A Government Operating Under A Faulty Argument for Competition

Timm Bechter, Waddell & Reed....................................................................................….24

Determining Factors As To Why Companies Do/Don't Go to FTTH

Max Kipfer, Connexion Technologies........................................................................……27

Publicly Funded “Open FTTH”

Mitch Shapiro, Independent Consultant......................................................................….30

Fiber To Every Home Is Our Ground-Breaking Strategy

Christy Batts, CDE Lightwave....................................................................................….39

Improving Economics and Quality of Life Through FTTH

John George, OFS Optics...........................................................................................…..40

Corning Connected Communities

Ginger Stevens and Bernhard Deutsch
Corning Cable Systems...............................................................................................…..51

Columns
Net Neutrality
Mario Rossi, Independent Consultant......................................................................……54

Software, of All Things

Jim Farmer, Wave7 Optics......................................................................................…….56
Verizon FiOS Joins The Military
David Chaffee, FTTH Prism............................................................................................59
GPON Concerns; This Issue
David Chaffee, Editor, FTTH Prism
cdcfiber@aol.com

There are problems in the GPON industry and it appears to be focused on the chips, although
other concerns including Verizon's economics and standards issues also are factors. Verizon FiOS has
already missed two rollout deadlines and both Alcatel-Lucent and Motorola are struggling to get out
First Office Applications (FOAs).

But is the problem with the chips or with Verizon's unrealistic demands for sophisticated
products at bargain-basement rates?

We no longer condemn those vendors that stayed out of FTTH because they felt they were
going to be crunched by the margins. Of course the classic example of that was poor Tellabs selling
products for less than they cost to produce. And that was on BPON.

GPON is considered a level of sophistication greater and yet various sources are telling us that
Verizon is holding to unrealistic pricing demands.

We understand the pressure Verizon is under. While FiOS generally is recognized on Wall Street
as a good thing, Verizon is still a ways from making money on it. And, like it or not, AT&T's U-Verse
FTTN architecture is picking up steam with the power of Cisco behind it. Analysts are pressuring
Verizon to keep the GPON costs down!

In some respects, does it make sense for Verizon to go to GPON now? We are not hearing a lot
of customer complaints about BPON at this time (we have BPON and it is more than satisfactory) and
quite honestly Verizon is in a leadership position against the other major telecom and Cable TV carriers
with it. It definitely does not make sense to roll out a more expensive product if the result is that FTTH
is not going to look economically competitive.

What the GPON slowdown has done has curbed Verizon rhetoric for the time being about 100
Mbps to every premise, or at least it should. While it hurts for me to say it, an enormous GPON rollout
does not make sense if people are not clamoring for it and it would lengthen Verizon's day to being
fiscally accretive with FiOS.

(Continue on Pg. 4)
 There may be reasons why Verizon's three approved optical networking vendors—Tellabs,
Motorola and Alcatel-Lucent—also may not be in a hurry to roll GPON out. If Verizon really is forcing
the margins down, and they are making a living on BPON, what's the hurry?

Not surprisingly, EPON and GEPON advocates are hopping on the issue, saying the GPON
chips are no good and that it is only a matter of time before EPON takes over the North American
market with China's Huawei leading the way, which is somewhat akin to every Western vendor's worst
nightmare.

GPON vendors claim the technology is going through its normal iterations prior to commercial
rollout and that the advantages over BPON, EPON, etc, make it worth the wait. However, consolidation
already is occurring and some companies are giving up. EPON proponents point to the fact that EPON
and GPON standards were both approved by FSAN in 2004 but that EPON is enjoying far greater
penetration, due in part to its easy acclimation into an Ethernet environment. “They have the standards
down, while GPON is still kicking the tires when it comes to standards, so to speak,” said one source.

There is no secret that there is anxiety regarding the chips, although several sources said
Freescale is making adequate to good chips. It has become Alcatel-Lucent's curent supplier. There is a
whole list of others burning the midnight oil to get it right, including Occam Networks (recently bought
Terawave), Mindspeed, Ikanos, Broadlight, Broadcom, PMC Sierra, Iamba, Texas Instruments, Hitachi,
etc.

But while Freescale is said to have made 40,000 workable chips, an EPON manufacturer
scoffed at the number. “We provided 40,000 chips in three days for our Japanese line,” this source told
us.

In the meantime, some companies such as Hitachi are hedging their bets, offering both an
EPON and GPON system.

We don't believe GPON will go away. There are too many good vendors working on the
problem. But its full-scale rollout is no doubt keyed to Verizon's affirmation. The question is whether
Verizon will relent a little bit on price or whether basically this is going to be a vendor community—at
least an optical components vendor community--largely screwed from the git go.

* * * * * * * * * * * * * *

I am tremendously proud of this issue because of the intellectual honesty and capability and
courage of its contributors

Founding editor Timm Bechter had as a priority the importance of always seeing both sides of
an issue, generally by getting people with opposing viewpoints to square off. Timm, in fact, finds
himself in that role this issue, advocating the dominant builder-carrier model recently dissed by the
FCC while the popular Diane Kruse applauds the Commission up to a point. Thanks to both for a
spirited debate and the courage to represent their viewpoints.
 In fact, our lead articles take contrasting points of view on passive optical networking (PON) vs.
point to point (P2P), an enduring industry topic of head-butting. Dr. Mario Rossi (formerly of Bell
Labs) and Calix executive David Russell take the PON position, while Telco Systems Irit Gillath
provides a spirited offensive for P2P. This issue will not go away and in fact will continue to take on
increased significance as the industry develops. (Remember that Cisco is a P2P advocate.)

In addition to terrific work from regular contributors James Farmer, John George and Bernhard
Deutsch (with help from Ginger Stevens), we also welcome FTTH pioneer Max Kipfer, long-time
telecom industry market research guru Mitch Shapiro, and Christy Batts of Clarksville, Tenn.'s
Department of Electricitiy. CDE is doing nothing less than redefining how a utility does FTTH. We
also would like to offer congratulations to John and Max for being elected to the FTTH Council's
Board of Directors.

One final note: We appreciated the commentary (largely positive) concerning our efforts last
issue to put the brakes on the ongoing chatter suggesting that broadband is growing far faster than it
actually is, as exemplified by the Exaflood nonsense. (We like Diane Kruse's definition of Exaflood,
which is something that will occur after Global Warming.) Such overly aggressive thinking has and is
hurting the fiber optics industry. Case in point: It is this kind of crazy hype that is not allowing the
optical components industry to gracefully consolidate and succeed, according to Avanex Chairman and
CEO Jo Major. Vendors that have not been profitable in many years can continue to go out and get
funding as the result of this type of hype, which still apparently gets the attention of financiers. As the
result, it limits everyone's potential.

Editor's Note: We don't believe that PON has definitely won the FTTH architectural battle, despite that
opinion shared by many in our industry. The second largest telco TV FTTH rollout in the United States,
SureWest, uses a point-to-point architecture and there is reason to believe that a lot of non-FSAN
carriers are at least seriously considering it. Following Dr. Mario Rossi and David Russell review the
benefits of PON while Irit Gillath develops strong arguments for point to point. We found it compelling
reading:

PON - The Technology of Choice of Service Providers Worldwide

Mario Rossi, Consultant, msrossi1@gmail.com
and
David Russell, Solutions Marketing Director, Calix, david.russell@calix.com

Summary
Passive Optical Networking (PON) is the technology of choice of service providers worldwide
deploying next generation Fiber-to-the-Home (FTTH) networks. PON technology has unique
advantages that make it ideal for transitioning from legacy to emerging IP over Ethernet based services.
Two types of PON technology are widely deployed: GE-PON and GPON.
 Two different international standards organizations developed GE-PON and GPON. The IEEE
802.3ah EFM ratifies the GE-PON (Gigabit Ethernet PON) technology and the ITU-T G.884/FSAN
ratifies the GPON (Gigabit PON) technology. Because the IEEE 802.3 committee developed GE-
PON, it is a Layer 2 standard. Service Providers developed GPON and, as a result, the standard
includes both Layer 2, as well as service and management layer standardization. The other difference
between GE-PON and GPON is that the first uses 1 Gbps symmetrical transmission rates whereas
GPON uses 2.5/1.25 Gbps asymmetrical (downstream/upstream) transmission rates. Both standards
bodies are working on next generation 10 Gbps technology. GPON supersedes an older standard,
BPON, which was not an Ethernet-based standard. Today, GPON is the preferred technology in the
US and Europe, whereas Asian service providers have favored GE-PON. PON uses three laser
wavelengths to provide services. Two wavelengths transport signals from the central office Optical
Line Terminal (OLT) towards the subscriber’s Optical Network Terminal (ONT) and one from the
subscriber to the central office (figure 1).

Video Broadcast
1550nm

Data + Voice
Optical Line  1490nm Optical 
Network 
Terminal Terminal 
(OLT) (ONT)
Data + Voice
Splitter
1310nm
Optical 
Network 
Terminal 
(ONT)

Figure 1

One wavelength (1550nm) is dedicated to RF video broadcasting, freeing the other wavelength
(1490nm) for downstream data traffic that includes IPTV, signaling and voice transport. The 1310 nm
wavelength transmitted by the subscriber back to the central office is responsible for upstream data
traffic, voice, signaling, ordering services, etc.

Having a full wavelength to broadcast RF video enables PON networks to support more video
bandwidth, enabling, for example, more HDTV content, when compared to competing technologies
such as Active Ethernet- (P2P), Hybrid Fiber Coax- (HFC) and DSL-based technologies. This
capability offers considerable advantage in markets where video is critical to offering a competitive
service (e.g. residential markets) and the service provider may have to compete with incumbents. The
video-dedicated wavelength uses the same RF technology used by cable TV companies and allows the
subscribers to use the coaxial cables in their homes, without adding any extra devices.

Some cable TV companies are starting to evaluate PON technology to upgrade their networks
and compete against telephone companies, such as Verizon and AT&T, which have been steadily
gaining market-share in video services. To date, this has primarily been in master-planned
communities where FTTH is required by the developer.
 Topology Options
When compared to P2P FTTH, the greatest strength of PON architecture is its flexibility.
Service providers have no control over the dynamics of development patterns in the areas they serve.
Network planners have to adapt as new housing and business parks are developed and change over
time. This means that deploying an architecture that can adapt to change and has many service options
is ideal.

P2P Ethernet has a fixed outside plant architecture, which either requires home running fiber
from the serving office or home running fiber from active electronics cabinets placed in the feeder
network. This places too many limitations on network planners. PON networks can be designed as
P2P networks, but offer many other choices.

Passive Vs. Active

The root of the outside plant flexibility is the passive nature of PONs. PONs makes use of
light’s ability to be split and manipulated without the need for active electronics. As a result, PONs can
deliver the full complement of bandwidth to every home and business on the network. For example, in
a GPON network every subscriber has access to 2.5 Gbps that is available in the ONT at the outside of
their home. By leveraging this ability to split light passively, PONs can significantly lower the capital
and operational costs of deployment.

Moreover, no electronics in the outside plant compounds savings in a number of areas:

1. No expensive outside plant cabinets with concrete footers, land lease issues, etc.
2. Less overall electronics required; PON OLTs serve more customers than P2P Ethernet
switches
3. Fewer fibers and less construction costs in the feeder plant
4. No ongoing powering required for outside plant electronics-network or battery/generator
backup
5. No support and maintenance required for outside plant locations
6. Less splicing and fiber cable management required

These economic differences have real impacts on the economic viability of FTTH deployments, as
evidenced by the shutdown of the Optical Entertainment Network (OEN). While a variety of reasons
are cited for the closure, one may have been the shift from PON to an active Ethernet architecture.

PON Architectural Choices

PONs allow for a variety of architectural types. By changing the location of the splitters,
service providers can even start with one model and adapt it as their needs change.

P2P from the Central Office – By co-locating splitters with the PON OLT in the serving
office, network designers can create a point-to-point architecture with the same design as a P2P
Ethernet topology. Operators who have a single serving office and have sufficient fiber capacity in
their feeder plant to have P2P architecture use this design. This design is popular because it allows for
100 percent utilization of OLT ports.
 P2P from an Outside Plant Cabinet – Most PON OLTs utilize hardened electronics and built-
in SONET or GE transport, this enables placement of PON OLT and splitters in feeder-distribution
cabinets. Service providers who have existing DLC cabinet locations and want to add FTTP to those
cabinets frequently use this architecture.

Local Convergence Point Architecture – This is the most commonly used PON architecture.
PON OLTs are placed in the serving office and splitters are placed in passive cross-connect pedestals
located in the feeder-distribution plant. This is a very flexible architecture for areas of new
construction where the service provider cannot predict where houses will be built and in what order.
This architecture also enables 100% utilization of PON OLT ports.

Distributed split architecture (also called dual-star topology) – This is the least commonly
used architecture, but for the right application can be very cost effective. Two or more splitter points
are placed between the OLT and the ONT. This architecture can strand OLT port capacity if not
designed properly. This approach uses the least amount of fiber and is ideal for serving video cameras
along a freeway or serving houses lining a rural road. Distributing splitters allows for using PONs
efficiently in sparsely occupied regions or low take rates areas. For example, installing four 1x8
splitters in the field and one 1x4 splitter in the central office (CO) may enable a service provider to
utilize most, if not all, of the OLT’s capacity. As the customer base grows, this Service Provider can
add another OLT in the CO and replace a couple of the splitters in the field taking the older ones to
another area with a lower take-rate.

Services and Revenue Generation Options

Transitioning Networks
As exciting as FTTH is for consumers, the technology has to make sense from a business
perspective for service providers. Today, most service providers are somewhere in a transition
between a legacy network and where they hope to be in the future. The legacy networks are generally
copper-based, offering a mixture of TDM, ATM and other legacy services. No one disputes the world
is moving to a fiber-based, IP-over-Ethernet future. The question is not if, but how and when service
providers will migrate their networks. PONs provide a seamless transition from legacy services to IP-
over-Ethernet-based services.

Slower Transition Faster Transition

Figure 2
Many PON OLTs support both copper- and PON based-services. This allows a service provider
with already deployed copper service, such as DSL, to upgrade customers to PON in an orderly and
economical fashion.
 Service Migration
Twelve years ago, seven of the world’s largest telephone companies came together to develop
the first standards for FTTH technology. The Full Service Access Network Group (FSAN) grew over
time to encompass 24 network operators with over 450 million subscribers. That means many
subscribers that need to be transitioned from the services they get today to the services they will want
in the future.

With this in mind, the FSAN members were particularly careful to develop a standard that
accommodates both legacy and emerging IP-over-Ethernet services. This explains why most service
providers in the United States are adopting GPON over P2P Ethernet for next-generation technology.

The FSAN committee recognized that they needed to define a standard that supported both
TDM voice as well as VOIP. Today, service providers can provision GPON ONTs for one or the other
service, or even both services simultaneously. Service providers can then turn up VOIP services when
it makes sense for their operational and support logistics, while still maintaining the revenue stream
coming from legacy voice.

The support of TDM-based services also means that service providers deploying GPON can
offer traditional T1 services. Serving business customers is a primary revenue source for service
providers deploying Fiber-to-the-Premise (FTTP), a term that encompasses service to the home,
apartment, condominium and businesses. Today, businesses require a mix of T1 and Ethernet-based
services. Both services continue to grow rapidly and while Ethernet-based services are the future of
business services, capturing revenues from T1 services are critical to any fiber-based business plan. As
Ethernet-based services mature, many of these customers shift to pseudo wire. PON designs are
becoming available that support pseudo wire; unfortunately, there are business applications, such as
cell-site backhaul to CDMA base stations, that still require the low latency and jitter offered by a
traditional T1 circuit.

The business case for deploying FTTH to residences requires significant numbers of customers
to take the triple play of voice, video and data services. Voice and data are the greater contributors to
margins. Video is not a great margin business, but drives revenues and also is critical for marketing to
and retaining customers. In most cases, the desire to deploy ultra high-speed Internet and video
services drive FTTH deployments.

While IPTV services are successfully being deployed over DSL-based networks, the advent of
HDTV and the need for very large bandwidths for video are driving service providers to transition from
copper to fiber. Here PON offers significant advantages over P2P Ethernet by again offering service
providers flexibility and choice.

