1

2
3
4
5
6
IN THE UNITED STATES DISTRICT COURT
7
FOR THE NORTHERN DISTRICT OF CALIFORNIA
8
9
DATA RETRIEVAL TECHNOLOGY, LLC, No C 08-5481 VRW
10
Plaintiff, ORDER
For the Northern District of California

11
United States District Court

v
12
SYBASE, INC and INFORMATICA
13 CORPORATION,
14
Defendants.
15
/
16
17
18 This suit focuses on United States Patent Nos 6,026,392
19 (“’392 Patent”) and 6,631,382 (“’382 Patent”). Both patents
20 describe computer-implemented methods for retrieving information
21 stored in databases without the need for human analysis of the
22 source data. Data Retrieval Technology LLC (“DRT”) alleges that
23 Sybase Incorporated (“Sybase”) infringes claim 2 of the ’382 Patent
24 and that Informatica Corporation (“Informatica”) infringes claim 1
25 of the ’392 Patent. Doc ##7, 47. This order addresses the claim
26 construction of the ’392 and ’382 Patents and GRANTS Informatica’s
27 motion (Doc #108) for summary judgment of invalidity of the ’392
28 Patent.
 1 I
2 Claim construction is an issue of law and it begins “with
3 the words of the claim.” Nystrom v TREX Co, Inc, 424 F3d 1136,
4 1142 (Fed Cir 2005). Claim terms are “generally given their
5 ordinary and customary meaning” unless the patent specification or
6 file history contains a clearly stated “special definition.”
7 Vitronics Corp v Conceptronic, Inc, 90 F3d 1576, 1582 (Fed Cir
8 1996). Moreover, “the ordinary and customary meaning of a claim
9 term is the meaning that the term would have to a person of
10 ordinary skill in the art in question at the time of the
For the Northern District of California

11 invention.” Phillips v AWH Corp, 415 F3d 1303, 1313 (Fed Cir
United States District Court

12 2005). Such a person understands the claim term by “looking at the

13 ordinary meaning in the context of the written description and the
14 prosecution history.” Medrad, Inc v MRI Devices Corp, 401 F3d
15 1313, 1319 (Fed Cir 2005). References to “preferred embodiments”
16 in the written description and prosecution history are not claim
17 limitations. Laitram Corp v Cambridge Wire Cloth Co, 863 F2d 855,
18 865 (Fed Cir 1988).
19 It is appropriate “for a court to consult trustworthy
20 extrinsic evidence to ensure that the claim construction it is
21 tending to from the patent file is not inconsistent with clearly
22 expressed, plainly apposite and widely held understandings in the
23 pertinent technical field.” Pitney Bowes, Inc v Hewlett-Packard
24 Co, 182 F3d 1298, 1309 (Fed Cir 1999). Extrinsic evidence
25 “consists of all evidence external to the patent and prosecution
26 history, including expert and inventor testimony, dictionaries, and
27 learned treatises.” Phillips, 415 F3d at 1317. All extrinsic
28 evidence should be evaluated in light of the intrinsic evidence.

2
 1 Id at 1319.
2 With these principles in mind, the court now turns to the
3 construction of the disputed claim language of the ’392 and ’382
4 Patents.
5
6 II
7 Both the ’392 and the ’382 Patent describe “a computer-
8 implemented system which is able to retrieve information stored in
9 one or more of a number of different sources and which may be in
10 any number of different formats and/or provide reports and analysis
For the Northern District of California

11 based on the information.” ’392 Patent at 1:10-14; ’382 Patent at

United States District Court

12 1:11-17. The parties agree that the terms of both patents should
13 be construed identically, as the patents are closely related. Doc
14 #104 at 7; Doc #106 at 6. The parties dispute the meaning of two
15 terms: “data source,” which appears in both disputed claims, and
16 “optimization,” which appears only in claim 1 of the ’392 Patent.
17
18 A
19 The court first turns to the term “data source” found
20 both in claim 1 of the ’392 Patent and claim 2 of the ’382 Patent.
21 Claim 1 of the ’392 patent states a method claim:
22 A computer-implemented method comprising;
23 providing a driver which issues instructions for
accessing data which may be stored in either
24 of first and second different data sources,
said driver containing program instructions
25 configured for use in connection with said
first data source; and
26 using said driver to automatically obtain
information about the data structure of said
27 first data source without the need for human
analysis of the first data source by accessing
28 content of information stored in said first

