`Case 1:14-cv-02396—PGG-MHD Document 153-10 Filed 06/28/19 Page 1 of 74
`
`EXHIBIT I
`
`EXHIBIT I
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 1 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 2 of 74
`
`Appeal Nos. 2016-2509, -2510, -2511, -2512, -2575
`
`
`
`
`
`
`
`United States Court of Appeals
`
`for the
`Federal Circuit
`
`
`
`Appellant,
`
`GOOGLE INC.,
`
`– v. –
`
`NETWORK-1 TECHNOLOGIES, INC.,
`
`Cross-Appellant.
`
`CROSS-APPEAL FROM THE UNITED STATES PATENT AND TRADEMARK
`OFFICE, PATENT TRIAL AND APPEAL BOARD, IN CASE NO. IPR2005-00345
`AND APPEALS FROM THE UNITED STATES PATENT AND TRADEMARK
`OFFICE, PATENT TRIAL AND APPEAL BOARD, IN CASE NOS. IPR2015-00343,
`IPR2015-00345, IPR2015-00347 AND IPR2015-00348
`
`BRIEF OF CROSS-APPELLANT AND
`APPELLEE NETWORK-1 TECHNOLOGIES, INC.
`
`
`
`GREGORY S. DOVEL
`MATTHAEUS MARTINO-WEINHARDT
`DOVEL & LUNER, LLP
`201 Santa Monica Boulevard, Suite 600
`Santa Monica, CA 90401
`Phone:
`(310) 656-7066
`Facsimile: (310) 656-7069
`greg@dovel.com
`matthaeus@dovel.com
`
`CHARLES R. MACEDO
`JUNG S. HAHM
`AMSTER, ROTHSTEIN & EBENSTEIN LLP
`90 Park Avenue
`New York, NY 10016
`Phone:
`(212) 336-8000
`Facsimile: (212) 336-8001
`cmacedo@arelaw.com
`jhahm@arelaw.com
`
`Counsel for Cross-Appellant and Appellee Network-1 Technologies, Inc.
`
`
`
`Dated: February 8, 2017
`
`
`
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 2 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 3 of 74
`
`Certificate of Interest
`
`Counsel for Cross-Appellant and Appellee Network-1 Technologies, Inc.
`
`certifies the following:
`
`1.
`
`The full name of every party or amicus represented by me is:
`
`
`
`Network-1 Technologies, Inc.
`
`The name of the real party in interest (if the party named in the caption is not
`2.
`the real party in interest) represented by me is:
`
`
`
`Network-1 Technologies, Inc.
`
`3. All parent corporations and any publicly held companies that own 10 percent
`or more of the stock of the party or amicus curiae represented by me are:
`
`
`
`None.
`
`The names of all law firms and the partners or associates that appeared for
`4.
`the party or amicus now represented by me in the trial court or agency or are
`expected to appear in this court are:
`
`
`Dovel & Luner, LLP: Gregory S. Dovel, Sean Luner, and Matthaeus
`Martino-Weinhardt
`
`Amster, Rothstein, & Ebenstein LLP: Charles R. Macedo, Brian A. Comack,
`and Jung S. Hahm
`
`
`February 8, 2017
`
`
`
`
`
`
`
`
`
`
`Respectfully submitted,
`
`/s/ Matthaeus Martino-Weinhardt
`Matthaeus Martino-Weinhardt
`Gregory S. Dovel
`Dovel & Luner, LLP
`201 Santa Monica Blvd., Suite 600
`Santa Monica, CA 90401
`(310) 656-7066
`greg@dovel.com
`matthaeus@dovel.com
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 3 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 4 of 74
`
`
`Charles R. Macedo
`Jung S. Hahm
`Amster, Rothstein & Ebenstein LLP
`90 Park Avenue
`New York, NY 10016
`(212) 336-8000
`cmacedo@arelaw.com
`jhahm@arelaw.com
`
`Counsel for Cross-Appellant and
`Appellee Network-1 Technologies,
`Inc.