PONs are designed to enable both RF video and IPTV services. As described above, the RF
video uses the 1550 nm wavelength and supports the channel lineups and channel spacing used by the
cable TV industry. This has proven to be a critical advantage for service providers over the past few
years. While the situation is improving, accessing content and deploying IPTV services has remained
more challenging than deploying RF video. Another problem faced by IPTV deployments is wiring the
house to reach all the IPTV set tops. Since most U.S. homes already have coaxial cabling, RF video is
easily distributed around the house.
 Many PON-based service providers are deploying RF video for their low tier service, and IPTV
for their high-tier, interactive services. Because IPTV services always require a set top, this two-tier
approach avoids the expense of set tops for low-end video customers. This creates a smooth transition
from legacy to advanced IP services, all provided over the same PON ONT.

Coaxial Cabling

Cat 3 Cabling

ONT

Cat 5e Patch Cord to Wireless Router‐Switch

Figure 3
The diagram depicts a sample household with three HDTV sets, three telephones
and a wireless router-switch serving two computers. The network uses existing
home cabling that minimizes overall installation costs.

PON and the Subscribers – High Bandwidth Reaches the Doorsteps, Now What?

Service Providers evaluate in great detail how much it will cost to enable the high-bandwidth
services they are bringing to a subscriber’s house but also need to evaluate how much it will cost to
distribute this content throughout the house. That can be a deal breaker if the Service Providers
transfer this cost to the subscriber; and they typically provide the ihome wiring installation for a
nominal fee.

For the subscriber, PON technology is advantageous since it enables the use of the existing
wiring in the home without disruptions. The existing coaxial cable carries the RF video, while the
Cat-3 twisted pair cabling connects standard telephone sets to the OLT POTS ports. If the house has
categorized unshielded twisted pair (UTP) cabling then data traffic connects directly from the ONT’s
Ethernet port to the in-home data wiring. If the home is not pre-wired for high-speed data, the service
provider runs a UTP cable from the ONT Ethernet port to a wireless 802.11 router/switch (Diagram 2)
that retail in the range of $25-50 dollars.

The previous example demonstrates how PONs are flexible enough to allow for service
providers connecting and providing content to subscribers without having to install extra electronics or
cabling in their premises.
 Other ways to distribute content within a house allow for added features, but will cost more
because of the electronics required or additional cabling to do the job. The service provider should
consider the benefits of each solution since they can affect installation costs. The easiest way to
prepare a house for the high-bandwidth provided by PON is during its construction or a major
remodeling where a contractor can install new cabling complying with TIA 570-A standard. In the
case of existing homes that require wiring for data services and video services, it may be less expensive
to use one of the home networking standards, such as MOCA (Multimedia over Coaxial Alliance -
http://www.mocalliance.org/en/index.asp) or HomePNA (http://www.homepna.org/en/index.asp).
Many PON vendors have included these technologies in their ONTs, as have IPTV set top vendors.

PON Standards are Comprehensive in Scope

PON standards define much more than just the physical and transport layers. Because the
largest telephone companies in the world defined GPON, they took special care to define standards for
dynamically managing bandwidth, quality of service, forward error correction and AES encryption. As
a result, PON is a robust and very secure technology. Vendors have incorporated functionality in the
ONT to ensure that upstream traffic that might be malicious is discarded. Because the PON ONT is a
trusted network element-this enables PONs to foil denial-of-service attacks and ensures that hackers
cannot spoof their way on to the network.

Subscribers need to keep in mind that security threats come mostly through the upper
layers of the OSI model (http://en.wikipedia.org/wiki/OSI_model) via spyware, viruses and other
methods since this allows for the hackers to probing many more users.

(Continue on Pg. 13)

 On another security layer, PON technology is designed for lifeline quality service with five 9’s
reliability. PONs support E911 services and Communications Assistance for Law Enforcement Act
(CALEA). By having a TDM capability as part of the standard, GPON can support customers not
interested in IP phones, such as seniors and persons living in places with poor mobile phones coverage,
who cannot use their mobile phones as a landline backup.

Economics of the Business Case

The financial success of FTTH is dependent on a model that expands the revenue-generating
ability of the network, while at the same time lowering the cost of operating the network. Service
providers who have gotten this equation right have seen excellent financial performance and are
generating positive cash flow ahead of their own projections. Operators who have gotten the equation
wrong, like Optical Entertainment Network, have had to sell or have shut down entirely.

A critical concept in understanding the economics of FTTH networks is to distinguish between

the cost to pass a home and the cost to serve a home. Ideally, service providers want to pass as many
homes as possible for the lowest possible cost. In doing so, they preserve their budget and spend it on
“success-based” capital - meaning costs the service provider incurs when the customer has agreed to
order the service.

PON networks have large cost advantages over P2P Ethernet when it comes to lowering the cost
to pass a home, meaning more of the capital budget shifts to success-based spending. In Burlington,
Vt.’s GPON FTTH network, which was recently featured in a case study published by the Minneapolis-
based Institute for Self Reliance, the cost to pass a home with GPON is $250 per home. This is
roughly equivalent to the cost of HFC systems, which have always had an advantage in cost per home
passed over traditional telephone networks.

Once a customer signs up for the service, the cost for PON electronics and the fiber line to the
house runs anywhere from $900-1500, depending on density and whether the outside plant is aerial or
buried.

Some very-hard-to-reach locations can run even higher, depending on the civil construction
involved. While the success-based cost for the P2P fiber cable will be the same, PON systems tend to
have lower costs for electronics on a per-subscriber basis, particularly at the central office. PON OLTs
share one laser over 32 subscribers. P2P Ethernet switches require 32 lasers, typically packaged in SFP
pluggable modules. These can be very expensive, particularly when hardened. If the Ethernet switch
is not hardened then the P2P architecture also has to bear the cost of outside plant cabinets with
environmental controls for both heat and cold. The corresponding costs can lead P2P systems to cost
as much as $3,000 per home - over twice the capital costs of PON-based systems. Added to these
extra costs, P2P network owners also have the costs of disposing old batteries (waste hazard, non eco-
friendly materials, etc.), spare batteries and emergency power equipment such as generators and the
crews and trucks to roll them to a remote cabinet in case of a power failure. The ongoing cost for
electrical power on an annual basis is also considerable. The list of spare parts for P2P networks grow
with switches and other electronics installed in the field that have a mean-time-between-failures
(MTBF) much lower than the passive optical splitters that PON use, which have an almost unlimited
life-span.
 When added up, the operating costs for an active versus a passive plant can be significant.
Verizon and many other service providers that operate PON networks have reported significant
operating savings on their lines using PONs. Reports have indicated annual savings of 75 percent and
more on the average cost to maintain a PON FTTH line versus a copper line with active electronics.
While the operating differences between PON and P2P will not be as great, the elimination of
powering, servicing and upkeep of outside-plant cabinets and electronics is significant.

Add to this the cost for service providers to rewire homes for video services. Many PON
operators achieve savings inside the home by offering RF video, a service PON can deliver that P2P
Ethernet cannot. Again, PON gives service providers choice and flexibility that P2P Ethernet does not.

Conclusion

The successful Service Providers of the future must be able to support and deliver a broad range
of services. They have to deploy a network that is flexible, scalable and economically viable since
their customers will always require new features at competitive costs.

The Internet and the IP technology underpinning it have proven to be very robust given the rate
of growth. However, some experts are calling attention to the need to add more bandwidth at the
Internet’s core to accommodate all the growth in “information consuming” at the edge.

The demand for high-bandwidth services is increasing as consumers adopt new technologies
and bandwidth-intensive applications. Due to their diversity, these networking technologies and
applications require both convergence and openness.

That means a network must be able to provide seamlessly:

1. Video – both in RF and IP formats since consumers use both technologies

2. Data – able to provide high-bandwidth reliably and economically
3. Voice – both Voice over IP (VoIP) and TDM-based Plain Old Telephone Service (POTS)
since consumers have different needs and requirements; but all require E911
4. Management and monitoring of increasingly complex systems in the home – always-on
security, energy management devices, etc.
5. Wireless/wireline network convergence – consumers are able to hand-off between a
wireless network and their in-home broadband residential network
6. Low operating and upgrading/scalability costs – competition will continue to be high and
the network must support a broad range of new services
7. Reliability – due to the nature and diversity of services running on that network the
downtime, if any, must be near to zero, meeting carrier class standards. That means as few
active components as possible since their use means exponential increase in the network’s
potential downtime.
8. Low environmental impact – the network operation shall have minimum impact by
minimizing power consumption and hazardous waste such as batteries, old/defective
electronics, etc.
9. Standards-based – the equipment and services should follow international standards
10. Security – the network shall provide robust security tools and safeguards to guarantee its
users privacy
 PON meets these requirements while also being the most cost effective architecture. That is
why it is the technology of choice of service providers worldwide.

Mario Rossi, MBA, BS Mechanical Engineer, BICSI RCDD, has extensive international experience in
fiber networks. Currently working on software development at Microsoft, he designed products and
systems encompassing central office, outside plant and enterprise solutions for customers worldwide
and is published extensively.. Professional experience includes working at Lucent Technologies, OFS
and Alcatel.

David Russell has over 20 years of experience in business development and marketing of broadband
access systems. Mr. Russell came to Calix through the acquisition of Optical Solutions, a leading
FTTH supplier, where he was responsible for strategy and business partnerships. Prior to joining
Optical Solutions in 2003, Mr. Russell worked in ADC’s CMTS Group, now owned by Big Band
Networks. He began his career at ADC working on optical connectivity and broadband access systems.

Future-Proofing Your Network: The Case for Active Ethernet

By Irit Gillath, Vice President of IP Product Line Management, Telco Systems
The two predominant technologies in the FTTH market are Active Ethernet networks and
Passive Optical Networks (PON). While both technologies use an all-fiber infrastructure (see
FTTP/H/B in Figure 1), there are some significant differences.
APON and BPON are ATM-based solutions with APON serving primarily business
applications and deployed by Verizon and SBC. BPON provides enhancements to APON with overlay
capabilities for delivering services like broadband video.
GPON uses a different, faster approach (up to 2.5 Gbps in current products), encapsulating
traffic in a version of the SONET-compatible Generic Framing Protocol (GFP).
 Figure 1 - Common Wireline Access Architectures

CO RT Node Curb Customer

(Neighborhood
Premise
DSLAM or DSL
Current DSLAM / ATM from DLC DSL modem
telco aggregation Up to 3660 meter of copper ADSL

Up to 1500 meter of copper DSL modem

Aggregation DSLAM
FTTN (100s HH) + set-top +
ADSL2+ / VDSL2

Aggregation 150m DSL modem

Aggregation ONU (6 - 12 HH)
FTTC + set-top +
VDSL2
ADSL2+

OLT Active Ethernet, BPON / GPON + optional RF overlay ONT + set-

FTTP/H/B Passive optical splitters / combiners in outside plant top + home

Fiber node Coax (6MHz DS channels,

Headend (500 - 1500 N*6MHz in future) Cable
HFC
CMTS / QAM Fiber Node Modem +
(50 HH)
Today ~700- 800 MHz DS / ~40 MHz US
Fiber Copper In the future ~900MHz DS / ~80MHz US

Fiber is the key enabling media to support the higher bandwidth and new services, such as VoIP and
VOD. Fiber enables providers to improve their competitiveness, achieve operating efficiencies, address
the demand for faster Internet service, and pursue new revenue-generating opportunities such as
digital video.

EPON, or Ethernet PON, is a technique that uses Ethernet as the main transmission method for
the PON. EPON runs at gigabit rates and has its own standardization process underway at the Institute
of Electrical and Electronics Engineers Inc. (IEEE).
Active Ethernet is a technique that uses Ethernet as the main transmission method over fiber
using active components based on approved and widely used IEEE standards. Active Ethernet runs
gigabit rates.
Active Ethernet topology is flexible and can be built in any network configuration - ring, star
and tree. This allows the service provider to adjust the network structure to the deployment budget, the
topological structure of the served area and the redundancy and quality of service the service provider
wishes to provide.
In most cases one of the three following architectures will be selected:
Option A is a simple star with a “home-run” fiber connection between the house and the CO,
this obviously “wastes” more fiber, but requires no outside-plant work and no active component in the
field. It also gives a dedicated line per user which provides an independent service for each user. So if
one fiber is cut, only one household is affected.
 Option B is a tree/star topology where multiple households (typically in the range of 400-1000)
are connected to one aggregation cabinet, which will include multiple aggregation switches. This not
only aggregates the bandwidth for better utilization of the fiber and the available bandwidth but also
allows for quality of service mechanisms like traffic shaping
Option C shows an enhancement of Option B where the aggregation switches are connected in a
ring topology which increases the resiliency of the service. Even if a fiber is cut or if a switch failis, the
service for the vast majority of the services area will remain uninterrupted. Ethernet-based protocols
allow for sub 50ms recovery time making voice and video services recovery completely unnoticeable
to the end-user
So the choice becomes, what technology will best propel companies to maximize profits and
meet future consumer demands?
PON has significant limitations. PON technology provides shared bandwidth, so the amount of
bandwidth varies widely depending on the type of PON (EPON, APON, BPON, or GPON) and the
type of splitters that are deployed (1:16, 1:32, or 1:64). Even in an optimal deployment, the highest
bandwidth of a PON network is approximately 60Mbps, and in many cases will be lower. This
bandwidth is available to all customers connected to the same splitter and there is no way to allocate
higher bandwidth to one customer
over another. This presents a major 1000
obstacle to offering higher revenue 900
800
generating services for customers
700
who request it. Uplink (Mbps)
600
Downlink (Mbps)
An Active Ethernet architecture is 500
based on active components in the 400
central office and at the customer 300
200
premises with an option to aggregate
100
a group of houses in the outside plant 0
using active switches. This
Ac

AP
Ac

GP

GP

EP

EP
technology can offer above 1 Gbps
tiv
tiv

ON
ON

ON
ON

ON
eE
eE

now, and more as the need arises. It

1:3

1:6

1:3
1 :3

1 :6
10
1G

4
2

is typically price competitive to 2

0M
igE

PON, but allows for the most future

growth.
 Active Ethernet or PON?
While many FTTH deployments today are based on PON, the advantages of point-to-point
active Ethernet are well documented, not only from a standards-based view, but also in cost savings,
flexibility, and ability to address the bandwidth needs of tomorrow..

Active Ethernet components are 802.3 compliant and offer full interoperability with other
standards-based solutions from a broad base of vendors. This means the provider is not locked into a
single vendor’s solution. From a deployment perspective, this enables the provider to choose the best,
most cost effective products from a wide range of vendors to construct the network.
Since Ethernet is standards-based and is used universally, the cost of Ethernet-based products
and components is decreasing rapidly. Residential and small business subscribers with FTTH service
will benefit from the cost savings as service providers continue to deploy additional equipment. Active
Ethernet offers subscribers dedicated bandwidth based on their individual requirements. This means
that the provider can capitalize on revenue-generating services for both residential and business
customers in the same area. In a PON network, the available bandwidth per subscriber terminal is
shared with all other subscribers in that area without being able to give different service levels.
New subscribers can be easily added to an Active Ethernet network within the 100 km
geographic area at minimal cost. This pay-as-you grow philosophy means that providers are not
making capital investments in their network until they have the subscribers to offset it. Customers can
be added one at a time as needed. PON networks require one splitter for every 32 ONUs within the
20km service area and are cost-effective only when at least 26 ONUs are connected.

There’s little doubt that the bandwidth and traffic engineering features of active Ethernet make
it a perfect solution for Triple Play applications. Many Active Ethernet solutions support integrated
VoIP protocols - H.323, SIP, and MGCP, as well as various standards-based mechanisms, which ensure
the voice, video, and data quality of service - IEEE802.1p, voice priority, and TOS marking, and IGMP
and RF to support video.
PONs require detailed pre-planning to place splitters in the ideal location from the OLT that
support the maximum number of subscribers and ensure that the ONU reaching the maximum number
of subscribers. On the other hand, Active Ethernet is easy to configure in either a mesh or ring topology
using a standards-based RSTP and MSTP based solution to ensure the shortest convergence time in the
industry (less than 50mSec). To keep costs low, services can be delivered using a single strand of fiber
or over CWDM using up to 16 unique wavelengths over fiber pairs. This ability to provide dedicated
bandwidth over a single fiber or fiber pair to multiple subscribers maximizes fiber capacity.

