Xerox Corporation v. Google Inc. et al

Filing 168

REDACTED VERSION of 164 Objections to Claim Construction Tutorial by Xerox Corporation. (Attachments: # 1 Exhibit A part 1, # 2 Exhibit A part 2, # 3 Exhibit B - E)(Mayo, Andrew)

Download PDF
IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE XEROX CORPORATION, Plaintiff-Counterclaim Defendant, v. GOOGLE INC., YAHOO! INC., RIGHT MEDIA INC., RIGHT MEDIA LLC, YOUTUBE, INC. AND YOUTUBE, LLC, Defendants-Counterclaim Plaintiffs. ) ) ) ) ) ) ) ) ) ) ) REDACTED PUBLIC VERSION C.A. No. 10-136-LPS PLAINTIFF XEROX CORPORATION’S RESPONSE TO DEFENDANTS’ CLAIM CONSTRUCTION TUTORIAL ASHBY & GEDDES Lawrence C. Ashby (I.D. #468) John G. Day (I.D. #2403) Lauren E. Maguire (I.D. #4261) 500 Delaware Avenue, 8th Floor P.O. Box 1150 Wilmington, DE 19899 (302) 654-1888 Attorneys for Plaintiff Xerox Corporation Of Counsel: Richard J. Stark Andrei Harasymiak CRAVATH, SWAINE & MOORE LLP Worldwide Plaza 825 Eighth Avenue New York, NY 10019-7475 (212) 474-1000 Dated: April 26, 2011 {00511788;v1} A. Defendants Improperly Use Their Tutorial as a Forum For Advocating Their Claim Construction Positions. Despite the Court’s admonition in its February 15 Amended Scheduling Order that “[t]he tutorial[s] should focus on the technology in issue and should not be used to argue claim construction contentions” (D.I. 115 ¶ 8, emphasis added), Defendants use their tutorial to argue for specific constructions of disputed claims. In their opening brief, for example, Defendants argue that a query formulated in accordance with the claims must be confined to performing a search only in a single category or “folder” of information. Defendants attempt to support their position by citing both Figure 39 of the ’979 Patent and the patent’s prosecution history. (See Defs.’ Opening Br., D.I. 141, at 4-5, 8-9.) But as discussed in Xerox’s responsive brief, the specification’s explanation of Figure 39 vitiates Defendants’ attempt to equate the asserted claims with the single sample query shown in that figure; the specification expressly teaches that the formulated query need only be “focused” on (not confined to) a category. (See 4/26/11 Xerox Resp. Br., at 10.) In addition, black-letter patent law, the language of the claims and the specification all contradict Defendants’ argument that only a single classification label may be used in query formulation. (Id. at 3-5.) Finally, Xerox never distinguished the ’979 Patent from the Rennison patent (or any other prior art) on the basis of Defendants’ incorrect construction of “restrict”. (Id. at 11-12.) Nor, as Defendants imply in their tutorial, did the examiner allow the patent on the basis of Defendants’ construction. (See Ex. A, at 28/11:15-12:08.)1 Instead, the examiner allowed the patent because Claim 1 was novel as a whole. (See Ex. B.) Nevertheless, Defendants’ tutorial consistently attempts to repackage these claim construction fallacies into seemingly neutral statements of fact. Thus, the tutorial repeatedly and 1 Citations to Defendants’ tutorial refer to Defendants’ hard copy slides (attached as Ex. A) and their accompanying video, as follows: “[hard copy page number]/[time in video]”. {00511788;v1} incorrectly implies that the technology of the ’979 Patent is confined entirely to the sample query shown in Figure 39. (See, e.g., Ex. A at 13/5:59-6:44; 16/8:01-8:22; 22/9:10-9:45.) The narration also repeatedly asserts that a classification label is “used to formulate a query to restrict a search to one assigned category”. (Id. at 8/3:47-4:11; see also id. at 16/8:01-8:22; 22/9:10-9:23; 33-35/13:21-13:45.) Finally, Defendants wrongly imply that Xerox distinguished the Rennison prior art patent during prosecution on the basis of Defendants’ proposed claim construction (id. at 28/11:15-12:08) and further wrongly imply that the examiner allowed the patent on this basis (id.). These assertions are all incorrect for the reasons above, as further explained in Xerox’s responsive claim construction brief. (See Xerox Resp. Br. at 3-5, 10-12.) Defendants also promote other incorrect claim constructions—e.g., that a query is a “request for search results” and the steps of Claim 1 must be performed before the step of Claim 2 (D.I. 141, at 14, 18)—by discussing parts of the ’979 Patent specification that have tangential relevance to the asserted claims. Thus, the tutorial devotes considerable time to a “Document Processing Flowchart” in which the system first generates and performs a query using entities and classification labels. Then, if the results obtained by using that query are “not sufficiently accurate”, the system creates an “aspect vector” using terms surrounding the entities identified in the document. (Ex. A at 36-37/13:45-14:30.) At that point, according to Defendants, “[t]he process then attempts the search again, using the more precise query”. (Id. at 37/14:25-14:30, emphasis added.) Defendants also refer to “search results” repeatedly throughout their discussion of this flowchart. (See, e.g., id. at 33-36/13:20-14:01.) The apparent purpose of this exercise is to imply that the claimed query must be a request for search results or that the query of Claims 1 and 18 must be formulated before the terms of