`
`2
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 4 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 5 of 74
`
`Table of Contents
`
`STATEMENT OF RELATED CASES ..................................................................... 1
`
`STATEMENT OF THE ISSUES............................................................................... 2
`
`STATEMENT OF THE CASE .................................................................................. 5
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`The claimed “non-exhaustive search.” .................................................. 5
`
`The asserted prior art. ............................................................................ 7
`
`The Board’s decisions. .......................................................................... 9
`
`Issues not appealed by Google ............................................................12
`
`Issues not decided by the Board. .........................................................12
`
`SUMMARY OF THE ARGUMENT ......................................................................14
`
`ARGUMENT ...........................................................................................................17
`
`I.
`
`The Board correctly defined a “non-exhaustive search” as “a search
`that locates a match without a comparison of all possible matches.” ...........17
`
`A.
`
`B.
`
`C.
`
`D.
`
`Because no intrinsic evidence supports Google’s assertion that
`the “all data” clause should be part of the construction of “non-
`exhaustive search,” Google must rely on extrinsic evidence of
`the meaning of “non-exhaustive search” to one of ordinary skill
`in the art. ..............................................................................................17
`
`The Board’s determination of a claim term’s ordinary meaning
`to those skilled in the art is a factual finding reviewed for
`substantial evidence. ............................................................................19
`
`The Board’s factual finding that the “all data” clause is “not
`part of the ordinary meaning” of “non-exhaustive search” is
`supported by substantial evidence. ......................................................20
`
`The specification does not redefine or disclaim the ordinary
`meaning of “non-exhaustive.” .............................................................26
`
`II.
`
`The Board’s findings that Iwamura, Ghias, and Conwell do not
`disclose a “non-exhaustive search” are supported by substantial
`evidence. ........................................................................................................29
`
`A.
`
`Iwamura does not disclose a “non-exhaustive search” of
`“extracted features.” ............................................................................29
`
` i
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 5 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 6 of 74
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`A “possible match” is a record in the database, not each
`group of data within each record that could potentially be
`compared. ..................................................................................30
`
`The construction of “non-exhaustive search” applies
`identically when the search is performed on “reference
`extracted features.”....................................................................32
`
`The Board’s finding that Iwamura does not disclose a
`“non-exhaustive search” is supported by substantial
`evidence. ...................................................................................35
`
`Google’s “reference extracted features” not “media
`work” argument fails. ................................................................39
`
`Google waived and never preserved any contention
`regarding a “non-exhaustive search” of “reference
`extracted features.”....................................................................41
`
`B.
`
`Ghias does not disclose a “non-exhaustive search.” ...........................43
`
`1.
`
`2.
`
`3.
`
`The Board’s finding that Ghias’s multiple-search feature
`does not disclose a “non-exhaustive search” is supported
`by substantial evidence. ............................................................43
`
`The Board properly declined to consider Google’s
`argument based on Ghias’s “packing all songs into one
`file” feature as a new theory presented for the first time
`in Google’s reply. ......................................................................49
`
`Ghias’s “packing all songs into one file” feature does not
`disclose a “non-exhaustive search.” ..........................................53
`
`C.
`
`Conwell does not disclose a “non-exhaustive search.” .......................54
`
`1.
`
`2.
`
`3.
`
`The Board did not focus improperly on inherent rather
`than express disclosure; it found that Conwell neither
`expressly nor inherently discloses a “non-exhaustive
`search.” ......................................................................................54
`
`The Board’s finding that Conwell does not expressly
`disclose a “non-exhaustive search” is supported by
`substantial evidence. .................................................................55
`
`Google’s “skilled artisans would have immediately
`understood” argument fails. ......................................................58
`
`4.