In looking at both Active Ethernet and PON, the choice between the two is becoming an easy
management decision as well. Because traffic is pure IP, no transition between protocols is required.
Network management of all network elements is very simple using off-the-shelf products and
standards. Active Ethernet can also send information to the NOC using SNMP, EFM OAM, MPLS, or
any number of management protocols used by the provider. Passive splitters cannot transmit data so
troubleshooting a problem requires dispatching a technician to the site – or multiple sites.
 Active Ethernet - Meeting the Demands of Tomorrow’s Network
For providers who want to future-proof their network, who want to offer a wide range of service
options, and who want to maximize their ROI, Active Ethernet is the logical choice for today’s Web 2.0
world.
Editor's Note: The Federal Communications Commission recently ruled against cable TV companies
that already have exclusive carriage arrangements with the owners of multiple dwelling units (MDUs),
saying such relationships hurt competition. Following Diane Kruse and Timm Bechter have contrasting
views on whether that is a good thing. Kruse later shifts gears concerning whether such arrangements
are good for developers and FTTH providers.

Weighing in on Exclusive Contracts for Multiple Dwelling Units

By Diane Kruse, CEO, Zoomy Communications,

On November 13, 2007, the Federal Communications Commission (FCC) issued a Report and
Order (Order) and Further Notice of Proposed Rulemaking declaring null and void any "building
exclusivity" clauses between franchised cable operators and owners of multiple dwelling units (MDUs)
and other centrally managed real estate developments. In this article, I will weigh in on the effects of
the FCC’s rule making, and discuss the FCC’s request for more input on extending the rule making to
other video providers and further restraints on exclusive contracts.

Before I comment on the FCC’s position, I will provide a detailed description and background
information on the Order and Proposed Rulemaking. This was eloquently summarized by the law firm,
Latham and Watkins; and therefore, I will provide the attorney’s summary herein; and then provide
comment on the summary. I am not a lawyer, and as a “broad-brush gal” – (read “big picture”), I rely
on others to give me the details. So as I do not get sued for plagiarism – the following description is
verbatim from Latham and Watkins on the Order and Proposed Rulemaking.

Background Information – Why the FCC Stepped In

In its 2003 Inside Wiring Order, the FCC chose to permit the very types of exclusive arrangements
between cable providers and MDUs that the FCC's new Order declares null and void. The FCC
explained that its about-face is justified by recent developments that were not part of the record in
2003. In particular, telecommunications carriers have made significant efforts to compete in the video
market, and they assert that exclusive access arrangements constitute a barrier to entry. Exclusivity
clauses bar competition not only for video services, but, these providers assert, also for the
increasingly popular "Triple Play" of voice, video and broadband Internet access service. Competition,
the FCC reasons, will likely result in lower prices, more channels, and greater diversity of information
and entertainment. In contrast, exclusivity clauses deny these benefits to consumers, according to the
FCC.

The FCC was not persuaded by claims regarding the pro-competitive effects of building exclusivity
clauses. For example, some building owners and cable operators argued that exclusivity often is
necessary to ensure that a building is wired for service. Similarly, commenters claimed that MDUs can
induce MVPDs to bid to provide the best mix of low price, service quality and choice, and facilities
improvements in return for building exclusivity. However, the FCC found that, on balance, MVPD
competition for individual residents within the MDU—rather than MVPD competition to exclusively
serve the entire MDU—could better achieve those consumer benefits. The FCC also reasoned that
MDU owners could not be counted on to enter exclusivity agreements consistent with the interests of
the MDU tenants, either because the MDU owners' interests differed from those of the tenants or
because the MDU owner failed to understand the implications of granting building exclusivity.
 Let me chime in – it is true from my experience that in some cases, MDU owners’ interests differ
from their tenants; especially if there is a big check given to the MDU owner for the ability to serve
their property exclusively. The revenue share that MDU owners have come to demand is now between
$200-$1,000 one time per door, or a significant ongoing monthly revenue share based upon the
percentage of monthly revenues generated by the service provider, for an exclusive contract and bulk
billing through the HOA, or Homeowner’s Association. I would have to argue that the monetary
incentive for MDU owners proposed by the cable, telecommunications, Triple Play and fiber optic
amenity companies indeed sway the interests of the MDU owner. This practice of paying off the MDU
owner seems to be more anti-competitive than exclusive access or bulk billing. It seems as though if
the FCC banned this practice, the playing field would level, and the consumer would better benefit.

I will also say that most experienced developers and MDU owners realize the potential impact of
being paid by the service providers for exclusive access. They realize that the tenants of their
communities may not agree with this practice, and therefore, many developers that we work with at
Zoomy do not want a big check; but rather, want Fiber to the Home and a robust triple play and
technology package from their providers.

The Order Applies to Cable Operators

The prohibition on exclusive access provisions applies to franchised cable operators, including
telecommunications carriers that provide MVPD service. Private cable operators (PCOs) (also known
as Satellite Master Antenna Television (SMATV) providers) and Direct Broadcast Satellite (DBS)
providers are not covered by the prohibition. The FNPRM seeks comment on whether to impose the
exclusivity ban on these and other MVPDs not covered by the Order.

I believe that the costs for providing video services to the MDU or to a single family home
development are still a barrier to entry for new entrants. Without the reassurance of high take rates (i.e.
exclusive access and bulk billing for video services) the costs for installing a new video headend are
cost prohibitive. Many new entrants install a mini-headend to each new MDU or “other centrally
managed real estate development” in order to offer video services to the development. The cable
operators have the advantage of extending their centralized video headend to serve the property. The
costs for installing a mini-headend are between $30,000 - $100,000; depending upon capabilities; and
$250,000 - $500,000 for an IPTV- based headend.

The large cable operators also continue to have an unfair advantage with their access to
programming, and pricing leverage with many channels that provided programming. If video content
firms want carriage to the home, they have to negotiate a deal with the cable guys.

I think the FCC was right in banning exclusive access for the cable companies; however, in order to
promote competition, and better yet, to promote FTTH, the FCC should not extend this ban to fiber
optic amenity companies, Private Cable Operators, SMATV providers or Direct Broadcast Satellite
providers.

So how does this promote Fiber to the Home? The telecommunication companies, Triple Play
providers and fiber optic amenity companies are indeed gaining traction by building FTTH networks.
Developers and MDU owners want FTTH to future-proof their properties, knowing there is an extreme
need for speed and bandwidth. (Let’s agree not to say “exaflood” – the developers and MDU owners,
and ultimately the consumers, have no idea what the exaflood is; and may more likely link that term
with Global Warming, rather than the need for bandwidth. I digress….). As I have discussed before,
building a FTTH network is capital intensive; and again, without the reassurance of high take rates for
services (again read “exclusive contracts and bulk billing”); it is difficult to justify the costs of building
a next-generation, advanced network. The cable guys are not building FTTH; their capital costs are
lower than the new entrants that are building advanced networks. I think the FCC is on-target for
prohibiting the cable guys from entering into an exclusive agreement, but I do not think the FCC should
extend this ruling to the other PCO’s, SMATV, DBS or companies that are building FTTH networks.
We want better networks, right?

FCC Defines "MDU" Broadly for the Purposes of the Exclusivity Ban

The Order defines "MDU" quite broadly for the purposes of this prohibition. The term "MDU"
includes not only apartment, cooperatives and condominium buildings, but also "other centrally
managed real estate developments," such as gated communities, mobile home parks and garden
apartments. The key feature of all of these buildings and developments is that the residents share some
common spaces requiring central management. The FCC excluded from the definition of MDU certain
types of housing/institutions characterized by short-term residency or institutional living, such as time
share units, academic campuses and dormitories, military bases, hotels, prisons and hospitals. The
Order does not prohibit exclusivity clauses in those settings.

One distinction that the FCC excluded from their definition is single-family homes that are not gated,
but are still controlled by a “centrally managed real estate development.” I think the FCC should ban
cable companies from exclusive contracts with these developments as well. In order to promote FTTH,
better video services and pricing to consumers, the FCC needs to realize the economics of the new
entrants into the market. The economics don’t work for new developments to have many providers;
especially if the network is FTTH. The cable companies have enjoyed a monopoly for years; let the
other guys play! FTTH providers need to have higher take rates for the buildout to be feasible. I’ve
been on this soap box before; the FCC needs a good lesson in the capital cost of building an FTTH
network. I am not saying the FCC people are uneducated. I am saying that we, as industry folks
promoting advanced networks, need to provide the FCC with information on the economic model, and
the reality of the capital costs and need for revenue to justify the capital expense. I personally, along
with other members of the FTTH Council, will be involved in providing this information to our
lawmakers.

Certain "Exclusive" Arrangements Continue to Be Permitted, But Are Subject to Further Notice

The FCC's Order prohibits only what the FCC terms "the most exclusionary exclusivity clauses,"
which "prohibit any other MVPD from any access whatsoever to the premises of the MDU building or
real estate development." The FCC expressly recognized as beyond the scope of the Order: (i) "wire
exclusivity" clauses, which are not a complete ban on access to the MDU by other MVPDs, but
prohibit them from using existing wires; (ii) "marketing exclusivity" clauses, which prohibit the MDU
owner from marketing the services of another MVPD; and (iii) "bulk billing arrangements," pursuant to
which residents may receive a discounted bulk billing rate, but may be required to continue to pay that
bulk billing rate even if the resident chooses to take service from (and pay the subscription fee of) a
different MVPD. The FCC asks for comment on these three types of arrangements in the further notice.
In addition, while the Order prohibits exclusive access clauses in contracts, nothing in the Order
requires MDU owners to allow new MVPDs to access the MDU.
 Belts and Suspenders. These are ways to ensure the network feasibility is “held up”. These belts
and suspenders are important to providers to cost justify the capital expense, especially in capital
intensive Fiber to the Home builds. I’m all for belts and suspenders. If there is a chance to have
competition within a development, and there is potential for erosion of take rates, then the providers
need some other assurances; these come in the form of marketing exclusivity, bulk billing
arrangements, and – “you can’t use my wires”.

Order Likely to be Appealed

It is expected that the cable industry and possibly others will appeal the FCC's Order. The Order is
likely to be challenged on at least two grounds: (1) the FCC's legal authority to prohibit exclusivity
clauses; and (2) the FCC's legal authority to abrogate those clauses in current contracts.

The FCC primarily relied on its authority under Section 628(b) of the Communications Act. Section
628(b) prohibits "unfair" practices that "hinder . . . any [MVPD] from providing satellite cable
programming or satellite broadcast programming to subscribers or consumers." Opponents of the
prohibition argued that Section 628 prohibits only exclusive programming contracts between cable
operators and vertically integrated programming providers, to ensure that MVPDs have access to
programming. The FCC's reliance on Section 628 to limit MVPDs' ability to contract with MDUs
arguably is outside the scope of the statute. Notably, the FCC does not rely solely on Section 628(b),
but seeks to bolster its statutory authority by referring to its "ancillary" power to regulate cable
operators under Titles I and III of the Communications Act.

Second, the FCC may have overstepped its legal authority by abrogating existing contracts. Related
to that, those appealing the Order might claim that the prohibition on exclusivity constitutes a
regulatory taking. The Order counters this concern by asserting that the public has long been on notice
that exclusivity clauses were under scrutiny, and might be found unlawful. In his separate statement,
Commissioner McDowell voiced his concerns regarding abrogating current contracts as follows: "My
concern is this: after unanimously inviting cable companies and building owners to strike such
[exclusive] deals in [the 2003 Inside Wiring Order], the FCC may now be abrogating those exact same
agreements immediately rather than waiting for them to expire and without providing a grace period."
Commissioner McDowell also recognized that arguments that the FCC's action constitutes a regulatory
taking "may also have merit."

The prohibition on exclusive access arrangements, which represents a significant shift in FCC
policy, was strongly opposed by the cable industry. Many cable industry proponents are stating the fall
or even more strongly stated - the death - of the cable industry will result from this order. Give me a
break. The cable companies have enjoyed a monopoly for decades! Rather than have their revenues
handed over on a silver platter, they ought to wake up, quit their griping, and start offering better
services, better customer service, better networks, and better pricing. Their stock prices are down, yes,
this is true. Comcast and Time Warner’s shares are down 20 percent; and Charter’s stock has
plummeted 60 percent. Many cable industry proponents say the stock has tumbled because of the FCC
orders. Consequently, Verizon’s stock has increased 20 percent since introducing the FiOS concept;
their revenues, number of customers, and their video subscribers have increased dramatically over the
past two years. The satellite TV companies offer more high definition channels, which has become
attractive to many consumers, and DirecTV’s stock has bumped up 20%. The cable companies need to
realize that the FCC does not solely control their destiny; and that they too have power to
influence…….. by simply being better.
A Government Operating Under a Faulty Argument for Competition
Timm P. Bechter, CFA, Investment Analyst, Network Equipment and Services, Waddell & Reed

FCC Chairman Kevin Martin and his fellow commissioners recently voted that personal property
rights should take a back seat to the telecom lobby’s wishes to immediately break the cable MSOs’ so-
called grip on the MDU pay-TV market. Never mind that the telcos have had every opportunity to
compete for exclusive arrangements with MDU property owners for TV services without any
regulatory ban on exclusive arrangements and never mind that the FCC’s argument that the retroactive
ban on exclusive contracts for MDUs is built on a flawed argument that competition necessarily means
lower prices over time.

Chairman Martin, in his statement accompanying the ruling, asserts that the ruling is pro-
competition. He goes on to use the rise in cable TV prices to imply that cable TV has little or no
competition. His statement is wrong on both counts. First, there has been competition in pay-TV for
years between the two satellite players and cable. In some cities there are even two cable companies
competing. Moreover, in the case of MDUs, telcos have always had the right to compete with cable
TV for exclusive contracts. Second, the rise in cable TV prices has not been due to a lack of
competition.
That the telcos have always been able to compete for exclusive contracts with MDUs deserves
emphasis. Can’t they compete on the merits of their service offering alone? If theirs is truly a superior
TV service, wouldn’t an MDU owner negotiate his or her way out of the exclusive contract he or she
has with the cableco to up the value of their rental property and provide their tenants with the better TV
service at a better value? Wouldn’t any new MDUs being built be the easy pickings for the telcos now
that they have the ability to get their own exclusive agreements with the property owner?

A little history is in order here, and a little is all that is needed. Before the cablecos started taking
the telcos’ landline telephony subscribers with their VoIP service, the largest telcos had no interest in
providing television services and made no lobbying effort to have the FCC rule on exclusive contracts
for MDU television services. Point being that the former telco monopolies dragged their feet
successfully to kill the FCC’s pro-competitive UNE-P regime (when they themselves were threatened
by what they saw as a regulatory taking of their facilities) now have no ethical dilemma in having their
lobbyists push for a regulatory taking of a different sort in hopes of taking business from the cablecos
in the MDU arena.

What about Chairman Martin’s implication that the rise in cable TV prices in recent years reflects a
lack of competition? This is misleading at best and bad logic at worst. Even if cable rates went up 93
percent from 1995 to 2005 the reason was not a lack of competition but the rise in the cost of content.
Additional evidence of the Chairman’s poor logic is his comparison of cable TV prices to those of long
distance and wireless telephony prices over this same period. This comparison is simply irresponsible
– both of these commodity markets have seen pricing go down due to their excess capacity. With
ample extra capacity for traffic on our nation’s long-distance and wireless networks, it is hardly
surprising that prices declined as service providers competed for the incremental customer to contribute
revenues to their high fixed cost networks.

A more valid comparison would be of cable TV prices with those of satellite TV. Content
programming costs have grown to cable and satellite (and now telcos feel the pain too) at no less than
low-double digit rates over the same period! While cable doesn’t
enjoy the natural economies of scale satellite does, both types of
pay TV providers, cable and satellite, have had to pass on the
content provider's relentless increases in prices in order to remain
profitable. [Satellite has been able to keep their prices down
somewhat as they took pay-TV market share and customers from
cable thanks to the natural scale advantages of satellite broadcast
technology.]

Excess capacity, such as that for long-distance and cellular

telephony, simply does not exist for pay-TV services. Telcos are
spending hand-over-fist on new last mile infrastructure to provide
pay-TV services while the satellite providers are launching new
satellites and the cablecos are upgrading their plants and moving
rapidly toward DOCSIS 3.0 technology.