3
 1 data source wherein said information about the
data structure leads to optimization of a new
2 database in which information from said first
database is to be stored.
3
’392 Patent at 19:33-45.
4
Claim 2 of the ’382 patent states a method claim:
5
A computer-implemented method usable in connection
6 with accessing data which may be stored in either
of first and second different data sources, the
7 method comprising:
8 a first step in providing a first driver containing
program instructions configured for use in
9 connection with said first data source in the
absence of said first driver being configured
10 for use in connection with said second data
source;
For the Northern District of California

11 a second step for using said first driver to

United States District Court

automatically obtain first information about

12 the data structure of said first data source
without the need for human analysis of the
13 first data source by automatically accessing
content of information stored in said first
14 data source;
using said first information to define a structure
15 for a first database different from said data
sources wherein said first database did not
16 exist before said step for using said first
driver to automatically obtain first
17 information; and
generating at least a first report based on
18 information in said first database.
19 ’382 Patent at 23:25-45.
20 DRT proposes that “data source” be construed as “a
21 collection of computer readable information such as a relational
22 database, a flat file, and the like.” Doc #128 at 6. Sybase and
23 Informatica propose a similar construction: “a collection of
24 computer readable information such as a flat file, a relational
25 database, a spreadsheet and the like.” Id. The dispute thus
26 centers on whether a spreadsheet should be included as an example
27 of a collection of computer readable information or whether a flat
28 file encompasses a spreadsheet. Doc #104 at 15; Doc #106 at 24.

4
 1 DRT’s objection to the inclusion of “spreadsheet” as an
2 example, as explained by counsel at the claim construction hearing,
3 appears to be based solely on the concern that it would be
4 redundant given that “flat file” already includes a spreadsheet.
5 Doc #136 at 5-6. Sybase and Informatica, on the other hand, want
6 to make clear that “data source” includes a spreadsheet and
7 foreclose any future debate on the point. Id at 9. Sybase and
8 Informatica point to language in the specification of the ’392
9 Patent listing as examples of data sources “flat file source data,
10 hierarchical databases, relational databases, spreadsheets, and the
For the Northern District of California

11 like.” ’392 Patent at 9:46-47. More importantly, the parties

United States District Court

12 agree that “data source” does in fact include a spreadsheet. For

13 this reason, and to avoid any further confusion or debate on the
14 issue, the court construes the term “data source” as “a collection
15 of computer readable information such as a flat file, a relational
16 database, a spreadsheet and the like.”
17
18 B
19 DRT and Informatica dispute the meaning of
20 “optimization,” found in claim 1 of the ’392 Patent. DRT asserts
21 that the term should be construed as “providing a database whose
22 performance with respect to a given characteristic (e g speed or
23 flexibility of output) is superior to the performance of the data
24 source with respect to that characteristic.” Doc #128 at 2.
25 Informatica does not propose a construction and instead moves for
26 summary judgment of indefiniteness pursuant to 35 USC § 112 ¶ 2,
27 which requires that a claim “particularly point[] out and
28 distinctly claim[] the subject matter.” Doc #108.

5
 1 1
2 In reviewing a summary judgment motion, the court must
3 determine whether genuine issues of material fact exist, resolving
4 any doubt in favor of the party opposing the motion. “[S]ummary
5 judgment will not lie if the dispute about a material fact is
6 ‘genuine,’ that is, if the evidence is such that a reasonable jury
7 could return a verdict for the nonmoving party.” Anderson v
8 Liberty Lobby, 477 US 242, 248 (1986). “Only disputes over facts
9 that might affect the outcome of the suit under the governing law
10 will properly preclude the entry of summary judgment.” Id. The
For the Northern District of California

11 burden of establishing the absence of a genuine issue of material

United States District Court

12 fact lies with the moving party. Celotex Corp v Catrett, 477 US
13 317, 322-23 (1986). Summary judgment is granted only if the moving
14 party is entitled to judgment as a matter of law. FRCP 56(c).
15 The nonmoving party may not simply rely on the pleadings,
16 however, but must produce significant probative evidence supporting
17 its claim that a genuine issue of material fact exists. TW Elec
18 Serv v Pacific Elec Contractors Ass’n, 809 F2d 626, 630 (9th Cir
19 1987). The evidence presented by the nonmoving party “is to be
20 believed, and all justifiable inferences are to be drawn in his
21 favor.” Anderson, 477 US at 255. “[T]he judge’s function is not
22 himself to weigh the evidence and determine the truth of the matter
23 but to determine whether there is a genuine issue for trial.” Id
24 at 249.
25
26 2
27 “[T]he limits of a patent must be known for the
28 protection of the patentee, the encouragement of the inventive

