IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF CALIFORNIA SAN FRANCISCO DIVISION
January 20, 2011
VASUDEVAN SOFTWARE, INC.,
INTERNATIONAL BUSINESS MACHINES CORPORATION AND ORACLE CORPORATION,
The opinion of the court was delivered by: Richard Seeborg United States District Judge
CLAIM CONSTRUCTION ORDER
Plaintiff Vasudevan Software, Inc. contends that defendants International Business Machines Corporation and Oracle Corporation (collectively "defendants") infringe three patents it owns. The 21 patents-in-suit are United States Patent Nos. 6,877,006 (the '006 Patent), 7,167,864 (the '864 Patent), and 7,720,861. The parties agree that all three patents are entitled to the same priority date 23 of July 19, 2000. On October 20, 2010, the Court held a claim construction hearing and heard 24 argument on the following claim terms: database(s), OLAP cube, and stored retrieved data. For the 25 reasons stated below, the following constructions are adopted: (1) Database: a structured set of data; (2) OLAP cube: a data structure having more than two dimensions; (3) Stored retrieved data: stored 27 retrieved data that does not consist solely of metadata.
the analysis of complex information including business data. For example, a user may wish to 4 analyze the sales results for specific items, periods, and locations. The data necessary to answer the 5 user's query may reside on multiple, geographically dispersed databases. The OLAP cube is an 6 analysis tool for capturing such data and making the information available to display to the user.
The technology in this dispute involves online analytical processing (OLAP) capabilities for
According to both parties, one prior art limitation of OLAP technology was that data residing in 8 different databases were often stored in incompatible formats or schemas. Where disparate 9 databases*fn1 were involved, an OLAP cube could not be constructed dynamically*fn2 on "live" data.
of a user's request and stored in an intermediate data repository described as a data warehouse.
Therefore, a business organization might extract data from disparate databases each night and store 13 the results in a data warehouse. In that case, any resulting OLAP cube constructed from data in the 14 warehouse would consist of "stale" data. By contrast, one aspect of the claimed invention involves 15 creating the OLAP cube dynamically in response to a user's request. Thus, data from disparate 16 databases are accessed directly to assemble the OLAP cube without going through an intermediate 17 repository of stale data.
Construction of the scope and meaning of disputed claim terms is a matter of law. Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed. Cir. 1995). In construing claim terms, the 21 Instead, the data from disparate databases needed to be transformed into a compatible format ahead
III. LEGAL STANDARD
Court focuses on how a person of ordinary skill in the art would have understood them at the time of 22 the invention. Phillips v. AWH Corp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) ("The inquiry into how 23 a person of ordinary skill in the art understands a claim term provides an objective baseline from 24 which to begin claim interpretation."). Accordingly, the Court considers "sources available to the 25 public" in order to understand how one of skill in the art would interpret the disputed claim terms.
Id. at 1314. These sources include the claims themselves, the specification as a whole, the 2 prosecution history, and extrinsic evidence shedding light on "scientific principles, the meaning of 3 technical terms, and the state of the art." Id.
most relevant evidence in construing disputed claim terms. In particular, the focus "begins and ends 6 in all cases with the actual words of the claim." Renishaw PLC v. Marposs Societa' per Azioni, 158 Courts consider the intrinsic evidence found in the patent and prosecution history to be the F.3d 1243, 1248 (Fed. Cir. 1998). The claims, however, do not stand alone. They must be read in 8 light of the entire specification to understand what the inventors invented and intended to claim. Id. 9 at 1250. The prosecution history also forms part of the intrinsic record, but often "lacks the clarity" 10 of the specification and thus may be less useful. Phillips, 415 F.3d at 1317. Finally, extrinsic evidence including expert testimony, dictionaries, and technical treatises may shed light on the state of the relevant art at the time of the invention. Markman, 52 F.3d at 980.
structured data." Defendants contend that "a structured set of data" is sufficient to define the term. 17
Thus, the parties essentially agree that a database is a structured set or collection of data.*fn3 In dispute 18 is whether the collection of data, or the data itself, must be "persistent." At oral argument, plaintiff 19 clarified its position that it is the data within the collection that must be persistent. It offered the 20 construction "collection of structured data that is persistent," as an alternative. Tr. at 33:9-19. 21
Furthermore, plaintiff claims that persistence is not a matter of the medium on which the database is 22 present (e.g., whether the data reside in memory or on hard disk), Tr. at 33:20-23, but instead that 23 persistent describes "the nature of the data that populates the databases." The data, according to 24 plaintiff, must be "at least semipermanent." Pl. Reply at 4:5-7. Therefore, plaintiff reasons, the
At oral argument, plaintiff defined structured data as that "arranged in some orderly fashion so you can use it." Tr. at 11:18-19.