`
`Google’s “genus discloses species” argument fails. .................60
`
` ii
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 6 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 7 of 74
`
`CONCLUSION ........................................................................................................63
`
`
`
`
`
`
` iii
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 7 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 8 of 74
`
`Cases
`
`Table of Authorities
`
`
`Ariosa Diagnostics v. Verinata Health, Inc.,
`805 F.3d 1359 (Fed. Cir. 2015) ............................................................................29
`
`AstraZeneca LP v. Apotex, Inc.,
`633 F.3d 1042 (Fed. Cir. 2010) ............................................................................59
`
`Eli Lilly & Co. v. Zenith Goldline Pharm., Inc.,
`471 F.3d 1369 (Fed. Cir. 2006) ............................................................... 60, 61, 62
`
`In re Gleave,
`560 F.3d 1331 (Fed. Cir. 2009) ............................................................................60
`
`In re Magnum Oil Tools Int’l, Ltd.,
`829 F.3d 1364 (Fed. Cir. 2016) ............................................................................19
`
`In re Petering,
`301 F.2d 676 (CCPA 1962) ..................................................................................60
`
`In re Schaumann,
`572 F.2d 312 (CCPA 1987) ..................................................................................60
`
`Intelligent Bio-Systems, Inc. v. Illumina Cambridge Ltd.,
`821 F.3d 1359 (Fed. Cir. 2016) ................................................... 49, 50, 52, 53, 63
`
`Minkin v. Gibbons, P.C.,
`680 F.3d 1341 (Fed. Cir. 2012) ............................................................................59
`
`Perfect Surgical Techniques, Inc. v. Olympus Am., Inc.,
`841 F.3d 1004 (Fed. Cir. 2016) ..................................................................... 19, 20
`
`Phillips v. AWH Corp.,
`415 F.3d 1303 (Fed. Cir. 2005) (en banc) ............................................................19
`
`Prolitec, Inc. v. Scentair Techs., Inc.,
`807 F.3d 1353 (Fed. Cir. 2015) ............................................................................19
`
`Sage Prod., Inc. v. Devon Indus., Inc.,
`126 F.3d 1420 (Fed. Cir. 1997) ............................................................................41
`
`SAS Inst., Inc. v. ComplementSoft, LLC.,
`825 F.3d 1341 (Fed. Cir. 2016) ............................................................................19
`
`Structural Rubber Prods. Co. v. Park Rubber Co.,
`749 F.2d 707 (Fed. Cir. 1984) ....................................................................... 58, 59
`
` iv
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 8 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 9 of 74
`
`Synopsys, Inc. v. Mentor Graphics Corp.,
`814 F.3d 1309 (Fed. Cir. 2016) ............................................................................58
`
`Teva Pharms. USA, Inc. v. Sandoz, Inc.,
`135 S. Ct. 831 (2015) .................................................................................... 18, 20
`
`Thorner v. Sony Computer Entm't Am. LLC,
`669 F.3d 1362 (Fed. Cir. 2012) ............................................................................26
`
`Trintec Industries, Inc. v. Top-U.S.A. Corp.,
`295 F.3d 1292 (Fed. Cir. 2002) ............................................................................59
`
`Trustees of Columbia Univ. v. Symantec Corp.,
`811 F.3d 1359 (Fed. Cir. 2016) ............................................................................27
`
`Statutes
`
`35 U.S.C. § 316 ........................................................................................................28
`
`Rules
`
`Office Patent Trial Practice Guide, 77 Fed. Reg. 48756 .........................................50
`
`Regulations
`
`37 C.F.R. § 42 ..........................................................................................................49
`
`
`
`
` v
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 9 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 10 of 74
`
`STATEMENT OF RELATED CASES
`
`
`
`Network-1 had filed a notice of cross-appeal involving IPR2015-00345 in
`
`this case, but on February 8, 2017, filed an unopposed motion to dismiss that cross-
`
`appeal.
`
`No other appeals involving these inter partes reviews have been before this
`
`or any other appellate court.
`
`
`
`Network-1 has asserted the four patents at issue in a district court action,
`
`Network-1 Techs., Inc. v. Google, Inc., No 1:14-cv-02396-PGG-MHD (S.D.N.Y.).
`
`
`
`Network-1 has also asserted a related patent, U.S. No. 8,904,464, in another
`
`district court action, Network-1 Techs., Inc. v. Google Inc., No. 1:14-cv-09558-
`
`PGG (S.D.N.Y.). Google challenged that patent in a separate covered business
`
`method proceeding, CBM2015-00113. The Board issued a final written decision
`
`in that proceeding on October 18, 2016, determining that Google had not shown
`
`the claims of the ‘464 patent to be unpatentable. Google is appealing that decision
`
`to this court, Google Inc. v. Network-1 Techs., Inc., No. 17-1379.