6
 1 genius of others and the assurance that the subject of the patent
2 will be dedicated ultimately to the public.” Markman v Westview
3 Instruments, Inc, 517 US 370, 390 (1996) (quoting General Elec Co v
4 Wabash Appliance Corp, 304 US 364, 369 (1938)). The § 112 ¶ 2
5 definiteness requirement “focuses on whether the claims * * *
6 adequately perform their function of notifying the public of the
7 [scope of the] patentee’s right to exclude.” Honeywell
8 International, Inc v International Trade Commissionn, 341 F3d 1332,
9 1338 (Fed Cir 2003) (quoting S3 Inc v nVIDIA Corp, 259 F3d 1364,
10 1371-72 (Fed Cir 2001)).
For the Northern District of California

11 Because Informatica moves for summary judgment of

United States District Court

12 indefiniteness, it must show by clear and convincing evidence that

13 the claim is “not amenable to construction” or “insolubly
14 ambiguous.” Haemonetics Corp v Baxter HealthCare Corp, 607 F3d
15 776, 783 (Fed Cir 2010) (internal citations omitted). A claim is
16 indefinite only if “a person of ordinary skill in the art could not
17 determine [its] bounds.” Halliburton Energy Services, Inc v M-I
18 LLC, 514 F3d 1244, 1249 (Fed Cir 2008). A person of ordinary skill
19 in the art may rely on the claim language, the specification, the
20 prosecution history and his or her knowledge to determine the
21 bounds of the claim. Id at 1249-1250. Although the court may look
22 to extrinsic evidence if necessary to its understanding of the
23 patent, such evidence in general, and expert testimony in
24 particular, is disfavored. See Texas Digital Systems, Inc v
25 Telegenix, Inc, 308 F3d 1193, 1212 (Fed Cir 2002).
26 The court should not hold the claim to be indefinite
27 simply because “it poses a difficult issue of claim construction;
28 if the claim is subject to construction, i e, it is not insolubly

7
 1 ambiguous, it is not invalid for indefiniteness.” Bancorp
2 Services, LLC v Hartford Life Ins Co, 359 F3d 1367, 1372 (Fed Cir
3 2004) (citing Honeywell, 341 F3d at 1338-39). Evaluating the
4 indefiniteness question in this fashion serves to protect the
5 statutory presumption of patent validity. Bancorp, 359 F3d at 1372
6 (citing 35 USC § 282); see Honeywell, 341 F3d at 1338-39. When the
7 question of indefiniteness is close, it should be resolved in favor
8 of the patentee. Bancorp, 359 F3d at 1372 (citing Exxon Research &
9 Engineering Co v United States, 265 F3d 1371, 1375 (Fed Cir 2001)).
10
For the Northern District of California

11 3
United States District Court

12 DRT first argues that Informatica is procedurally barred

13 from contending that “optimization” is indefinite because
14 Informatica did not claim indefiniteness as part of its invalidity
15 contentions filed on July 23, 2009. Doc #121 at 7-11. DRT asserts
16 that Informatica should not now be permitted to amend its
17 infringement contentions, because any amendment would not cure the
18 prejudice DRT faces resulting from Informatica’s delay. Id.
19 Informatica moved to amend its infringement contentions on August
20 20, 2010. Doc #130. Regardless of Informatica’s delay in
21 asserting indefiniteness, the court is required to construe the
22 claims at issue and declare them indefinite if they cannot be
23 properly construed. Furthermore, DRT has had ample opportunity to
24 argue in favor of its proposed claim construction and has not
25 suffered any apparent prejudice from Informatica’s assertion of
26 indefiniteness at this stage of the proceedings. Accordingly,
27 Informatica’s motion to amend its infringement contentions, Doc
28 #130, is DENIED AS MOOT.