The bottom line is that competition hasn’t brought down

pricing on the cableco side of the pay-TV market because of the
cable MSOs’ relative lack of pricing power relative to the content
providers (a shared problem with satellite and telco). Please note
that when it comes to TV service, the telcos have the same scale
problem as the cablecos, and they are third (from a platform
perspective) or fourth or fifth (from a competitor standpoint) to
the pay TV party. More competition in the last mile probably
increases the relative power of the content providers!

Now I am for healthy competition, let me be clear on that

point. I am for market-set pricing on facilities-based services
where natural advantages and disadvantages of the infrastructure
and service offering and customer service determine market
share. What I am not for is taking away someone’s property
rights in order to promote competition under false premises, not
to mention a not-so-subtle anti-cable agenda.

So, in his arguments about competition and prices, the FCC

Chairman has gone too far to justify the ruling. I believe the
courts will likely overturn the FCC’s ruling on the basis that it has
overreached its regulatory authority. But there is an even better
reason for overturning the ruling: it constitutes an
unconstitutional governmental taking of private property.

Amendment IV (part of the Bill of Rights) to the U.S.

Constitution: “The right of the people to be secure in their
persons, houses, papers, and effects, against unreasonable
searches and seizures, shall not be violated, and no warrants shall
issue, but upon probable cause, supported by oath or affirmation,
and particularly describing the place to be searched, and the
persons or things to be seized.”
 Chairman Martin makes the following claim in his written opinion: “All consumers, regardless of
where they live, should enjoy the benefits of competition in the video marketplace.” The statement
seems innocent enough, helping consumers in their pursuit of happiness and all that. But there is just
one minor problem . . . Let’s look at another statement, a hypothetical one that is similarly innocent
and on the surface all about the pursuit of happiness.

“All citizens, regardless of where they live, should enjoy the peace of knowing that they will not be
killed by a handgun. Therefore the (pick an agency, ATF?) hereby bans handguns.” Why doesn’t this
happen? Because of the Second Amendment to the Constitution: “A well regulated militia, being
necessary to the security of a free state, the right of the people to keep and bear arms, shall not be
infringed.” The right to individual property, just like the right to bear arms, is in the Bill of Rights and
therefore protected by the Constitution. My point is that there are basic freedoms guaranteed by the
Constitution.

The FCC is attempting to make the “common good argument” to take away a property owner’s right
to make exclusive contracts in the area of pay TV. This argument does not hold water because, as I
have pointed out, from the moment the telcos put themselves in the pay TV game they had every
opportunity to get exclusives for MDU pay-TV service. So the telco isn’t being discriminated against
and it isn’t being barred from winning the MDU pay-TV business. There are no physical barriers to the
telcos competing for MDU pay-TV business now and there never have been! Therefore the ruling
doesn’t even pass the necessary test, let alone the pro-competitiveness or flawed pro-consumer (lower
prices will result) argument.

Martin: “Exclusive contracts between incumbent cable operators and owners of “multiple dwelling
units” (MDUs) have been a significant barrier to competition.” Huh? Since when? Where is the white
knight of video competition when it comes to rural pay-TV or rural broadband? Absent. The telcos
make the argument that economics are the reason they won’t bring broadband to the rural areas and
would rather sell off their rural lines than have to build out pay-TV capable plant. Martin buys that
argument but doesn’t buy the exclusive contract argument as being economics based for the delivery of
pay-TV? Am I the only one that sees the inconsistency in this positioning?

Some may ask, is the ability of an MDU property owner to make an exclusive contract with a pay-
TV provider really a “property right?” A well known economist and personal property rights expert,
Armen Alchian, speaks to property rights:

A property right is the exclusive authority to determine how a resource is used, whether that
resource is owned by government or by individuals….. Private property rights have two other attributes
in addition to determining the use of a resource. One is the exclusive right to the services of the
resource. Thus, for example, the owner of an apartment with complete property rights to the apartment
has the right to determine whether to rent it out and, if so, which tenant to rent to; to live in it himself;
or to use it in any other peaceful way. That is the right to determine the use. If the owner rents out the
apartment, he also has the right to all the rental income from the property. That is the right to the
services of the resources (the rent). Finally, a private property right includes the right to delegate, rent,
or sell any portion of the rights by exchange or gift at whatever price the owner determines (provided
someone is willing to pay that price). If I am not allowed to buy some rights from you and you
therefore are not allowed to sell rights to me, private property rights are reduced. Thus, the three basic
elements of private property are (1) exclusivity of rights to the choice of use of a resource, (2)
exclusivity of rights to the services of a resource, and (3) rights to exchange the resource at mutually
agreeable terms. (http://www.econlib.org/LIBRARY/Enc/PropertyRights.html)
 Clearly, the FCC’s ruling constitutes an illegal transfer of private property rights – the right to
sell any portion of the rights by exchange at whatever price the owner determines as long as someone is
willing to pay that price (the exclusive contract to provide pay-TV services) – from the MDU owner to
the government. Private property rights have been reduced here – the government, with the FCC being
the enforcing agency of the executive branch in this case, has seized a property right.

Continuing with my focus on the fourth amendment – it clearly states that the government
cannot seize property unreasonably. So this is where the real argument lies and whenever you debate
reasonableness you are clearly in the opinion zone. Is it reasonable to diminish private property rights
to further some veiled national goal of greater broadband penetration through increased competition? It
is clear that because of rising content prices, ample competition in pay TV has not reduced prices.
Chairman Martin conveniently ignores the rise in content prices when he reports the increase in pay TV
prices. It is also clear that the telcos have had every opportunity to compete on the merits of their
service for the next exclusive pay-TV contract that comes along. The telcos also have every reason to
pursue new contracts where exclusive contracts are in place, if they can offer up a better deal and
persuade the MDU owner to buy out the existing contract with the cable MSO he or she is under
contract with. Finally, I question the motivations of the ruling.

From the standpoint of services available at the curb – what is becoming known in the
telecom/cable services business as a “home (apartment) passed” – seizing private property rights does
not change the available options, it merely removes the choice of pay-TV provider from the owner of
the property and gives it to the renter. So that leaves only one justification – greater choice for the sake
of having more choices. Is that what the FCC’s mission has become? To put more players into the
market so that the profitability of the business is reduced? Is this the same FCC that has allowed
consolidation of the telecom industry with the understanding that economies of scale can be realized
through mergers?

Determining Factors as to Why Localities Do/Don't Go To FTTH

Max Kipfer, Connexion Technologies mkipfier@cnxntech.com

Having had hand's on experience with literally hundreds of communities that decide to or not to
bring fiber to their residences, I believe I have a pretty good handle on why a locality decides to or not
to adopt FTTH technology.

First of all, each community is different. Certainly the attitude of the developers/town
planners/zoning officials, etc is important. How economically minded are they? Are they aware of the
importance of broadband to the success of their community? What sort of budget do they have?
 Lifestyle Changes Represent Compelling Reason
One positive is the lifestyle change that fiber optics brings. People that understand something
about broadband can get turned on by the idea that they have access to a next-generation medium that
is going to give them the best opportunity to operate effectively from home. If a builder is selling the
idea that this community represents a step up, then FTTH can certainly play an important role in the
selling equation.

Hand in hand with that is the reality that home values connected by FTTH will increase above
those without it. If communities, and individuals, view FTTH as an amenity like a swimming pool in
the backyard or hardwood floors in the den, they will be more willing to “invest” in it.

Builders are concerned about moving the homes they have. If FTTH can act as an agent to get
to yes, then of course they want to include it.

On the downside, builders believe FTTH costs more, and quite honestly, it does cost more.
Price is certainly a consideration and if people feel as though they cannot pay for amenities then FTTH
may not be seen as a good thing.

Education Continues to Be an Important Consideration

All of these conditions depend on the knowledge of the developer or buyer. That is why to this
day a lot of my job continues to be education. People may have limited knowledge that some things are
easier to download off their computer than others, but may not understand that the quality of their
HDTV and ability to use multiple HD channels is directly dependent on the kind of pipe they have to
the home. Some people are tremendously savvy in this area. To others, it is barely on their radar screen.

Companies like Cnxn Technologies are important because when the developer does hesitate, we
front the cost for him. They generally are in it for the short term, but if they upgrade the telecom
offering and largely eliminate the loss of the higher cost, its a win-win situation. Cnxn Technologies
knows that FTTH is going to work and be worth it. Our footprint is the southern smile to California,
where most of the new growth is ongoing currently. We find one community, we blitz the community. It
can move very quickly. ..

Selling FTTH is not going to work from my perspective if you charge the homebuyer $7,000 to
$10,000 more for a fiber connection. What will work with a developer is that his houses will sell faster
if they have FTTH. A prospective buyer who has two comparable properties with one connected to
fiber is not going to hesitate. The fiber connection is going to make it happen.

Green-Friendly A Major FTTH Positive

Interestingly, the FTTH selling criteria is different than it was 12 months ago. We now stress a
lot more the green community theme. You won't have big ugly copper rusting in your backyard if you
go FTTH, you will have something smaller, tighter, more efficient, that operates better. FTTH will help
you maintain a green community. We still stress the fact that homes hold their value better with FTTH
and telecommuting may be an important consideration.
 Is the health of the overall real estate market a factor in
whether communities go to FTTH? The answer is of course. It is
far easier to sell amenities like connected fiber when people are
flush in cash and the market is booming. However, to put a
positive spin on the current market, people might feel they are
getting more for their money to buy an FTTH house compared to
one that is wired with copper.

Another negative argument we hear is that the bandwidth

capabilities of FTTH are exaggerated.
The casual observer may not see the difference in speed at which
their computer operates or the signal their TV is being delivered
over. We show them data suggesting otherwise but there is no
major differentiator that will immediately open their eyes.

Forecasters like Mike Render help us tremendously. We

go to developers and show them that two million homes in the
United States already have been connected with fiber optics. That
is leverage. IT shows the reality of FTTH.

We do believe the idea of futureproofing is important to

people. Most people recognize that they are using the Internet
much more than they used to and that the video signal and video
capabilities are becoming more sophisticated and important. They
can also grasp the idea that a pipe into and out of their homes
may have limitations, especially if it is based on technology
discovered over 100 years ago.

Do homebuyers generally know what FTTH is? They

have probably heard of it, but.don't know what it means. They
would probably say it translates into faster Internet, more reliable
Internet. That is where you start to add value. We take it to the
next level, say it is a premium product, that you get
associated benefits from it such as 24x7 support, etc. It is all part
of a premium bundle of power.

While fiber's claims may not be immediately evident

today, if the consumer is being educated to expect 8-10 Mbps as
the minimum standard, and they are willing to pay for it, we often
find that they want more. As a carrier or developer, we want to be
in a position of saying we will give you as much as you need. We
never want to be in the position of saying this is the best we can
do for you.

People may have heard of 100 Mbps, may even be excited

by the idea. We need to let them know they can get that through
fiber optics. We need to let them know that whatever turn their
new infrastructure takes, whether it be HDTV or IPTV or
whatever, that fiber optics is going to help them have no worries.
 The Triple Play infrastructure, the wholly converged solution, is what people are starting to get.
Add to that security and access control, the idea that you can control any device in your home remotely,
and the blinders often come off.

The homebuyer may think they can add it later. In fact, builders are now stranding parcels. For
example, a prospective homebuyer on a new lot says the community across the way is offering FTTH
but the earlier lots you have offered have not been FTTH. In order to stay up, your new section is going
to be FTTH. Homebuyers are becoming savvy enough where they recognize they at least want a choice
between fiber or not. They don't want to spend a substantial amount of money—sometimes their life
savings—only to find that their new home does not have the same telecom capabilities as the one down
the road.

Income is always a factor. Some of the younger first-time homebuyers who may not want to
spend $150,000 on a home but need a an Internet connection may not even require a landline
connection. To someone who is spending $500,000 on a home another $10,000 for a fiber connection
may not matter.

More and more service providers are following the Verizon model and going with FTTH. There
are still a lot of stubborn people who say they are going to use the existing infrastructure because it is
going to work. We point out that people who make that decision are going to have to suffer the
consequences. Sometimes its other carriers when the incumbent can't or doesn't want to provide FTTH.
Lafayette, La. was important because it demonstrated that.

The battle is being won community by community. We are confident this trend will continue
and grow.

Publicly Funded “Open FTTH”

An Opportunity for ILECs and the Communities They Serve
by Mitch Shapiro, Independent Consultant

There is virtually universal agreement that FTTH provides the greatest capacity and best
technical performance of any wireline network option. At the same time, however, there remain
serious questions about the economics of non-greenfield FTTH investments. The fact that only Verizon
among the nation’s three RBOCs has committed to a large scale FTTH deployment—and for only a
little more than half of its footprint—is testimony to the fact that, for most of the country, incumbent
telcos do not perceive adequate economic justification for FTTH investments.

This article considers one relatively unexplored model for overcoming the economic challenges
of bringing FTTH to the tens of millions of American homes that currently lie beyond the scope of
today’s FTTH business cases. This model combines a wholesale muni-fiber approach with the
participation of the incumbent LEC as an “anchor” retail service provider. Other potential ILEC roles
under this “open-FTTH” model could include provision of services and network resources to support
network construction and operations.
 Incumbent LECs could benefit from this model because it would give them access to a state-of-
the-art FTTH network without the capital investment burden and risk associated with building the
network themselves. At the same time, the participation of an ILEC would reduce the risk and improve
the economics of the muni-FTTH business case.

This model of FTTH deployment would apply especially well to capital-constrained ILECs that
have yet to launch aggressive fiber-deployment strategies (e.g., Qwest among the three RBOCs). It
could also apply to ILECs like Verizon and AT&T that are planning FTTH and/or FTTN upgrades in
only some of their markets. For these companies, the model could be applied to the large number of
local markets in which ILEC management cannot cost-justify a fiber-based upgrade. The model could
also apply to cable operators, particularly in markets where the cable network has not been fully
upgraded to the industry’s standard 750-860 MHz HFC (hybrid fiber coax) architecture.

Will ILECs Embrace This Model?

In the October 2006 issue of FTTH Prism, James Salter raised a key challenge facing the model
proposed here. In his view, ILECs “have no incentive to embrace these networks and become the
Service Provider ‘middle man.’”

Why? They can’t afford it. They would have to do one of two things—abandon their existing
copper/coax infrastructure or compete against themselves for customers. If they abandon their
existing infrastructure in order to use the ‘better’ municipal fiber network, they would have to
write-off their plant asset, take a huge hit against net income, lower the equity on their balance
sheet, and adversely affect their stock price. If they compete against themselves—copper
customers on their old network versus fiber customers on the new network—they would have to
replace a high-margin copper customer…with a low-margin fiber customer.

While the above may be an accurate description of historic ILEC attitudes, it remains to be seen
whether these attitudes will withstand the mounting competitive pressures facing ILECs in the large
number of markets in which they are not planning to deploy fiber-rich, video-capable networks. In
these markets, the combination of cable VoIP and Triple-Play bundles, wireless replacement, and low-
cost web-based services will increasingly turn what were once “high-margin” copper customers into
either low-margin copper customers, or negative-margin non-customers.

Weighing The Relative Risks

Though Salter is correct that the financial risks of accelerated write-offs of ILEC copper plant
are significant, this is also increasingly true of the risks associated with other ILEC options.

For example, it seems increasingly likely that ILECs lacking a robust on-network video strategy
will face a future of residential revenue stagnation and even decline.

Evidence of this can be seen in the graph below, which compares second and third quarter
revenue growth rates for the three RBOCs and three top-tier cable operators. As it shows, year-over-
year growth in cable revenue has ranged from the high single digits to the low teens. In contrast,
RBOC growth rates have languished in the low single digits.
 Residential Revenue Growth Rates

AT&T (1) 0.4%

0.1%

Verizon (2) 3.4%

3.1%

Qwest (3) 1.4%

1.6%

Comcast 11.8%
11.3%

Time Warner 8.5%

7.5%

Cablevision 13.0%
9.7%

0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0% 14.0%

3Q07 2Q07

The graph also shows that Verizon, which has been ramping up its FTTH footprint and FiOS
customer base for several years, has been reporting residential market growth rates well above its peers.
During this year’s second and third quarters, it’s in-region residential growth rate (excluding MCI mass
market revenues) was in the 3.1-3.4% range, versus comparable growth rates of just 0.1-0.4% for
AT&T (which is still very early in its FTTH-based U-verse rollout) and 1.4-1.6% for Qwest (which has
yet to undertake a broad video-capable network upgrade).