`
`
`
`
`
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 10 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 11 of 74
`
`STATEMENT OF THE ISSUES
`
`1.
`
`Construction of “non-exhaustive search.” Google contends the
`
`Board’s construction of “non-exhaustive search” should include an additional
`
`clause: “and all data within all possible matches.” Google admits no intrinsic
`
`evidence supports its “all data” clause, and the Board found Google’s only
`
`extrinsic evidence (conclusory expert testimony) to be unpersuasive. After
`
`receiving contrary expert testimony supported by objective sources, the Board
`
`found that to those in the art, the ordinary meaning of “non-exhaustive search” did
`
`not include the “all data” clause. Is the Board’s factual finding supported by
`
`substantial evidence?
`
`
`
`2.
`
`Prior art findings. Under the Board’s construction, a search is
`
`exhaustive if it performs a comparison of all possible matches, i.e. all records in
`
`the dataset to be searched; it need not examine all data within each record. A
`
`search is non-exhaustive if it locates a match without a comparison of all possible
`
`matches.
`
`
`
`
`
`a.
`
`Iwamura. The Board quoted Iwamura itself and testimony by
`
`both sides’ experts that the possible matches for Iwamura’s search are reference
`
`melodies—not, as Google now contends, melody segments (i.e., smaller portions
`
`of reference melodies). The Board found based on uncontradicted evidence that
`
`2
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 11 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 12 of 74
`
`Iwamura’s search examines all reference melodies in the database and thus is
`
`exhaustive. Are the Board’s findings supported by substantial evidence?
`
`
`
`
`
`b. Waiver. Google never argued below—either in its petition,
`
`reply, or at the oral hearing—that the construction of “non-exhaustive search”
`
`applies differently to a search of “extracted features,” as it now argues on appeal.
`
`Did Google waive that argument?
`
`
`
`
`
`c.
`
`Ghias. Ghias discloses that, after performing a first search with
`
`an initial query, one may perform new query that looks for matches to the new
`
`query among the results of the first search. Quoting from Ghias itself and
`
`testimony from both sides’ experts, the Board found that the “possible matches”
`
`for the second query are the records within the dataset to be searched—i.e., records
`
`among the results to the first search—and that Ghias performs a comparison of all
`
`such records. Are the Board’s findings supported by substantial evidence?
`
`
`
`
`
`d.
`
`Improper reply. In its petition, Google argued that Ghias
`
`discloses a non-exhaustive search because it teaches an “approximate pattern
`
`matching algorithm.” In response, Network-1 demonstrated that Ghias’s
`
`“approximate pattern matching algorithm” is exhaustive, not non-exhaustive. In
`
`reply, Google then argued for the first time that Ghias also teaches packing all
`
`songs into one file. Was the Board’s determination that this was a new theory
`
`improperly raised for the first time in reply an abuse of discretion?
`
`3
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 12 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 13 of 74
`
`
`
`
`
`e.
`
`Conwell. Google admits Conwell does not inherently disclose
`
`a non-exhaustive search. Citing Conwell itself and uncontradicted testimony by
`
`Network-1’s expert, the Board found that Conwell does not expressly disclose a
`
`non-exhaustive search, because it neither says to use a non-exhaustive search, nor
`
`describes any non-exhaustive search algorithm. Is the Board’s finding supported
`
`by substantial evidence?
`
`
`
`
`
`4
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 13 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 14 of 74
`
`STATEMENT OF THE CASE
`
`1.
`
`The claimed “non-exhaustive search.”
`
`
`
`This case involves four patents: No. 8,640,179 (the ‘179 patent), No.
`
`8,205,237 (the ‘237 patent), No. 8,010,988 (the ‘988 patent), and No. 8,656,441
`
`(the ‘441 patent). Each of these four patents (collective, the “IPR patents”)
`
`involves a search that compares features from a given work with records in a
`
`reference database to locate a matching record and identify an action to be taken.