Plaintiff's proposed construction for the term database is "a persistent collection of opposite of persistent data is transient data, which includes data stored temporarily while it is 2 accessed. 3
ascribe any special meaning to the term database. Instead, they disagree over whether one of skill in 5 the art at the time of the invention would understand that term, used in its ordinary sense, to 6 encompass only structured sets of persistent data. In plaintiff's view, including persistent in the 7 construction does not exclude certain databases, but instead recites a property of all databases. See 8
Based on the parties' respective arguments, they appear to agree that the patentee did not
Tr. at 15:1-3 ("[Plaintiff is] not suggesting that there's such a thing as a transient database . . . ."). It 9 admits that the patent nowhere uses the word persistent, or even transient by contrast, but suggests 10 that the term database was well understood in the art and required no explication. Defendants, for their part, insist that plaintiff's construction reads a limitation into the term that will effect a hindsight change to the scope of the patent claims.
construction. For example, the '006 Patent claim 1*fn4
requires "accessing with a computer a plurality 15 of
disparate digital databases and retrieving with a computer requested
data from such databases."
Moreover, the claim requires the further steps of accepting a user update of specific data, accessing 17 a database relevant to the update, and "updating that database dynamically on demand with the 18 specific data of the user update." See '006 Patent claim 1(e) and 1(f). According to plaintiff, if the 19 data contained in the databases were transient, then the computer could not reliably retrieve the 20 requested data or execute the user update. At oral argument, plaintiff summarized its position as 16:2-3. This argument, however, appears only to require that a database, to be useful for the 23 invention, must contain data at the time of execution of the method. It does not suggest, as 24 plaintiff's construction would have it, that the data persist for any time before or after the execution 25 of the method. Although plaintiff contends that a database not retaining data between sessions
Plaintiff argues that the context in which database appears in the claims supports its "[t]he data needs to be reliably in the source databases or the invention simply won't work." Tr. at 22 would be useless, and therefore essentially not a database, none of the claims expressly require user 2 updates to be maintained for any period. So long as the steps of the patented method have been 3 performed, then the invention as claimed has been practiced.
construction during prosecution of the patent. A patentee's communications with the Patent and 6 PTO and the inventor understood the patent." Phillips, 415 F. 3d at 1317. In describing his 8 invention, the patentee explained that the claimed OLAP cube is assembled "without accessing any 9 multidimensional database(s) of stored or persisted retrieved data assembled a priori." Pl. Exh. 14 10 at 19. Plaintiff's position, however, is not that the inventor claimed databases with persistent data, but that by definition all databases contain persistent data. If its construction is adopted, then the patentee's reference to a "database of" persisted data would necessarily be redundant. Instead, a fair 13 reading of the passage suggests that the inventor understood that databases could contain other than 14 persisted data. Without placing too much emphasis on it, the inventor's reference to persisted data 15 is at least ambiguous with respect to the definition of database. Thus, the prosecution history fails to 16 support unequivocally plaintiff's construction.
Finally, both sides introduce extrinsic evidence in arguing that their respective constructions 18 better align with the meaning of database, as understood by one of ordinary skill at the time of the 19 invention. While both dictionaries and treatises may be consulted in claim construction, the Federal 20 (explaining that "dictionaries, and especially technical dictionaries, endeavor to collect the accepted 22 meanings of terms used in various fields of science and technology"). Defendants claim that, at the 23 time, database represented essentially a non-technical term and supply a definition from a general 24 dictionary: "a structured set of data held in a computer, esp. one that is accessible or that can be 25 arranged in various ways." Def. Exh. D, Oxford American Dictionary and Language Guide (1999), 26 at 240. 27 28 evidence includes only one technical dictionary, an undated "Free On-Line Dictionary of
Plaintiff also suggests that the inventor used the term database consistent with its proposed Trademark Office form part of the intrinsic record and may be useful in comprehending "how the 7 Circuit has "especially noted" the value of technical dictionaries. See Phillips, 415 F.3d at 1318 21 Although plaintiff objects to defendants' reliance on a general dictionary, its extrinsic
Computing." Pl. Exh. 18 (defining database as "[o]ne or more large structured sets of persistent 2 data, usually associated with software to update and query the data"). Primarily, it relies on excerpts 3 from four texts published between 1999 and 2009 that describe databases as consisting of persistent 4 data. For example, plaintiff's Exhibit 15 states that "[a] database is a collection of persistent data. 5
That is, database relations are stored permanently in the computer rather than being transient data of 6 some application program." M. Levine and G. Loizou, A Guided Tour of Relational Databases
Two of plaintiff's exhibits, however, suggest that whether data are persistent may be a context-dependent concept. In one case, plaintiff refers to a 1990 database management book for 10 which the IBM Editorial Board served as consulting editors. It suggested at oral argument that this extrinsic evidence contradicts IBM's present position. Tr. at 18:17-21. The text includes the following passage: "Of course, the distinction between persistent and transient data is not a hard and 13 fast one, but depends to some extent on context (i.e., on how the data is being used). . . . A database 14 consists of some collection of persistent data that is used by the application systems of some given 15 enterprise." Pl. Exh. 17. C. J. Date, An Introduction to Database Systems (1990), at 10.