While risking revenue stagnation if they don’t invest in fiber-rich upgrades, ILECs also face
significant risk if they do undertake such upgrades in markets with questionable ability to generate
returns that are attractive to investors.

Strong concerns about such risks were expressed by Wall Street analysts during Qwest’s
October 30, 2007 third quarter earnings call, following the company’s announcement of an increase in
2008 spending for FTTN deployments. Following the call, Qwest’s stock closed at $7.06/share, down
13.7 percent from the previous day’s closing of $8.18. And during the month following the call, the
stock was trading below $6.80 per share.

Even the nation’s two largest and strongest RBOCs have implicitly acknowledged that a
substantial portion of their service area cannot justify investments in fiber-rich upgrades. As noted
above, Verizon has targeted only a little more than half of its total footprint for its FiOS FTTH build.
And in a December 11, 2007 analyst conference, AT&T announced plans to pass a comparable
percentage of its total footprint (including BellSouth territories) with its U-verse FTTN network.

In terms of risk, its worth noting that AT&T’s U-verse Internet service offers top speeds of only
6 Mbps, to be upgraded next year to 10 Mbps. This barely matches what most cable operators offer
today, and should fall well short of cable’s next generation DOCSIS 3.0 technology, and even more so
what FiOS and other FTTH networks will be able to deliver. So, while per-home capital costs are
substantially lower for U-verse than for FiOS, the former’s much lower data rates entails its own
competitive risk. And from a public policy perspective, it could be argued that AT&T’s U-verse
strategy increases the risk that many American communities will fall far behind their counterparts in
other countries in terms of future Internet data rates and services and the associated benefits.

Given the risk perceptions of ILECs and their shareholders, it seems reasonable to consider
publicly funded FTTH models in the large number of markets where investments in FTTH—or even
FTTN--cannot satisfy private sector risk and return requirements. As discussed below, this is because a
“public infrastructure” model has fundamentally different investment criteria than a private sector
model.

It is natural that ILECs’ historical role and financial structure as facilities-based service
providers would make them cautious about the type of model proposed here. But, given their other
options, it is the author’s contention that ILECs should at least consider this model in markets where
they cannot justify their own FTTH or FTTN investments. While this would entail significant changes
in the traditional ILEC business model, these changes may very well offer the best (and perhaps only)
alternative to stagnation and even decline in these markets.

Given its relatively weak capital structure, its lack of on-network video strategy, and Wall
Street’s initial resistance to recent signs of accelerated FTTN spending, Qwest might be an especially
good candidate for consideration of such a strategy. This seems particularly appropriate in light of its
recent change of top management, which has triggered a period of strategic reevaluation.

Externalities Favor A Public Investment Model

In addition to offering ILECs a path to fiber-powered growth that is largely free of investment
risk, a publicly funded wholesale FTTH model has the potential to generate substantial public benefits
for the communities in which it is adopted. The nature and implications of these benefits are briefly
discussed below.

Common sense, anecdotal information and a growing body of research strongly suggest that
widespread availability of high-speed open-access IP networks has the potential to deliver substantial
positive “externalities”. For the purposes of this discussion, the term “externalities” refers to benefits
that do not accrue (or are not perceived to accrue) directly and fully to those paying for communication
services. As a result, their value is difficult for service providers to monetize in the form of service
fees.

The desire to promote externalities within their communities is one of the chief motivations
driving the growing interest among local governments in “municipal broadband” projects.
Externalities targeted by muni-broadband projects include:

*attracting new businesses and strengthening the local tax base;

*driving economic growth and increasing the global competitiveness of the community’s workforce
and business base;
*increased efficiency in the delivery of public and private services, including education and
healthcare;
*bridging the “digital divide;
*reducing traffic congestion and pollution; and
*enhancing the overall quality of life of a community’s citizens.
 In his 2006 book America at the Internet Crossroads: Choosing the road to innovation, wealth,
and a supercharged economy, muni-FTTH advocate Mike Bookey considered the potential cost
savings associated with widespread FTTH deployment. He focused on two areas well suited to take
advantage of the symmetrical HD-quality video transmissions enabled by FTTH: reductions in motor
vehicle travel and delivery of home-healthcare to chronically ill patients. As he notes, the latter
account for a very small percentage of the population, but a substantial share of total healthcare
expenses.

Potential FTTH-Enabled Transportation & Healthcare Savings

10-yr.
First Year Total*
Transportation
Savings from fewer miles driven $74 billion $829 billion
Social savings from less gasoline consumption $4.3 billion $48.6 billion
Travel time savings $47 billion $537 billion
Road building savings $6.4 billion $71.7 billion
Total transportation savings $128 billion $1.5 trillion
Healthcare
Total U.S. healthcare expense (2003) $1.5 trillion $22.8 trillion
Expense for chronically ill $500 billion $48.6 bil.
Savings from remote monitoring of chronically ill $175 billion $2.7 trillion
Total transportation and healthcare savings $303 billion $4.2 trillion

* Assumes 2.5% and 9% annual increases in transportation and

healthcare costs, respectively.
Source: "America at the Internet Crossroads," by Michael Bookey

Bookey estimated that these two areas alone could generate total annual savings of more than
$300 billion if FTTH networks were deployed nationwide. That’s enough savings in one year to more
than cover the cost of building these networks. And while it certainly can be argued that Bookey’s
estimates are overstated, a case can also be made that they may be understated, since they don’t include
other potential public benefits in areas like education and economic growth.

Though, as Bookey’s analysis suggests, the total value of public benefits, including
externalities, may be quite large, there are inherent challenges in trying to maximize this value in
access markets dominated by private, profit-seeking companies.

This is because profit-seeking access providers lack sufficient financial incentives to invest in
capabilities and services aimed at maximizing externalities. As privately owned companies serving
shareholders’ interests, it would be difficult (if not illegal) for access providers to justify significant
investments targeting externalities whose value they cannot monetize and return in some form to their
shareholders.

In sharp contrast to private access providers, local governments are motivated to invest in
communications infrastructure targeting externalities. In fact, such “public infrastructure” investments
are a central role of government. In keeping with this role, governments also have access to relatively
low-cost and “patient” funding sources not available to private network operators, including tax-free
bonds and taxes.
 Market Entry, Efficiency, Competition And Innovation
Due to their high and largely fixed costs, the market for access networks exhibits substantial
barriers to entry. As Diane Kruse noted in the October 2007 issue of FTTH Prism, this cost structure
makes facilities-based competition an economically inefficient and unreliable means to achieving
healthy competition in applications and services. These inefficiencies and barriers to entry are
evidenced by today’s highly concentrated access market and the history of failure and bankruptcy
among CLECs and “overbuilders.”

In contrast, the economics of non-facilities based competition (e.g., at the application and
device layer) is characterized by relatively low barriers to entry and potentially high and efficient levels
of competition. A good example of this is the Internet application market, which is characterized by
relatively low startup costs and other entry barriers, rapid cycles of technical and service innovation,
and intense and multifaceted competition.

As is strongly suggested by the Internet’s dynamic growth, high levels of non-facilities

competition and innovation are most likely to occur in an environment in which access providers are
not able to significantly extend their market power into application and device markets. Such vertical
extensions of market power are likely to generate anti-competitive and innovation-dampening market
dynamics, because access providers have both the motivation and ability to position themselves as
gatekeepers between service providers and end-users. To the extent they do so, they distort the
interaction of supply and demand in the service and device markets in ways that favor their own
financial goals at the expense of end users, independent service and device suppliers, and overall
market efficiency.

Concerns about such vertical extension of access providers’ market power is a key driver of
demands for “network neutrality” regulation. (ed. Note: See Mario Rossi column on “Net Neutrality.”)
As implementation issues and other potential pitfalls of this approach have become apparent, some net
neutrality advocates have begun calling for “structural separation” as a more effective solution. Such
an approach would limit access providers’ role to providing wholesale services and would prohibit
them from discriminating with regard to the terms of wholesale service offered to retail service
providers. As such, it would avoid the anti-competitive impacts associated with vertical integration of
access and services.

Getting There from Here

Our discussion thus far suggests that an optimal network model for maximizing total value is a
single publicly funded, high-capacity, IP-based FTTH network operated on a non-discriminatory
wholesale basis.

This model would avoid the economic inefficiencies associated with duplicate physical
networks. It would also achieve the high levels of efficiency, competition and innovation associated
with an all-fiber IP open-access wholesale network. And, because the network would be funded as
public infrastructure, its business model could target valuable externalities and low prices without
triggering conflicts with private shareholders.
 The obvious problem with this model is that it posits a single-FTTH solution when, in fact, a
substantial majority of American homes are already passed by two copper-intensive legacy networks,
owned and controlled by incumbent access providers which have so far preferred a different model.
Given this, a key question becomes, how can we migrate from today’s status quo to a model that
captures as much as possible the benefits of the ideal “open-FTTH” model.

It is at this point that ILECs’ competitive weakness in many local markets becomes a potential
virtue. This is because their inability to cost-justify fiber-rich upgrades means these ILECs must either
accept stagnant or even negative growth in these markets, or look for other means to counter cable’s
growing competitive advantage.

One such alternative would be to become a retail service provider on a publicly owned and
funded open FTTH network. This would avoid the cost of a network upgrade while at the same time
giving an ILEC access to the advanced capabilities of an FTTH network. This, in turn, would free up
ILEC capital for investments in developing and marketing innovative, high-value, FTTH-powered
services, including some that cable’s HFC networks could not cost-effectively deliver.

In addition to sharply reducing an ILEC’s capital investment risk, this model also reduces the
total capex requirement and increases overall economic efficiency when compared to a scenario in
which a community deploys a muni-fiber network while the ILEC pursues its own separate network-
upgrade strategy.

This model could also reduce the "build it and they will come" risk faced by municipalities
pursuing a wholesale model. As James Salter noted in his October 2006 FTTH Prism article, a key
challenge confronting the wholesale muni-FTTH model is the risk of relying on retail service providers
that are “relatively small [and] underfunded” and lack the customer loyalty, resources and brand ID to
achieve the necessary take rates.

Since the ILEC is one of two existing service providers in a local market, it would have both
strong motivation and proven capability to attract retail customers to the services it delivers on a muni-
FTTH network. And its efforts to do so would be enhanced by the fact that at least some of its FTTH-
powered services could out-perform the HFC-delivered services offered by its cable competitor.

The combination of a municipality marketing the network as a whole and the incumbent LEC
marketing its own retail services would increase the likelihood that an open-FTTH muni network could
attract sufficient customers to be financially self-sustaining. At the same time, the municipality would
be prohibited from favoring the ILEC’s retail services, thus providing a level playing field for entry by
other retail providers.
 Some might question whether any other entity would choose to provide retail services on a
wholesale muni-FTTH network carrying retail services from the incumbent LEC. While some
companies and business models might be discouraged from market entry in this scenario, others
probably would not. The latter could include a range of IP-based service providers, including:

1. independent VoIP providers, including those attracted to FTTH’s ability to support high-quality
video communications;
2. large and small video content suppliers, which are already migrating to IP-delivery modes
aimed at bypassing cable/telco gatekeepers;
3. the growing ranks of Internet-based companies, led by Google, whose revenue models differ
substantially from ILEC models in that they focus mainly on highly targeted advertising rather
than end-user fees.

An open-FTTH wholesale network might also attract out-of-market ILECs or cable operators
looking to expand their reach without the hefty financial risk associated with facilities-based market
entry. But the participation of these entities would probably not be needed for the network to support a
healthy competitive ecosystem and achieve penetration rates sufficient to pay for itself.

Other Potential ILEC Roles

In addition to being a competitive retail service provider on the FTTH network, a community’s
incumbent LEC might also play other roles in the project. Given its existing resources in the local
market, it might contract with the municipality to provide construction and network management
services. This might make particular sense in communities that do not already have well-developed
internal telecom or public utility units capable of handling these tasks.

Given the model’s goal of non-discrimination, it would be important to ensure that the network
and business structures were designed and managed to ensure that the incumbent LEC would not
receive preferential treatment, even if it was involved in some network management functions. This
could require safeguards to prevent such preferential treatment and, if need be, include limits on the
ILEC's network management responsibilities (e.g., in terms of what network "layers" it would be
responsible for managing).

The ILEC’s role in building and managing the network could even extend to providing the
muni-net with access to fiber, conduit, poles and other facilities, via lease, purchase or other forms of
access. This would appeal to a municipality to the extent it could reduce its costs and/or expand
network coverage and capabilities. From the ILEC’s perspective, such arrangements would represent
another component of their business model transition. To ensure they did not provide an ILEC with
unfair advantage in the retail market, these arrangements would need to be managed under the same
non-discrimination umbrella that would apply to other aspects of the ILEC’s involvement in the
project.
 Challenges Expected, But Manageable
Should ILECs and municipalities begin to develop win-win models along the lines suggested
here, it's reasonable to expect that cable operators in affected markets would launch aggressive
lobbying, legal and public relations challenges to them. This would follow the typical model of such
challenges being mounted against proposed muni-fiber networks, except that, in this case, the challenge
would come from only one incumbent, with the other being an active supporter of the project.

This could substantially shift the relative balance of power in such legal and political battles, by
applying the ILEC’s impressive resources in these areas to support of the muni-fiber project, rather than
in opposition to it.

Though an ILEC’s involvement in a muni-fiber project has the potential to add ammunition to
opponents’ PR attacks on the project’s fairness and competitive impacts, it does not seem likely to
significantly increase the project’s legal risks. This is especially so given the open wholesale model
proposed here, which would allow an incumbent cable operator to access the FTTH network on the
same terms as the ILEC and every other retail provider. In fact, some states only allow munis to deploy
fiber networks if their role is limited to provision of wholesale service.

If a pro-muni bill pending in Congress becomes law, some states’ current restrictions on muni-
fiber projects would be removed, making such projects more viable in these states. The prospect for
such passage are likely to improve somewhat if, as expected, Democrats increase their control of both
Houses, particularly so if they also win the White House.

Should federal legislation lessen existing barriers to muni-fiber deployments, the pressure on
ILECs to consider the model proposed here could increase, since the launch of a competitive muni-
FTTH network would greatly increase the risk of revenue decline facing a community’s facilities-based
ILEC.

While the financial, legal and political hurdles confronting this model are certainly not trivial,
the point of this article is that—at least in some markets--both ILECs and municipalities have
increasing incentives to seriously consider it.

To the extent these incentives intensify and ILECs’ other options become less attractive in some
of their markets—which the author believes are fairly likely--the more room there will be to negotiate
win-win arrangements that provide a relatively attractive return to ILEC investors, while achieving the
substantial public benefits promised by the open muni-FTTH model.

Mitch Shapiro is an independent analyst whose recently published reports have examined the
economics and market impacts of municipal broadband, Verizon’s FiOS and bundled service
competition. He is currently working on a report comparing RBOC network and residential market
strategies and their prospects for success. He welcomes feedback on the ideas presented in this
article, and can be contacted at mitchshapiro@cox.net.
Fiber to Every Home is our Ground-Breaking Strategy
Christy Batts, Telecommunications Marketing Manager, CDE Lightband

We believe that here at the Clarksville Department of Electricity (CDE) in Tennessee we are
developing innovative ideas that are more broadly using fiber optics to help utilities and consumers.

Ours is the largest community in the United States with every premise hooked up to fiber optics.
The entire project cost $55 million and was funded from a revenue bond.

By providing fiber to every home in our district, we are able to remotely perform monthly meter
readings as well as the Triple Play services that work best over an FTTH infrastructure.

The impact will be dramatic on both counts. With a service area of over 54,000 metered homes
and businesses in a very transient and high growth market, CDE runs over 130,000 truck rolls annually
to connect or disconnect meters or to perform meter readings. With those truck rolls costing an average
of $30, this operational cost was becoming more difficult to manage and control as the community
continued to grow.

We calculate that the remote reading, connection and disconnection through the fiber connection
will eliminate over 90 percent of these truck rolls, saving CDE over $3 million annually.