`
`Appx1508-1516 (¶¶14-27). 1
`
`
`
`Google’s appeal turns on the claimed “non-exhaustive search,” a core
`
`feature of several challenged claims of the IPR patents. The following challenged
`
`claims of the IPR patents include the element of a “non-exhaustive search”:
`
` Claims 1-3, 6, 8-14, 18, 19, 21-27, 29-31, and 34-37 of the ‘179 patent
`
` Claims 25-27, 29, and 30 of the ‘237 patent
`
` Claims 15-17, 21-28, 31-33, 38, 51, and 52 of the ‘988 patent
`
` Claims 1-3, 6, 8-14, 18, 19, 21-27, 29, and 30 of the ‘441 patent
`
`
`
`A search seeks to identify items that match a given query within a dataset of
`
`items that are possible matches, and each such item in a dataset is referred to as a
`
`“record.” Appx1541-1542 (¶¶63-64); Appx1547 (¶74).
`
`
`1 Many of the same issues and evidence were presented in each of the four
`IPRs. Where an issue appeared in multiple IPRs, the Board made similar findings
`and reached the same conclusions. Therefore, this brief for simplicity will just cite
`to one instance where a Board finding or document appears.
`
`5
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 14 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 15 of 74
`
`
`
`“A ‘non-exhaustive search’ is a search that uses an algorithm designed to
`
`locate a match without comparing the work to all records in the database.”
`
`Appx1554 (¶79). It does this by using “an intelligent algorithm to narrow the
`
`database to only a subset of potential matches.” Id. “For example, if there are 100
`
`records in a database, a non-exhaustive search could use an intelligent algorithm to
`
`exclude 75 records from the search such that only 25 would be searched during the
`
`comparison process.” Appx1554 (¶80). The specification identified specific
`
`examples of “non-exhaustive search” algorithms including “kd-trees, vantage point
`
`trees and excluded middle vantage point forests.” Appx100 (9:14-17).2 “Each of
`
`these examples uses an intelligent algorithm to narrow the database to only a
`
`subset of potential matches.” Appx1554 (¶80) (citing Appx100 (9:14-17)). Such
`
`intelligent algorithms achieve efficiency “by reducing the number of records being
`
`search[ed], e.g., by discarding clusters (buckets) of potential matches, not by
`
`reducing the length of an individual record being searched.” Appx1546 & n.13.
`
`
`
`A “non-exhaustive search” contrasts with an exhaustive search. An
`
`exhaustive (or “brute-force” search) “systematically checks whether each potential
`
`match matches the work to be identified until a match is found.” Appx1555. The
`
`
`2 The four IPR patents have identical written descriptions, because all are
`
`based on the same application and claim priority to the same provisional
`application. Where a statement appears in more than one specification, this brief
`will for simplicity just cite one instance.
`
`
`6
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 15 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 16 of 74
`
`specification includes the example of an exhaustive search that would “perform a
`
`linear search of all N entries, perhaps halting the search when the first match is
`
`found.” Appx128 (8:59-61). Similarly, the Wikipedia entry for “brute-force
`
`search” explains that “[i]n computer science, brute-force search or exhaustive
`
`search…consists of systematically enumerating all possible candidates for the
`
`solution and checking whether each candidate satisfies” the search criteria.
`
`Appx1393.
`
`2.
`
`The asserted prior art.
`
`
`
`Iwamura. Iwamura (U.S. Patent No. 6,188,010) teaches a method for
`
`searching a database of records, where each record corresponds to a “reference
`
`melody.” A “reference melody” (or “melody in the database”) corresponds to a
`
`particular portion of a classical musical work, such as “the beginning of the
`
`melody” or “one or more famous phrases in the middle of the music.” Appx1294-
`
`1295 (5:22-40; 6:35-58; 7:52-55). The data in each record in Iwamura’s database
`
`includes a “composer ID,” “title ID,” and “melody data.” Appx1294 (5:14-15;
`
`5:28-40). The “melody data” in each record may consist of a set of numerical
`
`values corresponding to pitch values for the reference melody. Appx1293-1294
`
`(6:31-54). For example:
`
`
`
`7
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 16 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 17 of 74
`
`Appx1294 (6:50-54).