Additionally, plaintiff's Exhibit 16 states, "[a] database may be defined as a collection of persistent 17 data. The term persistent is somewhat vague, but is intended to imply that the data has a more-or-18 less independent existence or that it is semipermanent. . . . Of course, this is a very general concept. 19
Roman, Access Database (2002), at 11 (emphasis in original). In a plausible interpretation, this 21 passage at least suggests that some databases may not require persistent data, depending on their 22 purpose. Thus, although plaintiff cites some texts stating that databases consist of persistent data, 23 the extrinsic evidence does not strongly support its construction. 24
25 patent claims, specification, prosecution history, nor extrinsic evidence sufficiently support a 26 construction requiring a structured set of data to persist for a length of time beyond the steps of the 27 claimed invention in order to qualify as a database. Ultimately, the patentee is entitled to the full 28 (1999), at 4 (emphasis in original). 8
Most real-life databases consist of data that exist for a specific purpose and are thus persistent." S. 20
While many databases no doubt maintain persistent data as plaintiff claims, neither the scope of the claim terms chosen and is therefore also bound by it, as the case may be. Construction 2 of database, therefore, will read: "a structured set of data." 3
Plaintiff's construction for OLAP cube is "an online analytical processing view of data that contains at least three dimensions." Defendants propose "data structure, which can be queried, 6 having more than two dimensions and providing an online analytical processing view of the data."
B. OLAP Cube 4
The parties agree that OLAP stands for "online analytical processing" and that the OLAP cube 8 involves at least three dimensions. The parties disagree, however, over whether the OLAP view 9 itself consists of a data structure that can be queried. 10
lexicographer. See Vitronics Corp. v. Conceptronic, 90 F.3d 1576, 1582 (Fed. Cir. 1996)
("Although words in a claim are generally given their ordinary and customary meaning, a patentee 13 may choose to be his own lexicographer and use terms in a manner other than their ordinary 14 meaning, as long as the special definition of the term is clearly stated in the patent specification or 15 file history."). In particular, plaintiff asserts that courts recognize the use of "i.e." in conjunction 16 with a claim term as providing an express definition. See, e.g., Abbott Labs. v. Novopharm Ltd., 323 F.3d 1324, 1330 (Fed. Cir. 2003). In this case, the patent states that "[t]he present invention, for the 18 first time, assembles an OLAP (online analytical processing) view of data (i.e., an OLAP cube) at 19 run time, in response to a data query by a user, by accessing a plurality of incompatible source 20 databases." '006 Patent at 2:38-42. Because OLAP cube follows "i.e.," plaintiff asserts that an
Even if the patentee acted as his own lexicographer in defining OLAP cube, the meaning of "OLAP view of data" is not self-evident. In other words, OLAP view of data still must be construed 24 within the context of the patent. Based on the specification, the OLAP cube is assembled from data.
The claims contain separate and distinct steps directed to "assembling with a computer an OLAP 26 cube of retrieved data" and "displaying the OLAP cube to the user using the GUI [graphical user 27 interface]." See, e.g., '006 Patent claim 1(c) and 1(d). The claims are consistent with descriptions 28 of the invention in other sections of the specification. For example, Figure 21 of the '006 Patent
CLAIM CONSTRUCTION ORDER
For this term, plaintiff's primary argument is that the patentee acted as his own
OLAP cube must be construed as a view of data and not as a data structure.
consists of a flowchart illustrating an embodiment of the invention. It contains two distinct steps 2 labeled "assemble OLAP cube" and "display OLAP cube on GUI." The corresponding description 3 of Figure 21 explains: "Once the databases are accessible, data are retrieved from the databases in 4 step 2103. The data are used to assemble the OLAP cube in step 2104. The OLAP cube is then 5 displayed using the GUI in step 2105." '006 Patent at 12: 25-29. 6
the view is not a data structure. It points to places in the specification where the patentee refers to 8 the OLAP cube of the invention using the "view" terminology. For example, in describing a 9 database interface in Figure 6, the specification includes the following passage: "During the 10 assembly of the view, this alias file 650 is used to map the terminology of the legacy database . . . .