Once the decision was made to build an 860 mile fiber optic network the next phase focused on
ways to maximize the benefits of the fiber. CDE went to the community voters in a referendum and an
overwhelming 72 percent of voters indicated that they wanted CDE to explore offering a choice in
video, Internet and phone providers for our community. Thus CDE Lightband was born. CDE
Lightband represents a new era in municipal power services with state of the art technology being used
to manage electric customers and to provide additional service choices in digital video, Internet and
phone.

We have taken advantage of the unlimited bandwidth capabilities of fiber optics to save money
for our utility and simplify how it does business, and to make it a cutting-edge provider of telecom
services. CDE Lightband will offer a full range of digital video services, high-speed Internet and phone
service all via an Internet protocol delivery.

By bringing fiber to the home to simplify meter readings, we also are doing something that not
even most FTTH companies are doing: We are going directly to the home with fiber even if the owner
has not ordered our telecom service package.

The exciting aspect of this type of deployment is that all of the homes passed with our electric
service will have the opportunity to choose the additional services we offer. If they do not choose our
optional services we are still providing a higher level of service and management of their electric
service while saving significant operational costs.
 We have made strategic partnerships with many companies to help us build, manage and
support this endeavor. Atlantic Engineering Group is our design and build company and has been the
driving force behind our construction of the fiber plant. Other partners include Amino for our set top
boxes, Kasenna for our middleware, World Wide Packets for our portals, Cornerstone Information
Systems for our Internet development and ETI will be our billing service. Programming is coming
from Telco TV with encryption support for Secure Media. Our VOD product is coming from
ViewNow. As of today we have all of the pieces in place except for our telephony provider. We are
currently in beta test with subscribers on our video and Internet product.

We have been encouraged by other communities that have gone to fiber and succeeded against
the stronger duopolies of large telco and cable TV provider.The cable TV provider in our area is
Charter Communications and they offer a full Triple Play while AT&T provides phone, Internet, and
dish services. Both entities were obviously not pleased with the prospect of our offering a competitive
choice in the market initially. Since the referendum and our beginning to offer services on a test basis
they have really focused their efforts on their business and in turn the consumer has benefited from
their renewed focus.

While it is using a facilities-based retail model, CDE did offer Charter Communications the
opportunity for open access, but Charter was not interested. They currently have an HFC plant in
place. Once we determined that we would have no one interested in a strategic partnership, and with
the support of the referendum vote, we felt obligated to the citizens of the community to provide them
the choice they have asked for.

Currently we will be launching with 194 channels of video with 20 of those channels in HD. We
will provide a full video on demand offering as well. Our high-speed Internet service will be delivered
at a full 10 Mbps. One of the comments we heard most from our customers about their desire for
additional service was they wanted assurances that we would maintain a local customer service staff,
local offices for support and that we would stay committed to providing a strong product at a fair price.
We feel certain that we can maintain our offerings and provide our customers with the local service and
support they desire.

Improving Economies and Quality of Life Through FTTH

John George, Director, FTTX Solutions, OFS, johngeorge@ofsoptics.com

Video has recently been anointed the killer application by leading industry experts. But an
application should only be described as “killer” if it enables high value services that improve our way
of life and help create what can be truly described as a new economy. Can video-based services offer
more value and possibly transform our economy? Yes is the answer, once there is enough symmetrical
bandwidth to support large screen, high definition, symmetrical video streaming and file transfers to
and from homes.
 Video Services that Can Improve the 21st Century Economy
Optical fiber-connected users could participate in a new economy facilitated by true global
collaboration through Very High Definition 1080p (VHD), later 2160p (Super HD), and in 10 – 15
years even 3D video telephony. Home-based workers could be able to quickly send and receive multi-
gigabyte video files used for training and presentations, while engaging in lifelike face-to-face video
meetings. Health care could be transformed by doctors assessing patients and medical telemetry though
crystal clear large video images. Online marketing and advertising could be video enabled, with higher
definition and 3D providing an attractive virtual shopping experience for consumers and businesses.
Personal and group experiences might be video recorded and shared in lifelike definition with other
fiber-connected users anywhere on the globe. These applications could help reduce transportation needs
and increase productivity, saving time and the environment. With enough bandwidth, the new video-
connected society could be more networked and productive than ever. Here are a few potential video-
enabled services, how they might evolve, and some keys to supporting these services over the optical
access networks that will carry us into a new economy.

Teleworking
Teleworking offers the benefits of improving the environment, lowering costs, increasing
productivity, and improved quality of life. In 2006, only about 11 percent of American workers
teleworked to any degree, even though 25 percent could and would telework if permitted by employers,
according to the 2005/2006 National Technology Readiness Survey. Those who do telework do so only
1.6 days per week on average. But would people telework more often, and receive more encouragement
from employers to telework, with better communications tools? Could HD video- enabled telework
triple participation and raise frequency to 2.5 days/week for 35 percent of the American workforce? If
so, the benefits could be astounding as projected in table 1.

Table 1
Teleworking Benefits Projection (USA)
2006 Baseline Potential 2006 Fiber Enabled
Teleworkers (%) 11.0% 25.0% 35.0%
Teleworkers (M) 16.0 36.3 50.8
Avg days/week 1.6 1.6 2.5
Miles saved/yr (M) 25,520 58,000 126,875
Business Travel Savings (%) 5%
GDP Increase/yr ($M) $ 37,429 $ 85,067 $ 186,083
Fuel Savings/yr $(M) $ 3,512 $ 7,982 $ 17,460
Business Travel Savings ($M) $ 8,500
Incemental over 2006 baseline
Productivity Improvement/yr ($M) $ 47,637 $ 148,654
Fuel Savings/yr ($M) $ 4,470 $ 13,948
Business Travel savings/yr ($M) $ 8,500
TOTAL ANNUAL BENEFIT ($M) $ 52,107 $ 171,102
Assumptions
Labor Force Total 145M
MPG 21
Miles/commute round trip 20
Gas Price $2.89/gal
Business Travel/yr US Corporations $170M
 The “potential 2006 case” assumes all who had the option from employers to telework or had a
job they perceived as amenable to teleworking participated. The “fiber enabled” case assumes that
fiber-enabled teleworking services drive fulfillment of the 2006 potential plus significant additional
participation involving about 1/3 of the workforce.

Fiber-enabled teleworking video services could provide an estimated incremental economic

benefit of $171B/yr from the 2006 baseline, translating to $3600 per teleworker annually, if
participation could be increased as shown. This does not include any benefits of reducing greenhouse
gas output by about 56M tons per year i , reducing road maintenance and construction cost, and reduced
office-space requirements. For a small city with a labor force of 10,000, these assumptions translate
into a potential economic impact of over $11M per year. However, such incremental benefits assume a
tripling of the number of teleworkers, and an almost 60 percent increase in the number of home work
days per week per Teleworker. How can we make that happen?

One of the often-cited inhibitors to teleworking is lack of face-to face contact as a

communications barrier, according to 32 percent of respondents in a large Federal Government survey.
ii
Another barrier is employer skepticism about productivity from home-based workers. From the same
survey, more than 60 percent of managers said they had misinterpreted a colleague through e-mail
communications. Additionally, disabled workers could get much more effective and productive access
to the workplace through improved video-based communications tools. iii
 Productivity can be much lower for workers at home due to slow access network speeds. These
concerns could be addressed by bi-directional, fiber optic-enabled, Very High Definition (1080p ) video
links between employers and home workers, and very high access speeds. Telephone calls and
teleconferences would become videophone calls and videoconferences, on large screens in very high
definition, with low latency, and with body language conveyed accurately. Upgrading these video
applications to 3D and very-high definition could truly emulate in-person interactions. E-mailing and
downloading large files, including video files, could be at corporate-network speeds. Unfortunately,
today most teleworkers are struggling with much slower connections than available in the office, as
shown in table 2. To greatly increase participation, teleworkers will likely need data rates comparable
to those available within the corporate LAN, plus additional bandwidth to support bi-directional HD
video.

Table 2 - Corporate LAN Intranet vs. Teleworker Typical Data Rates – 2007

Intra-company LAN Mbps Teleworker at Home

(Typical) Mbps*
Down- 100 1- 10
stream
Upstream 100 0.1 – 1

* Typical, not including fiber connected teleworkers. OFS data based on published industry
data. Teleworker at home case assumes no bandwidth sharing is required with other residents in
that home.

How can teleworkers keep pace when peers in the office are networked at rates 10 to 100 times
faster? Teleworking can be enhanced sufficiently to potentially drive much greater adoption by
providing fiber-enabled, VHD and large screen (>=42”) video conferencing to home-based workers.
Such “Telepresence” or “Video Conferencing” applications are already available, but are typically
constrained for home use by the very limited (typically 1 Mbps maximum) data rates provided
upstream by most broadband connections. Examples of Video file transfers and streams utilized by
teleworkers might include the following:

1. Video Teleconferencing--staying in visual touch would be a key to help increase teleworking

participation, in VHD on a large screen. Video meetings and interviews could significantly reduce
the need for business travel. Most households do not have the bandwidth, particularly upstream
bandwidth, to support even a single 8 Mbps MPEG-4/H.264 compressed HD stream.

2. Video Mail--another way of communicating with colleagues and superiors, to emulate the in
office experience. E-mailing a 2 minute 1080p VHD video clip would take about 32 minutes with
the fastest upstream connection (1 Mbps) available from most networks today, if no others in the
home were using the upstream capacity.

3.Video Uploads, Downloads--examples of this could be insurance workers uploading or

downloading damage assessment videos, real estate agents uploading virtual home tour videos,
downloading training or product videos, e-mailing PowerPoint files with embedded videos, etc.

Can Teleworking fulfill its promise? Very possibly, with high bandwidth optical fiber-enabled
services. But Teleworking is just one application that can be supercharged by optical fiber-enabled
access networks.

Health Care
Spending on Health Care in the United .States exceeded $2 trillion in 2005 and has been
growing at about 4 percent above the rate of inflation. iv Given the huge cost of healthcare, small
percentage improvements translate into large savings. Almost 30 percent of this cost is associated with
non-surgical interactions between patients and physicians. Fiber-enabled telemedicine, using very high
definition bi-directional video and medical telemetry connections, could enable home health care to
replace many in-person healthcare professional-patient interactions. In 2005 about $40B was spent on
home health care, $120B on nursing home care, and $420B on physician and clinical services. If fiber
enabled telemedicine could reduce each of these costs by just 10 percent, an annual savings of over
$50B could be achieved. Assuming the current growth rates in health care costs, a 10 percent savings in
2015 would equal over $100B.

Here is just one early example of a community leveraging fiber optic bandwidths to improve
health care cost and effectiveness. While seemingly like any other quiet southern community, new
telemedicine services are helping make Owensboro, Kentucky a leader in efficient, quality health care.
Fiber connections of 1,000 Mbps between hospitals and physicians are enabling quick and remote
diagnosis of patients through very-high-resolution imaging. The happy result is that patients now gain
faster diagnosis, in less time, with lower cost, and less commute mileage by the physicians. Owensboro
is just one of many communities benefiting from this technology.

Social Video
Social video is exploding onto the scene with countless websites for uploading and viewing of
videos. Examples of social video services include dating/matchmaking, entertainment, advertising, and
job searching on both public and subscription websites. Even with current bandwidth restrictions that
typically restrict these videos to fuzzy small images, social video is already immensely popular, with
over 100 million user-generated videos viewed each year and over 65,000 new user-generated videos
uploaded each day. While the business model for profitable social video is in its infancy, early
indications are that an advertising-based model could be viable. What is social video worth to an
economy? This may be difficult to quantify today, but it seems likely that a community having the
bandwidth to support social video services would be more valued by residents, thereby helping
communities retain and attract residents and their employers.

Entertainment Video: Television, Video on Demand, Pay per View, Movie Rentals
Internet video service revenues – including user-generated content, TV shows and movies – will
grow from $1B in 2006 to exceed $7 billion by 2010, according to Parks and Associates.

The economic benefit of entertainment video to a community could be translated into increased
home values and quality of life. For example, the $16B annual bricks and mortar video-rental model
can easily convert to a fully Internet-based distribution channel, saving building space, time, and
aggravation for renters (such as the author) who would rather spend minutes rather than hours per
month renting and “returning” videos. However, the building-based stores are likely safe for the
moment given that it would take eight hours to download a two hour HD-DVD at the current average
broadband download speeds of about 4 Mbps. At fiber optic speeds of 1,000 Mbps the same task would
take only about two minutes.
NBC and FOX recently announced a free, advertising-supported television service over the
Internet. This could be the beginning of a new distribution method by content generators. Will Internet
TV, VOD, and PPV supplement or bypass the traditional MSO, satellite, and emerging “Telecom”
service provider distribution channels? Or will the traditional and new entertainment video distributors
gain or increase their ownership in studios and content providers, and use private IP networks to
distribute content to customers? While the landscape for entertainment video distribution is rapidly
changing, one constraint that must be overcome regardless is the increasing bandwidth required to
support ever higher definition video on larger screens.

Distance Learning
Distance learning is a >$2B market in the U.S. alone, but has potential for much greater growth.
One traditional inhibitor of distance learning is the quality of the interaction possible between students
and instructors, and between students themselves. VDH, low latency video links could greatly enhance
distance learning and increase its adoption.

Videophone Communications
Wireline Videophone services can become much more desirable with enough bandwidth, and
low latency, to enable large-screen, high-definition communications. The global market for residential
video services will exceed $1B in 2007, but that could be just the beginning of a revolution from voice
to video communications. How much more growth would be possible with fiber-enabled video
telephony? Past implementations could in theory enable video communications with family and
friends, but these were severely bandwidth- and latency-limited, resulting in a fuzzy and
unsynchronized attempt at a conversation. A subscription-based model for this service providing an in-
person conversational experience, fiber-enabled with a low latency video pathway could generate
significant revenues and/or allow service providers to grow market share.

Supporting the Video Enabled Economy - Will the tools be available?

Higher Definition Video Evolution

Very High Definition (1080p) video available today is likely not the endgame for video
technology and utility. Screen sizes are growing quickly with 50 inch flat panel LCD and plasma 720p
HD monitors widely available for under $2000 and a 108 inch wall-sized 1080p LCD monitor currently
the top offering. But consider that with viewing distances unchanged, the resulting viewer experience
with larger screens is lower perceived definition Wall sized-screens could be in multiple rooms of the
home and will need even higher definition to maintain equivalent clarity. In response to the need for
large screen digital imagery, the industry is developing new 2D video standards, as described in Figure
1.
Figure 1 – Current and Projected 2D Video Formats and Estimated Data Rates
2D Video Applications Mb/s Native Mb/s (compressed)
Format per stream H.262 or MPEG- H.264 or
2 MPEG-4
Standard DVD
Definition Digital Broadcast 249 4 2
(SD)
Mature High HD Broadcast
Definition 1,493 16 8
(HD)
Very High HD-DVD Blue Ray
Growing Fast Definition Videoconferencing 2,986 32 16
(VHD) Camcorders

Super HD 14,930 100 50

New Standards Large Screens
Ultra HD 59,720 400 200
ITU-T J.601

3D Video
Multitudes of companies and universities are developing 3D video technology today. There are
three basic 3D technology categories under investigation. One requires the use of special glasses or
goggles, but due to the very strong industry desire to create a solution without visual aids, it will not be
considered in this paper. The second type of technology uses what appear to be 2D screen-equipped
“lenticular” pixels, so named because they are shaped like lentils, with each project light at different
angles, to create some perceived level of depth. The last type, volumetric displays, potentially provides
the greatest 3D depth and realism. Volumetric video displays also potentially consume the greatest
amount of bandwidth between the 3D technology options, and thus should be considered in bandwidth
planning for future application support.