Claim 2 are added. But the asserted claims concern query formulation only. None of the asserted claims requires performing a search of any kind, nor a request for search results, nor the return of {00511788;v1} -2- search results to users. (See Xerox Resp. Br. at 5-8.) Defendants ignore the actual teachings concerning query formulation earlier in the specification, which clearly indicate that a query is defined by its content. By focusing solely on portions of the specification that go beyond the subject matter of the asserted claims, Defendants distort the scope of those claims. Similarly, Claim 2 of the ’979 Patent says nothing about a query being formulated and executed only after a search performed in accordance with Claim 1. Claim 2 merely requires that the formulated query contain data corresponding to entities, terms and classification labels, all based on the same document content. As discussed in Xerox’s briefs, there is no reason that Claim 1 must be performed before Claim 2, and the specification expressly discloses an embodiment in which the step of Claim 2 begins before the steps of Claim 1 are complete—an embodiment Defendants ignore in their tutorial. (See Xerox Opening Br., D.I. 142, at 18-19; Xerox Resp. Br. at 20.) B. Defendants Make Incorrect Assertions in Discussing the ’979 Patent In addition to making incorrect claim construction arguments, Defendants’ tutorial makes various assertions that are inconsistent with undisputed aspects of the asserted claims and their supporting specification sections. For example, the tutorial incorrectly asserts that an entity “represents a concept about which a user may desire additional information”. (Ex. A at 9/4:10-4:25; 14/6:54-7:01.) But the ’979 Patent explicitly defines “entities” as “something recognized in a document (e.g., a person’s name, a location, a medical term, a graphics entity that may include image data, graphics data, audio data or video data) that can be in the form of an image, text, embedded data, HTML, etc.” (’979 at 6:65-7:2), and the parties adopted this definition as the agreed construction for “entity”. This definition says nothing about entities being “concepts” based on user desire for information. In that regard, the tutorial also incorrectly suggests that entity selection involves a user. (See Ex. A at 23/10:09-10:15.) In the asserted claims and corresponding specification sections, however, entity identification and {00511788;v1} -3- query formulation are performed “automatically”, i.e., without any user involvement, and Defendants have never argued that any user involvement is required by the claims. C. Defendants Mischaracterize the Relationship of the Asserted Claims of the ’979 Patent with the Prior Art. Apparently attempting to frame validity arguments, Defendants use their tutorial to mischaracterize the relationship of the ’979 Patent’s claimed technology with the prior art. Thus, Defendants begin their tutorial with a discussion of “agent-based systems” mentioned early in the specification and then segue into a discussion of systems that performed agent-based functions prior to the ’979 Patent. Defendants’ attempts to frame the asserted claims of the ’979 Patent in terms of these prior agent-based systems are misleading and inaccurate. The ’979 Patent’s specification—which it shares with 8 other issued U.S. patents, most not claiming query formulation—describes a comprehensive document management system.2 Only one portion of that system concerns formulating queries based on document content.3 (See ’979 Patent Section F.3, which also refers to Sections B.4 and F.1.) Generating queries in a novel and unique manner that combines information (“entities”) recognized in document content with categorization of the same document content is the only subject of the claims at issue in this case. In that regard, the ’979 Patent expressly teaches that its query formulation technique is intended “to improve the quality (e.g., in terms of precision recall) of information retrieval systems”—not, as Defendants would have it in their tutorial, “the asserted problem of document enrichment and management”. (See Ex. A at 4/1:46-1:52.) None 2 Indeed, the discussion of “agent-based systems” that Defendants cite in their tutorial is identical in all of those patents. 3 As the Federal Circuit has noted, “each claim does not necessarily cover every feature disclosed in the specification. When the claim addresses only some of the features disclosed in the specification, it is improper to limit the claim to other, unclaimed features.” Broadcom Corp. v. Qualcomm Inc., 543 F.3d 683, 689 (Fed. Cir. 2008) (quoting Ventana Med. Sys., Inc. v. Biogenex Labs., Inc., 473 F.3d 1173, 1181 (Fed. Cir. 2006)). {00511788;v1} -4- ASHBY & GEDDES, /s/ John G. Day Lawrence C. Ashby (I.D. #468) John G. Day (I.D. #2403) Lauren E. Maguire (I.D. #4261) 500 Delaware Avenue, 8th Floor P.O. Box 1150 Wilmington, DE 19899 (302) 654-1888 lashby@ashby-geddes.com jday@ashby-geddes.com lmaguire@ashby-geddes.com Attorneys for Plaintiff Xerox Corporation Of Counsel: Richard J. Stark Andrei Harasymiak CRAVATH, SWAINE & MOORE LLP Worldwide Plaza 825 Eighth Avenue New York, NY 10019 (212) 474-1000 Dated: April 26, 2011 {00511788;v1} -6-