The use of the page metaphor also enables a structured view of all relevant information in a collated and correlated manner." '006 Patent at 6:10-17. Additionally, plaintiff contends that the 13 prosecution histories demonstrate the PTO recognized the patentee's express definition of OLAP 14 cube. In communications from the PTO during reexamination of the '006 and '864 Patents, it 15 explains a step in the invention as: "An On Line Analytical Processing (OLAP) view of data 16 23-26. These cited passages, however, suggest only that the patentee consistently used the term 18
OLAP view as a synonymous expression for OLAP cube and that the PTO adopted the patentee's 19 choice of words. They do not establish that the inventor or the PTO defined OLAP cube in a 20 manner inconsistent with a data structure. 21
retrieved during its assembly, which is an awkward construction. Even plaintiff has difficulty 23 discussing the OLAP cube without referring to its underlying data. At oral argument, plaintiff 24 explained a step in the claimed invention as "the computer is going to receive the data retrieval 25 request and then it's going to reach back to the source, the plurality of source databases, and pull in 26 responsive live data to build the OLAP cube." Tr. at 6:17-20. In describing the multidimensional 27 nature of the OLAP cube, plaintiff stated that it is "multidimensional in the sense that it can add 28 many, many, many dimensions, and for each dimension, there will be many, many, many data
While acknowledging that the OLAP cube is assembled from data, plaintiff maintains that (referred to as an OLAP cube) is assembled responsive to a data query from a user . . . ." Pl. Exhs. 17
Plaintiff's position requires that the OLAP view of data is somehow untethered from that values." Tr. at 8:23-25. Moreover, in describing the assembly of the OLAP view, it explained that 2 "[y]ou assemble a view by populating or pulling in the responsive data, arranging it, and then you 3 display it." Tr. at 40-10-12. Altogether, plaintiff's descriptions indicate that the OLAP cube is built 4 from data, it involves multiple dimensions of data values, and it is arranged before being displayed. 5
Plaintiff offers a strained argument that construing the OLAP cube as a data structure would 6 exclude the inventor's preferred embodiment from the claims. First, it explains that the serialized 7 file is itself a multidimensional database of stored retrieved data, which acts as a compass for 8 locating data in the disparate source databases. Second, in the preferred embodiment, assembling a 9 data structure requires use of the serialized file. Third, the claims require that "the OLAP cube is 10 assembled dynamically on demand without accessing a multidimensional database of stored retrieved data." See, e.g., '006 Patent claim 1(c). Therefore, the argument goes, if the OLAP cube were a data structure, it could not be assembled using the serialized file of the preferred embodiment 13 without contravening the prohibition on accessing a database of stored retrieved data.*fn5 Plaintiff's 14 apparent argument, however, is that the "OLAP view" can be assembled without the serialized file.
The result, then, is that a data structure is never assembled in any step of the claims, despite claim 16 language referring to "assembling an OLAP cube of the retrieved data." '006 Patent claim 1(c).
While this interpretation avoids the putative conflict with the preferred embodiment, it does not 18 present a reasonable alternative construction of the claims. Accordingly, the OLAP cube, or OLAP 19 view of data, consists of a data structure.
Additionally, defendants argue that the OLAP cube "can be queried." Of course, in the context of claim construction, adding the limitation requires that every embodiment of the 22 inventor's OLAP cube must be capable of being queried. In defendants' view, the drill-down and 23 pivoting capabilities, described and claimed in the patent, suggest that the OLAP cube must be 24 capable of being queried. Figure 12 and its description in the specification are directed to "dynamic 25 drill-down operations, i.e., dynamic queries for increasing levels of details." '006 Patent at 10:14-26
15. Figures 13 and 14 and their descriptions illustrate database pivoting, characterized as "rearranging the primary axis of the database view so that the look of the data can be altered." '006 3
Patent at 10:60-62. Defendants claim that these operations are examples of queries that are directed 4 to the OLAP cube. Plaintiff maintains that the original source databases are queried.