Volumetric 3D displays are sold today for scientific, engineering, and other specialty
applications. Future versions of 3D volumetric displays could support the array of video services
discussed above. However, for volumetric 3D to meet these expectations, it could very well require
many times more bandwidth than 2D video, as projected in Figure 2 below, based on the stacked LCD
screen approach by LightSpace Technologies. Key assumptions include a significant improvement in
video compression from today’s MPEG-4/H.264 level with 720p HDTV of about 150:1 to a staggering
600:1 with future compression technologies. Even with 600:1 compression, 4 streams of Ultra
Definition 3D could consume 10 Gbps. Other 3D technologies could require different levels of
bandwidth than shown below. For example, it has been estimated that 3D video could require three
orders of magnitude, or 1000 times, more bandwidth than 2D video at an equivalent resolution, leading
to a staggering 125 Gbps for compressed Ultra Definition 3D video. The projection below is more
forgiving, and assumes something like the modulation technology employed by Lightspace
Technologies will allow only 20 stacked images to display a perceived 608 depth levels, reducing the
Ultra Definition 3D compressed data rate to 2.6 Gbps per stream. The key consideration is that optical
infrastructure installed today, which is expensive and disruptive to replace, should be able to cost
effectively support bandwidth upgrades to enable the 3D services likely in our future.
Figure 2: 3D Video Evolution – Volumetric Display (projected and estimated)

Image resolution Mega Frames/ Bits per Mb/s Native Comp. Mb/s
3D Video voxels
Horiz. Vert. Depth sec Voxel per stream Ratio compressed
Std Def.
768 480 20 7 60 40 17,695 280 63
High Def.
2160 1080 20 47 60 40 111,974 400 280
Super Def.
3840 2160 20 166 60 40 398,131 500 796
Ultra Def.
7440 4320 20 643 60 40 1,542,758 600 2,571

Video Content Generation

It is now possible for individuals to generate Very High Definition 1080p (VHD) video using
small camcorders costing less than $1000. While HD video content is broadcast today as 720p/1080i,
the traditional broadcast industry might upgrade to VHD (1080p) or even SHD (2160p) video in the
next 5–10 years as set-top box and video acquisition equipment life cycles shorten and if competition
from fiber-enabled VHD Internet-based video services impacts traditional broadcast video distribution
businesses.

Processing Speeds
The processing speeds required for 3D video might be 100 to 1000 times greater than that
required for 2D video Compare that demand to the most recent prediction by Gordon Moore that his
law, which says computer processing speeds double every 18 months and has been validated over 42
years, will cease to be valid in 10 to 15 years based on fundamental physical constraints. v This
translates, coincidentally, to processing speeds of 100 to 1000 times faster than possible today in 2007.

Storage
Storage is a key technology enabling video service proliferation today. The cost per MB of data
storage capacity has been dropping by about 35 percent to 40 percent annually for the past decade and
if this trend continues, the cost to store video files growing in size by about 40 percent annually should
keep storage economical as video file sizes explode over the coming years For example, the 14.4 GB
of magnetic disk capacity required to store a two hour VHD video today costs only about $12. If new
technologies enable trends to continue from 2007 at a 35 percent annual cost reduction per unit of
storage, the cost per GB will be <$0.045 in 2020, resulting in a storage capacity cost for a 2 hour, 252
GB HD 3D video of about the same $12 that is available today for business models utilizing video
storage technologies.
 Optical and Electrical Communications Data Rates
Today, an optical 10 Gbps per second transmitter/receiver module capable of 10 km
transmission costs less than $500 if purchased in high volumes. The theoretical capacity of a single
optical fiber is 50,000 Gbps over hundreds of km, leaving no doubt that it is capable or supporting
nearly any bandwidth demand for the foreseeable future. Copper-based electrical transmission systems
can support 10 Gbps over very short lengths of 10 meters to 100 meters on specialized indoor copper
cabling, and with future VDSL formats using multiple copper pairs may eventually be able to support
about 80 Mbps downstream and 4 Mbps upstream on 800 meters (2500 feet) of legacy outdoor copper
cables.

Network Bandwidth Demand Projections – will FTTH soon be a Requirement?

Will upstream bandwidth demands for high value video communications necessitate FTTH?
While “killer” video services are emerging, they will eventually and possibly soon require fiber optic
bandwidths to transform us into a new economy. Early copper networks supported 300 bits per second
(0.003 Megabits per second), and with 30 years of evolution the latest versions are approaching
theoretical bandwidth limits of copper cables, known as the Shannon limit. One of the constraints of the
xDSL and DOCSIS copper-based systems is a severe shortage of upstream bandwidth, limited to about
1 Mbps, resulting from the allocation of limited spectrum in favor of downstream bandwidth. While
this upstream bandwidth constraint did not significantly impact services during the early Internet web-
browsing era in which most traffic was low resolution and downloads, some copper systems could be
severely challenged to support upstream bandwidth demands in the near future.

First-generation FTTH systems range from about 5 Mbps to 100 Mbps downstream and 2 Mbps
to 10 Mbps upstream per home, enough to launch transformational video services, yet are using a tiny
fraction of optical fiber's potential bandwidth. FTTH can cost effectively evolve to the Gigabit per
second speeds that will continue enabling and expanding transformational video services, if one is
using a properly designed and specified optical fiber path. Figure 3 projects the household bandwidth
budget that might be required to support video based services at various levels of currently available
definition.

Figure 3 - Projected Video Bandwidth Budget for 2008 – 2015

Date Rate per Home (Mbps)

Video Applications SD 480p HD 720p VHD 1080p
Down Up Down Up Down Up
1 File Upload and Download 2 2 8 8 16 16
2 Conference or Video Phone 4 4 16 16 32 32
1 Shopping 2 - 8 - 16 -
2 TV/VOD/PPV/Gaming/Learning 4 - 16 - 32 -
TOTAL per Household (Mbps) 12 6 48 24 96 48
Assumptions: 6 person household with applications operating simultaneously. File upload and
download data rate assumes a 1 minute video with MPEG-4 compression level can be
transferred in 1 minute.
 Within seven years, video-enabled VHD applications could drive household bandwidth needs to
about 100 Mbps downstream and 50 Mbps upstream to support peak demands. While the temptation
may exist to oversubscribe fiber feeder or HFC fiber links using statistical algorithms, one should also
consider that real time video service customers with heavy symmetrical bandwidth demands will likely
be very intolerant of interruptions compared to traditional Internet surfers. Given the confluence of
continued exponential technology improvements and cost reductions, potential value of VHD video
applications, service providers and communities able to support this video bandwidth budget could be
in the best position to profit from high-speed network investments.

What about the future, more than 10 years hence? To what bandwidth levels should networks
being built today aspire to support video-driven bandwidths over the more than 25 year-life of an
optical cabling system? This becomes a more speculative exercise, but is worthy of consideration to
help avoid potentially costly, premature network infrastructure replacement. In figure 4 we take a
glimpse into future bandwidths, which might be valued on networks that we are building today.

Figure 4– Projected Video Bandwidth Budget per Household (Mbps) - 2015 – 2035
Video Applications 2015 - 2025 2025 - 2035
SHD 2160p 3D UDH 4320p HD 3D Super 3D Ultra 3D
Down Up Down Up Down Up Down Up Down Up Down Up
1 File Upload and Download 50 50 63 63 200 200 280 280 796 796 2,571 2,571
2 Conference or Video Phone 100 100 126 126 400 400 560 560 1,593 1,593 5,143 5,143
1 Shopping 50 - 63 - 200 - 280 - 796 - 2,571 -
2 TV/VOD/PPV/Gaming/Learning 100 - 126 - 400 - 560 - 1,593 - 5,143 -
TOTAL (Mbps) 300 150 377 189 1,200 600 1,680 840 4,778 2,389 15,428 7,714

Could 15 Gbps be required downstream and nearly 8 Gbps upstream in 25 years? Data rates
1000 times faster in 25 years seem extreme and unlikely today, but so did today’s leading edge 15
Mbps residential access speeds 25 years ago, when a rate of 0.0015 Mbps (1000 times slower than
today’s rates) was coveted by early home users. With continued exponential performance and cost
improvements in processing speeds, optical data rates, storage, and video technologies, the networks
built today should be cost effectively upgradeable to support the gigabit data rates and video services
on the horizon.

Optimizing FTTX Networks for Video Applications

The optical fiber cabling system is the pipe enabling the video revolution today, and must
support many generations of bandwidth upgrades. Future higher definition and 3D applications could
drive bandwidth demands to Gbps per home symmetrical. By the end of 2007, over 10M homes, about
9 percent of the total, will be passed by FTTH in the US. However, of the nearly 700M Multiple
dwelling units (MDUs) across the globe, only about 1 percent of these units are passed by fiber. There
have been many innovations helping to reduce the cost and increase the performance of FTTX
networks. Among key topics of interest today are how to best support current and future video
applications over the passive fiber network, and reducing the cost and time required to deploy fiber
inside apartments, condominiums, and homes.
 Supporting Video through the Bends in MDU and FTTH Networks.
In the face of video-driven data rates quadrupling every four years and tight optical power
budgets, the constraints for today’s forward-looking fiber network are primarily bandwidth and loss.
While the theoretical bandwidth of optical fiber is staggering, having more usable wavelengths on each
fiber helps reduce the cost of bandwidth upgrades by enabling lower cost per bit systems using coarse
wavelength division multiplexing (CWDM) systems. The usable wavelengths of the optical path can be
increased by up to 50 percent over conventional fiber using Full Spectrum G.652D single-mode optical
fiber. This fiber along with Full Spectrum splitters and low loss connections nominally solves the
bandwidth and loss concern. However, service providers have recently become very concerned about
added loss caused by the fiber bends– particularly for bend sensitive video wavelengths used in FTTX,
Enterprise, and HFC networks, when running fiber inside a premise or small enclosure. The
combination of tight power budgets and conventional single mode fiber can shut down video services
in today’s compact, bend challenged networks.

Conclusion
Video applications with increasing definition and personalization can help drive both bandwidth
growth and successful business cases in FTTX and ultimately most all-optical networks. The economic
impact of a suite of video-based services enabled by fiber optic bandwidths could exceed $200B for the
nation based on improved efficiencies and utilities for business, healthcare, entertainment, and other
services. The projected improvements in technologies supporting higher definition and 3D video
including computer processing, data storage, optical system data rates, and video appear to cost
effectively support continued increases in video clarity initially in 2D and later in 3D, likely driving the
need for Gbps of symmetrical bandwidth for each home and MDU over the decades long lifetime of the
optical- and copper- based cabling systems installed today.

Fiber-enabled video services better support the business case over a performance and cost
optimized optical path to help enable maximum current and future benefits from the network. A video-
dependent network should use video-optimized fibers for video applications, which primarily operate at
bend loss sensitive wavelengths from 1480 nm–1625 nm. Bend-optimized fibers are available that
enable use of the Full Spectrum with low loss for increased bandwidth, provide low-bending loss, and
are fully compatible with the installed base of standard G.652 fiber. These bend-optimized fibers are in
high demand anyplace in networks where tight bends are possible, to enable enhanced services and
increased application reliability for greater potential revenue and subscriber retention. To help meet the
challenge of reducing the installed cost for running fiber inside the MDU and premise, new solutions
are available that leverage bend-optimized fiber and space efficient packaging to reduce installation
labor and installed cost. Leveraging these innovations can help fiber-based video services and networks
fulfill their potential to create a new economy.
Corning Connected Communities:
Helping Developers and Infrastructure Providers to Market the Values of FTTH to Home Buyers

Ginger Stevens, Manager Market Communications

Dr. Bernhard Deutsch, Director Marketing and Market Development, Corning Cable Systems

Home buyers have three primary reasons to choose a certain neighborhood. The first two are
fairly obvious: location and price of the home. However, the home’s amenities are ranked third on this
list. Communication infrastructure has become such a significant part of our lives that it has drastically
increased in importance among the list of amenities. Home buyers have two main considerations here:
if the communication infrastructure will support their current and immediate future needs; and if it will
become outdated when they sell the home.

Understanding the different perspectives of home buyers and developers has defined the
marketing strategy required to sway potential home buyers. At Corning, we were first approached by
PAXIO Inc., an FTTH service provider in the southwest of the United States. Then a start-up, PAXIO
was working closely with the nation’s largest developer on a concept to market their FTTH services.
PAXIO’s CEO, Phil Clark and the developer’s marketing director had the vision to promote FTTH to
potential home buyers. This program would provide the benefit to the service provider by easily
reaching interested consumers and to the developer by promoting the advantages of the superior
communication infrastructure to potential home buyers. Best of all, consumers benefit from knowing
where their dream home will be built with the amenity of an FTTH network.

Home buyers as “consumers” need to feel comfortable with the technology, its benefits and the
providers of it – for both the communication services, as well as the network components. They also
need an easy-to-find information source to answer their questions. After several joint meetings between
PAXIO, Corning and the developer, it was decided that a marketing program would be best to raise the
awareness of the technology, educate consumers and ensure they feel comfortable with its performance
and quality. With this in mind, the Corning Connected CommunitySM program was created, and Pulte
Homes in California became the charter member.

The benefits of the Corning Connected Community program include several different tools to
meet the discussed requirements. Targeted advertising in print media that appeal to home buyers like
The Real Estate Book as well as online services such as Homebuilder.com are supported with the
Corning Connected Community brand. Developers who are program members have access rights to the
logo for use in individual community ads as well.
 The picture below shows an example of an advertisement that is targeted towards potential
home buyers to ask for more information about FTTH and where they can get it (left) and an example
from a developer's advertisement for the Kellswater Bridge development in Kannapolis, North Carolina
(right).

As part of the Corning Connected Community program, developers also receive a promotional
package for on-site use in their sales offices, as well as model homes. Special signage for the entrance
of the development and yard signs provide visibility and differentiate the community from those
without FTTH technology. The Fiber-to-the-Home Consumer Benefits and the Broadband Comparison
videos explain how FTTH works and how the technology will enhance the living experience in their
new home. Display advertisements, window
decals, brochures and other consumer education
materials complete the on-site marketing package.
Additionally, developers and service provider
members provide input for case studies analyzing
the benefits of the FTTH. These case studies will
be used to convince home buyers to select a home
in a development with an FTTH infrastructure.
Again, all parties win. Case studies and articles are
featured on a special website created as a resource
center for potential home buyers who want to
understand the technology and its benefits for
them: www.corningconnectedcommunities.com.

In return, developers and service providers commit to deploy optical fiber cable to each home in
the selected development via a quality outside-plant network. A Corning Connected Community will
provide home owners with the assurance that their communication infrastructure meets any future
broadband requirements.
 The success of the collaboration is very visible. PAXIO has passed 30,000 homes in the
southwestern United States already. Key developments in the San Francisco Bay Area have been
marketed as Corning Connected Communities: Mariani Square and Bentley Park in San Jose, Elevation
22 and City Limits in Emeryville, and 428 Alice and Jade in Oakland to name a few. Danbury Place in
Sunnyvale was the first Corning Connected Community. PAXIO receives between 1,400 and 1,600 hits
(through email, phone, and web) per month from interested consumers trying to find out how and
where they can find the FTTH amenity from PAXIO now.

Subsequently, several other service providers working with home developers approached
Corning about the Corning Connected Community program. Connexion Technologies quickly saw the
value in a joint marketing effort and joined the program in 2007. They market the program and its
materials package to developers as a way for them to gain exposure in the community. Developers rely
on the community signage for visibility and to differentiate themselves from their competitors. “It’s so
easy for them to take advantage of the materials and to use as much or as little as they want to fit into
their marketing plan,” says Jaime Cnossen, National Field Marketing Manager of Connexion. “With
the real estate market in a decline, it’s more important than ever for developers to differentiate
themselves, and the extra exposure from press releases, articles and industry awards provides them one
more advantage.”

The pictures below show the Corning Connected Community signage in front of a model home
and promotional material for Connexion Technologies and Corning Connected Community inside the
model home.

PorchLight Communications is the newest member of the Corning Connected Community

program. “After extensive research into several top suppliers of the FTTH industry, we chose to partner
exclusively with Corning because of the quality, reliability, availability, and innovation found in their
entire line of products,” said Corey Smith, Chief Technology Officer of PorchLight Communications.
“Becoming a member of the Corning Connected Community program was a natural progression of this
decision, because it helps us efficiently deliver high quality network elements in all of our projects.”
The Corning Connected Community program helps them gain instant credibility. Reliability, brand-
name recognition, and efficiency are paramount for PorchLight’s future success, because FTTH is a
relatively new product in the eyes of many builders, developers and home buyers. Regular updates of
new products, sales, marketing and engineering resources keep PorchLight focused on their core
business with confidence and efficiency.
 In summary, driven by developers’ and service providers’ needs to find a marketing tool that
will educate potential home buyers about the life-enhancing benefits of the FTTH technology and its
value for the home owner now and when the home is sold, the Corning Connected Community
program was developed. Utilizing the brand that best resonates with consumers and communication
infrastructure experts the program provides educational material for home buyers and a strong and
useful marketing package for developers and service providers. In turn, home buyers gain the assurance
that the communication infrastructure to their home is state of the art for many years to come. Several
examples exist where this cooperation benefited service providers such as PAXIO, Connexion
Technologies and PorchLight Communications.