`
`
`
`A melody entered by the user is then “compared” to “each reference
`
`melody” in the database. Appx1295 (7:52-54). The “reference melody” that is
`
`identified as a match “is returned as a search result.” Appx1295 (7:54-55). To
`
`make each individual comparison more efficient, Iwamura lines up peak notes (the
`
`highest pitch values) of the entered melody with peak notes of each reference
`
`melody, rather than comparing every single note within each reference melody.
`
`Appx1295 (7:52-54). The algorithm always performs a comparison to “each
`
`record” in the database. Appx1599-1601.
`
`
`
`Ghias. Ghias (U.S. Patent No. 5,874,686) relates to searching for digitized
`
`songs in a database. It teaches an “approximate pattern matching algorithm” that is
`
`applied so that the user’s input “is compared with all the songs” in the dataset to be
`
`searched. Appx1280 (5:66-6:2). What makes this algorithm more efficient is that
`
`it uses less than all data in the query (the user’s input) to perform the comparisons,
`
`such that search times do not increase as fast as the length of the query. Appx1280
`
`(6:23-35). See infra n.11.
`
`
`
`Ghias discloses that after performing a first search, “the user can perform a
`
`new query on a restricted search list consisting of songs just retrieved.” Appx1281
`
`(7:4-7). Ghias does not teach any different algorithm for performing the second
`
`search (with the new query) on the “restricted search list”; it applies the same
`
`8
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 17 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 18 of 74
`
`algorithm to a different dataset—the songs on the restricted search list. The second
`
`search thus performs a comparison to all songs on the restricted search list.
`
`Appx2111 (Google’s expert testifying regarding the second search); Appx2124
`
`(same).
`
`
`
`Conwell. Conwell (U.S. Patent No. 6,970,886) discloses associating media
`
`content such as MP3 files with identifiers and URLs. Appx1522. As relevant
`
`here, Conwell teaches locating a match using a look-up table. Appx1271 (3:43-
`
`44). Conwell does not identify any specific algorithm for performing the exact
`
`match comparison using its look-up table. Appx1523; Appx1734; Appx1738.
`
`
`
`In fact, because the indices or keys in Conwell’s look-up table are not
`
`consecutive—e.g., the index number following 112 is not 113 but 198 (Appx1269
`
`(Fig. 3))—the only evidence in the record shows that using a lookup search would
`
`require systematically checking each entry in Conwell’s look-up table. Appx1739-
`
`1742.
`
`3.
`
`The Board’s decisions.
`
`
`
`Google petitioned for IPR of certain claims of the IPR patents. Appx208-
`
`275; Appx2619-2686; Appx3353-3420; Appx4011-4078. The Board instituted
`
`review. Appx1432-1433; Appx3248-3249; Appx3945-3946; Appx4599-4600.
`
`
`
`As relevant here, the Board in its institution decision defined a “non-
`
`exhaustive search” as “a search that locates a match without a comparison of all
`
`9
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 18 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 19 of 74
`
`possible matches.” Appx3930-3931. The Board rejected the additional clause
`
`Google proposed in its petition, “and all data within all possible matches,” as not
`
`supported by the intrinsic and extrinsic record. Id. The Board found the “all data”
`
`clause “is not part of the ordinary meaning” of “non-exhaustive search.” Id.
`
`
`
`Network-1 submitted additional expert testimony supporting the Board’s
`
`construction along with its response. Appx1554-1558. In its reply, Google did not
`
`challenge this construction, nor submit any testimony or evidence on the issue.
`
`Appx67. The Board ruled in its final decision that “after considering the complete
`
`record” (Appx48), “we remain persuaded” that “non-exhaustive search” does not
`
`include “all data within all possible matches.” Appx24. It further found that
`
`Google “does not dispute the construction.” Id.
`
`
`
`Based on this construction, the Board found that neither Iwamura, Ghias, nor
`
`Conwell discloses a “non-exhaustive search.”