support either view. In discussing dynamic drill down, the patent explains that: "The query 1215 is 7 then executed and the results are displayed using the data display defined above. Queries 1215 can 8 also be aggregated . . . . Since these queries 1215 can act on more than a single data source, the user 9 is empowered with a new capability of data analysis over an entire set of databases . . . ." '006 10 primary databases are queried. Alternatively, if the OLAP cube itself consists of data from multiple databases, then a query directed to the OLAP cube could be characterized as indirectly "acting on" 13 multiple sources. In describing pivoting, the specification states that "the pivot is based on the 14 result of the preceding defined query." '006 Patent at 11:13-14. Thus, with respect to pivoting, 15 there is some support in the specification that a new query is not involved. 16 17 that the OLAP cube must be capable of being queried. Claim 21 of the '864 Patent is directed to 18 "[t]he method of claim1, where the displaying step provides drill down capability." Claim 21, 19 however, is necessarily narrower than the claim from which it depends. Thus, claim 1 must include 20 some embodiments of the invention where the displaying step does not provide drill down 21 capability. While disputing that the invention operates in this fashion, defendants admitted that a 22 drill-down operation could be performed by querying back to the underlying databases and 23 constructing the OLAP cube anew. Tr. 54:4-8. In this case, the specification is broad enough to 24 encompass an embodiment of the invention where the OLAP cube need not be capable of being 25 queried. In sum, the construction of OLAP cube is: "a data structure having more than two 26 dimensions."
The section of the specification cited by both parties, however, does not unambiguously Patent at 10:27-32. The reference to queries acting on more than a single source may mean that the
Defendants also argue that dependent claims encompassing drill-down and pivoting suggest
C. Stored Retrieved Data 2
"data values retrieved from the plurality of disparate digital databases."*fn6 Defendants counter that 4 the phrase consists of ordinary words and should be given its plain meaning. Alternatively, it 5 suggests the construction "data stored and retrieved before it is accessed." Although plaintiff's 6 construction is narrower than the plain meaning, it contends that stored retrieved data must be read 7 in light of its appearance within an express negative limitation in the claims. For example, in 8 the'006 Patent, the OLAP cube of the invention is assembled dynamically "without accessing a 9 multidimensional database of stored retrieved data." According to plaintiff, this negative limitation 10 serves to distinguish the patentee's dynamic OLAP cube over prior art OLAP cubes containing stale data from the plurality of disparate digital databases. Furthermore, plaintiff contends that the preferred embodiment of the invention utilizes a database of metadata to assemble the claimed 13
The final phrase in dispute is "stored retrieved data." Plaintiff argues that the phrase means
OLAP cube. Thus, unless data within the phrase "stored retrieved data" is construed as data values, 14 the preferred embodiment will be excluded from the patent claims.
be construed consistent with the language chosen by the patentee. See Chef America, Inc. v. Lamb-
The Court is not permitted to redraft the patentee's claims, even if the resulting construction renders 19 the claim inoperable. In Chef America, the appellant owned a patented process for producing a 20 dough product with a "light, flaky, crispy texture." Id. at 1372. The disputed construction involved 21 the phrase "heating the resulting batter-coated dough to a temperature in the range of about 400 22 degrees F. to 850 degrees F." Id. at 1371. The Federal Circuit held that the claim unambiguously 23 required the dough to be heated to the specified temperature range, and not the oven as the plaintiff 24 had argued. Id. at 1374. Even though the resulting claim was nonsensical and resulted in a process
As defendants urge, if a claim is "susceptible to only one reasonable interpretation," it must Weston, Inc., 358 F.3d 1371, 1374 (Fed. Cir. 2004) (internal quotation marks and citation omitted).
that failed to achieve the stated objective of the invention, the language chosen by the patentee was 2 binding. 3
exchange "data values" for "data" in the phrase "stored retrieved data." Although defendants do not 5 concede that their construction excludes the patentee's preferred embodiment, they argue that even 6 if it does, plaintiff may not fix the patentee's drafting error through the claim construction process. 7
4. In defendants' view, patentee's use of this phrase demonstrates that he knew at the time he chose 9 the claim language that metadata were included within the word data. Thus, plaintiff's present 10 suggestion to restrict the word chosen by the patentee to data values, defendants argue, constitutes an impermissible attempt to broaden the claim scope.