Acknowledgements: The authors would like to thank Phil Clark, CEO of PAXIO, Jaime Cnossen,
National Field Marketing Manager of Connexion Technology, and Corey Smith, Chief Technology
Officer of PorchLight Communications for providing their insights into the developer market and the
benefits of the Corning Connected Community program.

Net Neutrality
Mario Rossi, Independent Consultant

Introduction – What is Net Neutrality?

“Net Neutrality” is the short term for Network Neutrality and states that all Internet sites and
their content should be equally accessible to any Web user. It should not be confused with “Open
Network,” which is a network that lets all of its users interconnect to some or all of its functionalities
and interfaces equally. The Net Neutrality cares about content and Open Network cares about the
means to deliver such content.

Net Neutrality has been gaining considerable attention lately since it conflicts with one of the
basic Internet principles: the freedom of information. The subject gained prominence when a couple
years ago major carriers and ISPs declared that they were planning to start “tiering” the content
delivery based on premium services (prices). It means that a company not subscribing to the
“premium” service would see their packets slowed down or even rejected in a certain network, meaning
that end-users sitting at their computers would not be able to subscribe or use the services of a
competitor of that carrier or even know that it existed at all. It does sound like the Chinese
government blocking unauthorized content or George Orwell’s 1984.

What is the Status?

In the United States, the House of Representatives passed telecom bill H.R. 5252 (“To promote
the deployment of broadband networks and services”) in May 2007. The Senate Commerce
Committee passed telecom bill S. 2686 on June 28. An amendment to the bill adding meaningful net
neutrality safeguards failed. The end of the 109th Congress marked a victory for Net Neutrality
advocates since no bill allowing the carriers' and ISPs' planned practices passed. The debate will
continue next year and we can expect fiercer “battles.”
 This battle is spreading overseas and the debate is gaining momentum in the United Kingdom
where Tiscali and PlusNet are joining AT&T, Verizon and Comcast’s voices, and developing tiered
packages to their subscribers and charging extra fees to content providers.

Recent news about Comcast filtering and discriminating packets generated outcries and finally
opened the eyes of many people who believed the carrier and ISP statements that this would “never
occur.” To monitor such activities, the EFF recently developed new software to monitor the net
discrimination that is freely available on that organization’s website (4). Politicians are starting to
voice their electorate’s opinion and municipalities are considering this factor when evaluating their own
broadband networks.

Can Net Neutrality hinder progress as its opponents claim?

Anybody reading this article who had a computer connected to the Internet in the early 1990s
probably still remembers that, despite hosting a very rich content, it was just a mere glimpse of what
we can find today. Yet, it was available to all, since the “Net” was conceived to help spread knowledge
and connect people worldwide. The government was heavily funding the Internet and several
technologies such as WWW were not available yet. I can still remember using software such as
Gopher to look for information since the Web Browsers were very limited and information not ready
for them. Many companies did not believe it would become what it is today and several that helped
building its foundation are no longer around (e.g. Netscape). Yet, free competition and widespread
resources availability led the Internet and several industry sectors in unprecedented growth. Mindsets
shifted and new business models allowed for economical exponential growth worldwide.

One cannot deny that it took a lot of investment to bring the Internet to what it is today and I
can barely imagine what we will be doing with it ten years from now. Yet, information has been freely
available without discrimination and this is definitely one of the main Internet pillars.

Companies willing to make money in the “Net” must reinvent themselves and create new
business models. The old monopolistic school of thought does not have place in such dynamic and
creative environment. There are countless cases of companies that strove and became the new giants
of the World by doing that (Google, Yahoo, Microsoft and eBay are just a few). The Net also changed
society by enabling several new services such as online search, blogs, video and file sharing and online
sales by companies that make millions or billions of dollars primarily on advertisement and charge
nothing to its users.

So, why are the large carriers and ISPs complaining? They already charge connection fees on
both ends of the network (e.g. subscribers like you and me and companies connecting their data centers
to the Internet). The excuse of high traffic is not very strong since the subscribers paying the
connection fees are generating it.

The Internet grew to the stage we see it today by being “free” (we still have to pay for access)
and non-discriminatory. It is hard to imagine where it would be today if it were in the hands of a
single monopolistic service provider.
 Conclusion
The Internet has been demonstrating the need for new marketing tools that segment the market
to the individual level and no longer large groups or classes. Companies that saw this and succeeded
in developing these tools are thriving on the Net. Google and Yahoo are two examples that do not
need much detailed explanation. Google recently started to spearhead a wireless open network where
it can expand its business model.

People want to be free and to choose what suits them better. The iPhone is an excellent
example of this trend. Users did not want to be stuck with the carrier-of-choice’s (AT&T) network and
massively adopted device-unlocking techniques that were available shortly after the device went to
market. Verizon apparently is learning from this kind of consumer behavior and declared support to
open networks and Google’s new open device that will compete against the iPhone. The future does
not belong to the stronger or smarter but to the most adaptable.

Therefore, in short, Net Neutrality helps the Internet, which promotes competition and
economic progress. History has been consistently demonstrating that open platforms and standards
tend to succeed more often than the opposite (e.g. VHS vs. BetaMax, PC vs. Mac and the list goes on).
Discriminating and closing the networks just appears to be like going in the opposite direction of
evolution and that could end-up penalizing society. If companies like AT&T, Verizon and Comcast get
their way; their subscribers may end-up with the same cable TV business model offered today where
tiered subscriptions offer access to different channel packages. An Internet equivalent would be a
subscriber only being able to access certain websites or regions and tiered services with a similar
package system. That is definitely not what the Internet is all about.

Software, of all Things

Jim Farmer, CTO, Wave7 Optics

How about this? Me writing about software. Hey, I’m a hardware guy from way back; I think
software is what I get when the laundry forgets to starch my shirts. Nonetheless, our products today
run on software, so I guess I have to acknowledge it. Software has made our products smaller and
cheaper than if they were made completely out of hardware. And software has the neat property that
you can change it remotely as you fix bugs and add features.

So what’s the fundamental difference between hardware and software? With software, you
control a limited amount of hardware and make it do different things at different times. Say we want to
multiply two numbers together. We can build a complex set of adders and other logic sets to operate on
each bit that needs to be manipulated. The pile of hardware could get pretty big. It would be fast, but
it would take a lot of logic elements (hardware) to do it, and it couldn’t do anything but multiply two
numbers. Alternatively I could gather up a couple of memory slots to hold the intermediate and final
results, combine that with a limited number of logic elements and some software that told the pieces
what to do at each step in the process. It would be slower, but it would take less hardware. And when I
was finished with the multiplication, I could use the same hardware to do something else. I store the
instructions on what to do in memory, and when I need to change the instructions, I simply replace
them in the memory.
 So that’s kind of how the hardware/software decision is made: hardware is faster at doing
things, but software allows me to get by with less hardware, and lets me change the instructions
remotely when I need to. As hardware gets faster, I can do more in software, and this generally lets me
reduce the amount of that fast hardware I need, reducing the size and cost of what I’m building. If
software is fast enough (given the available hardware on which it runs), then I do the job in software.
If I need more speed, I do the job in hardware.

Take modern GPON or GEPON systems. There are folks who specialize in building
application-specific integrated circuits (ASICs - hardware integrated into an integrated circuit) that
support the transmission standard, doing all of the fast stuff to format the data for transmission per the
standard. This includes adding (at transmit) or detecting and using (at receive) so-called header
information, that tells the system what to do with the packet of data. It also includes encrypting and
decrypting the data so prying spies can’t figure out what’s being sent to another subscriber. And it
includes figuring out what type of data is in a particular packet, and handling that data so that it gets the
quality of service (QoS) it needs.

These tasks are handled by a combination of hardware and software integrated into the ASIC.
Most ASICs tend to be hardware-oriented. But ASICs themselves have certain processes that are best
handled in software, so there is built into the ASIC the ability to execute (meaning to perform, not to
kill, though sometimes we are tempted) certain steps in software. The ASIC manufacturer may supply
the software to do these tasks. Sometimes it’s called firmware rather than software, based on what it
does.

Us guys who manufacture the FTTH systems start with the ASICs, then surround them with a
bunch of other hardware needed to make an OLT and an ONT. Part of the hardware is inevitably a
processor of some sort to execute more software, which we write. Or in some cases we buy parts of the
software and integrate it into the software we write. A piece of software we buy or write is sometimes
known as a stack. No, I have no idea where the name came from. For instance, we will buy an
operating system that performs a lot of low-level functions in the system, just like a Windows or MAC
operating system on your personal computer. We also may buy a stack that performs most of the
functions required to implement, say, the voice over internet protocol (VoIP) feature. But there’s a lot
of additional software that has to get written by the system manufacturer.

For many reasons, software may have to be changed after a product is in the field. This is one
of the neat features of software, as opposed to hardware. Can there be bugs in the software? You bet.
The bugs can creep in at any level, the ASIC manufacturer, outside stack vendors, internally generated
software and so on. Planned features may not make it into early releases of software, but become
available later. Feedback from customers leads to better ways to do things. Modern systems can
handle this very neatly. A low-level piece of software is written that controls downloading newer
software to a device. While new software is downloaded, the old software continues to run. The new
software is loaded to a different place in memory than that occupied by the old software. We call the
downloaded set of software an image. So I can have two images loaded in a piece of hardware at the
same time: an old image which is and has been running, but maybe doesn’t have some features of the
new, and I have an image of the new software that has been downloaded but is not running yet.

When I am ready, I send a command to the hardware, telling it to restart using the new image.
This is almost always a service-interrupting operation, so I’ll probably do it late at night when few
people are affected. When the new software starts running, I can test to make sure it is doing what it is
supposed to do. But the old image still remains in the equipment and can run again if something goes
wrong with the new software load. Things can go wrong, even though you do extensive testing of the
new software before you ever release it to the field. Some customer may have hardware hooked to an
ONT that you didn’t expect, and it causes a problem. Or the hardware may have an old version of an
operating system you didn’t know was out there, and may not have a feature you were expecting to be
there. Or something else may go wrong.

If something goes wrong with a software load, you can send a command to go back to the old
version. Sometimes the system will automatically revert to the older version if you don’t send it a
commit command within a certain length of time. This protects you from a problem where for some
reason, the new software locks up the equipment so that you can no longer send commands – if it didn’t
revert to the old software, your only alternative at that point would be to roll a truck to the affected unit.
Sometimes, if you don’t send a commit command, when power goes down and is subsequently restored,
the software will reboot with the old image, not the new. Again, this can protect you from problems,
but can cause problems, sometimes months down the road when the device reverts to the old image
unexpectedly, because it rebooted for some reason.

But the division between software and hardware is not nearly as clean as we’ve made it out to
be. For instance, in many of our triple play ONTs, we provide a feature called RF return, which takes
upstream transmissions from cable TV set top boxes and sends them to the headend (central office to
you bellheads). Most functions that must be done to support this feature are done in an integrated
circuit called a field programmable gate array, or FPGA. The FPGA is an IC with a bunch of logic
blocks (hardware) in it that will do anything you tell them to do, within reason. You tell them what you
want them to do using, of course, software installed in the FPGA.

This software is very specialized and is close to the operation of one subsystem, so we often call
it firmware rather than software. It can be modified and new firmware downloaded just as with any
other software. This lets us perform functions that need to run too fast to be done completely in
software, yet gives us the versatility of software in that we can reconfigure the “hardware” after a box
is in the field. Unlike software, firmware lets us configure a set of hardware elements to do whatever
we need to do, and do it fast. Since we are not building the hardware from scratch (we buy the FPGA
from vendors who specialize in such), the one-time cost is low.

There are other types of hardware that run software in an FTTH system. For example, the
interface between the voice over IP (VoIP) section and the analog telephone lines into which you plug
the subscriber’s telephones, is handled largely by a pair of integrated circuits; the digital portions are
handled in a device called a subscriber line interface circuit (SLIC) and the analog in a device called
the subscriber line analog circuit (SLAC). But there are some functions that are better handled in
software running on a special microprocessor called a digital signal processor (DSP). The DSP is a
processor with built-in hardware that is designed to perform certain mathematical functions very fast,
but controlled by software. The echo cancellation function is one that comes to mind that is best
handled in software on the DSP chip.

So there is a lot of software running the hardware in pretty much every piece of equipment you
have today, from your mp3 player, to your car, to the OLTs and ONTs in your FTTH system. That’s
kind of a hard thing for this old hardware boy to admit, but we need every trick in the book to drive
cost down and performance up on every piece of gear around. The right combination of hardware and
software helps us with that optimization.
Verizon FiOS Makes Its Way Into U.S. Military
David Chaffee, Editor, The FTTH Prism
The Verizon FiOS program is having an enormous impact on Western World communications
and undoubtedly is hastening the day of universal FTTH. But beyond the independent telcos, PTTs,
utilities, etc, that it has energized, it also has gotten the attention of the U.S. military.

Like its broad FTTP rollout, Verizon deploys a passive optical network (PON) architecture for
the U.S. military using single-mode fiber. The installation is from the edge of the campus to the
desktop.
“There are no active electronics, nothing to break and it saves a lot of space and power and has
a lot of headroom,” says Bill Kite, director and general manager of the program. By headroom, Kite
means the network can grow into the applications the military will require across the fiber. “This
network will last the next 50 to 100 years,” he observes. “We offer the government a clear path future-
wise.”
Critical to the program is the additional security it offers. “A problem with copper distribution
systems is that they radiate information,” says Kite. These “compromising emanations” can be tapped
and security lost, he notes. “Fiber optics does not radiate signals like copper distribution,” says Kite.
“We don't have to shield it.” Adds Kite: “Glass reduces the risk that someone could intercept the
signal.”

“We are focused on getting inside the building and operating at Layer 3 and above,” says Ed
Hill, director of engineering. Verizon works with SAIC, which provides the value add. Often it is
government agencies with “very high security needs.” Adds Hill: “We bring the strength on the front
end infrastructure and they add the applications and tool.”

The governmental program was actually ahead of the civilian FTTP program in that it already
began incorporating small desktop optical network terminals (ONTs) in 2007 in significant numbers.
The ONT is the size of a small modem and itself sits on the desktop or underneath it, according to Kite.
The ONT has 4 Gbps capability.

“We work very closely with the Verizon engineering team doing the FTTP buildout and we are
using the same vendors that have been approved through that program,” says Kite. A key difference is
that testing is more rigorous. “We ensure that the supply lines are protected.”

“We essentially start from the interface point where service comes into the building,” says Kite,
“and then take care of everything in the building.”

Verizon Business provides a similar service to corporations which is not quite as expensive
normally because the security requirements often are not as great. Governmental contractors represent a
portion of that market. “The solution usually makes sense for buildings that have 500 users and above,”
Kite tells us.

While the starting point is the continental United States, the business unit already has gotten
busy on several international installations. It is not constrained to the Verizon telecom serving area.
“Video is a major driver,” says Kite, again suggesting an important parallel to the Verizon FiOS
program. “People want their data in real-time as well.”

“It is significant for a lot of our governmental customers that we are able to provide voice, video
and data on a single strand,” says Hill. “We are geared to the upward growth in bandwidth. These
offices can grow without ripping out or replacing the infrastructure.”

The FTTH Prism magazine is currently published on a quarterly basis. Back issues can be retrieved at
www.ftthprism.com. For advertising rates or to submit an article contact the publisher
at 410-988-2723 or cdcfiber@aol.com.

Editor and Publisher...................................................................David Chaffee

Director of IT..............................................................................Jason Scammell
Editorial Advisory Board: Timm Bechter, Bernhard Deutsch, Jim Farmer, John George, Larry Johnson,
Diane Kruse, Laurie Poole, Richard Moran, James Salter, Peter Westafer