8
 1 The parties agree that DRT’s proposed expert, Paul
2 Bertucci, has the level of ordinary skill in the art. Doc #125 at
3 5. Mr Bertucci has defined a person of ordinary skill in the art
4 as “[o]ne with experience in database,” Doc #107-7 at 9, and
5 subsequently as “a person with a Bachelor’s in a computer related
6 field such as electrical engineering or computer science, and one
7 to two years of experience in software development or the
8 equivalent work experience.” Doc #123 at 4. Under either
9 definition, a person of ordinary skill in the art would be unable
10 to determine the bounds of claim 1 of the ’392 Patent.
For the Northern District of California

11 DRT’s proposed construction would find a database to be

United States District Court

12 “optimiz[ed]” when its performance “with respect to a given

13 characteristic” is superior to that of the data source from which
14 it was created. Doc #128 at 2 (emphasis added). The fundamental
15 flaw here is that no such characteristic is in fact given. Thus,
16 if one could identify a single characteristic of a database that is
17 superior to that of the original source, the database would be
18 “optimized” within the meaning of the claim. As DRT’s expert Paul
19 Bertucci stated in his deposition, “[t]here are hundreds of * * *
20 potential characteristics” of a database that may be improved. Doc
21 #107-7 at 11. Further, Mr Bertucci admitted in his second
22 declaration that “the process of optimization involves tradeoffs ——
23 i e, one design/configuration may provide improved performance for
24 certain types of queries while at the same time, diminishing
25 performance for other types of queries and database operations.”
26 Doc #105 at 4.
27 It is clear that DRT’s proposed construction of
28 “optimization” fails to define any meaningful limitation. Claim 1

9
 1 describes a method for using a computer to access a data source
2 automatically and create a new database using the data from the
3 source. ’392 Patent at 19:33-38. The claim further requires that
4 the new database be “optimiz[ed],” id, but under the proposed
5 construction any such database could be considered optimized. One
6 must only identify one of the hundreds of possible characteristics
7 of the new database that performs better than that of the original
8 data source. And because optimization necessarily involves
9 tradeoffs, a database that is demonstrably inferior to the original
10 data source with respect to the most important characteristics
For the Northern District of California

11 would still be superior with respect to some characteristic and

United States District Court

12 therefore “optimized.” A competitor would have no way to know

13 whether a process for transforming a data source into a new
14 database led to “optimization” and would be subject to an
15 infringement suit if the patent holder could locate a single
16 characteristic of the database that is superior to that of the data
17 source. Accordingly, DRT’s proposed construction is indefinite.
18 The court is unable to adopt any narrowing construction
19 of “optimization” that is consistent with the language of the
20 claim. DRT notes that “the patent provides a number of examples of
21 characteristics for which a database may be optimized” and suggests
22 that “the claim term is further limited when read in light of the
23 specification by the examples provided by the specification.” Doc
24 #121 at 16. This suggestion fails to grasp “the distinction
25 between using the specification to interpret the meaning of a claim
26 and importing limitations from the specification into the claim.”
27 Phillips, 415 F3d at 1323. Although the specification can be used
28 to help define unclear claim terms, it cannot be used to limit

10
 1 them. The patent specification provides examples of “different
2 systems which are optimized for different purposes,” specifically
3 those “optimized for data entry or storage vs speed or flexibility
4 or data analysis and reporting, optimized for accounting data vs
5 company data, and the like.” ’392 Patent at 1:34-37. But DRT does
6 not argue that these examples actually define “optimize,” which, as
7 discussed above, means to improve performance with respect to any
8 of a large number of possible characteristics.
9 DRT points out that the Random House Personal Computer
10 Dictionary defines “optimization” as “[i]n programming, to fine
For the Northern District of California

11 tune a program so that it runs more quickly or takes up less

United States District Court

12 space.” Doc #121 at 16. This does not support DRT’s proposed
13 definition of “optimization” as improving performance with respect
14 to any characteristic, and in fact further undercuts its argument
15 that the term “optimization” has a definite meaning.
16 Claim 1 is the only claim from the ‘392 Patent asserted
17 by DRT and is “insolubly ambiguous.” Informatica’s motion for
18 summary judgment of indefiniteness is supported by clear and
19 convincing evidence and is accordingly GRANTED. Doc #108.
20
21
22
23 IT IS SO ORDERED.
24
25 VAUGHN R WALKER
United States District Chief Judge
26
27
28

11