Despite defendants' argument that stored retrieved data consists entirely of plain English 13 words, the word "data," as commonly used and as reflected in the patent, is not devoid of ambiguity. 14
The phrase to which defendants point-metadata or data about the data-communicates that 15 metadata are "data," but they are different from "the data." One of skill in the art would understand 16 "metadata," "data," and "the data" all to have distinct meanings within the phrase that are rendered 17 unambiguous by the context. Accordingly, stored retrieved data is not a term capable of only one 18 reasonable interpretation and the Court must construe it within the context of the specification 19 including the claims. 20
be rejected. See Vitronics Corp. v. Conceptronic, 90 F.3d 1576, 1583 (Fed. Cir. 1996) (explaining 22 that the meaning of a disputed term should be interpreted such that the preferred embodiment falls 23 within the claim scope). In this case, the specification refers to the patentee's "MIDaS" program as 24 a preferred embodiment of the invention. This embodiment utilizes a preconfigured serialized file 25 that contains information about the locations and types of data in the disparate databases and acts as 26 a kind of compass during assembly of the OLAP cube. See, e.g., '006 Patent 7:36-40. ("MIDaS is 27 performing a user-defined multi-dimensional sort/analysis of the data at runtime using the definition 28 of the primary information in the serialized database interface as a compass for the operations.").
In this case, defendants contend that the same principle applies to plaintiff's request to
The specification contains explicit reference to "metadata or data about the data." '006 Patent 10:3-8
To that end, plaintiff argues that a construction excluding the preferred embodiment should
as a multidimensional database of stored retrieved metadata. The question then, is whether the 3 limitation "without accessing a database of stored retrieved data" prohibits use of the serialized file. 4
Given a reasonable alternative, one of skill in the art reading the specification would not likely 5 conclude that the inventor claimed his invention in a manner that excludes the preferred 6 embodiment. See Hoechst Celanese Corp. v. BP Chems. Ltd., 78 F.3d 1575, 1581 (Fed. Cir. 1996) 7
As the information in the serialized file consists of metadata, the file itself may be described (citation omitted). Here, using "data" to refer to data that does not include metadata is consistent 8 with its usage in the specification. Although plaintiff utilizes "data values" in its proposed 9 construction, it suggested at oral argument that it did not intend stored retrieved data to be limited to 10 data values, but that the claim term should not encompass metadata alone. Tr. at 67:3-7.
Plaintiff's proposed construction also limits the location from which the stored retrieved data are obtained to "the plurality of disparate digital databases." While the claim terms are construed in 13 light of the specification, at the same time it is impermissible to import limitations from the 14 examples into the claims. Phillips v. AWH Corp., 415 F.3d 1303, 1323 (Fed. Cir. 2005) 15
("[A]lthough the specification often describes very specific embodiments of the invention, we have 16 repeatedly warned against confining the claims to those embodiments."). In drawing the line 17 between construing terms and importing limitations, the Court focuses on how one of ordinary skill 18 in the art would understand the claim terms. Id. 19
In this case, plaintiff argues that the steps of the claims, as well as the rest of the 20 specification, indicate that the only databases from which the stored retrieved data can be retrieved 21 are the plurality of disparate digital databases. In the '006 Patent claim 1, step (b) recites accessing 22 "a plurality of disparate digital databases and retrieving with a computer requested data from such 23 databases." Step (c) subsequently entails assembling an OLAP cube "of the retrieved data. . . 24 without accessing a multidimensional database of stored retrieved data." In plaintiff's view, both 25 "the retrieved data" and "stored retrieved data" in step (c) can only be retrieved from the plurality of 26 disparate digital databases. With respect to the retrieved data, "the" serves as an express signal that 27 the phrase refers to data previously retrieved in step (b), i.e., from the plurality of disparate digital 28 databases. By contrast, stored retrieved data has no antecedent in the claim and therefore is not 2 explicitly restricted to data retrieved from the plurality of disparate digital databases. 3
Moreover, there is no support in the specification for narrowing the construction of stored
4 retrieved data in this fashion. Plaintiff raises no argument that the patentee intended a special 5 meaning in his use of the phrase, nor does it suggest that any further conflict with the preferred 6 embodiment is implicated. Restricting stored retrieved data to data retrieved from the plurality of 7 disparate digital databases would impermissibly import a limitation into the unambiguous claim 8 language. Therefore, the Court's construction of stored retrieved data is: "stored retrieved data that 9 does not consist solely of metadata."
For the reasons stated above, the Court adopts the following constructions for the disputed
Database: a structured set of data.
OLAP cube: a data structure having more than two dimensions.
Stored retrieved data: stored retrieved data that does not consist solely of metadata.
IT IS SO ORDERED