`
`
`
`Iwamura. The Board found that in Iwamura, each possible match (i.e., each
`
`record) corresponds to a “reference melody.” Appx33-34. The Board found that
`
`“Iwamura explicitly states that its search returns a reference melody as a search
`
`result,” and that Google’s expert, “Dr. Moulin…testifies that the possible matches
`
`in Iwamura’s search are the reference melodies in the database.” Appx34. The
`
`Board therefore rejected Google’s contention that the possible matches are
`
`“melody segments,” i.e. smaller portions of reference melodies. Appx35.
`
`10
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 19 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 20 of 74
`
`
`
`The Board also found that, as both experts agreed, Iwamura’s search
`
`performs a comparison of all reference melodies and thus discloses an exhaustive
`
`search, rather than a “non-exhaustive search.” Appx32-35.
`
`
`
`Ghias. The Board found that Ghias’s second search, the new query search of
`
`the “restricted search list” consisting of results to a first search, “still would be
`
`exhaustive of the dataset provided to the search algorithm,” because it compares all
`
`records on the restricted search list. Appx54. “A search is not non-exhaustive
`
`simply because it does not seek matches outside of the records that are available to
`
`the search algorithm.” Id.
`
`
`
`The Board also determined that Google introduced a “new argument
`
`improperly raised for the first time in the Reply”—that Ghias discloses a “non-
`
`exhaustive search” because it discloses “packing all the songs into one file.”
`
`Appx52. Citing to Google’s petition and reply, the Board found that because “this
`
`argument represents a ‘new theory’ beyond that presented in the Petition with
`
`respect to the asserted unpatentability grounds based on Ghias,” it would not
`
`consider the argument. Appx52.
`
`
`
`Conwell. The Board found that “Conwell does not explicitly disclose” a
`
`“non-exhaustive search” (Appx8), nor does so inherently (Appx8-10). It found
`
`“persuasive the testimony of Dr. Karypis,” Network-1’s expert, “that exhaustive
`
`methods may be employed to perform the database lookup in Conwell, and that
`
`11
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 20 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 21 of 74
`
`Conwell does not disclose any non-exhaustive approach.” Appx9. The Board also
`
`rejected Google’s argument that “the genus of searching is very limited, i.e.,
`
`exhaustive and non-exhaustive.” Appx9.
`
`4.
`
`Issues not appealed by Google
`
`
`
`Independent claims 1 and 5 of the ‘237 patent claim a “sub-linear time
`
`search”; independent claim 33 of the ‘237 patent likewise recites a
`
`“sublinear…search.” Appx137-138. The Board determined that neither of the
`
`asserted prior art references—Iwamura and Ghias—disclose a sub-linear search.
`
`Appx29-30; Appx37. Google does not appeal the Board’s construction of a “sub-
`
`linear search” nor its determination that the asserted prior art does not disclose that
`
`element.
`
`5.
`
`Issues not decided by the Board.
`
`
`
`The following claims challenged during IPR require a “non-exhaustive
`
`neighbor search”:
`
` Claims 1-3, 6, 8-14, 19, 21-27, 30, 31, and 34-37 of the ‘179 patent
`
` Claims 1-3, 6, 8-14, 18, 19, 21-27, 29, and 30 of the ‘441 patent
`
`
`
`Google asserted that Conwell and Ghias disclosed this element. Appx7,
`
`Appx10. Because the Board found that neither Conwell nor Ghias discloses a
`
`“non-exhaustive search” (Appx8; Appx12-13), it did not reach Network-1’s
`
`12
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 21 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 22 of 74
`
`argument that the asserted prior art also fails to disclose the claimed “neighbor
`
`search,” Appx4613-4628; Appx4649-4653.
`
`
`
`In addition, the following claims challenged during IPR require a “non-
`
`exhaustive search identifying a neighbor”:
`
` Claims 15-17, 21-28, 31-33, 38, 51, and 52 of the ‘988 patent.
`
`
`
`Google asserted that Ghias and Iwamura disclosed this element. Appx50;
`
`Appx55. Because the Board found that neither Ghias nor Iwamura discloses a
`
`“non-exhaustive search” (Appx52-54; Appx59), it did not reach Network-1’s
`
`argument that the asserted prior art also fails to disclose “identifying a neighbor,”
`
`Appx3970-3974; Appx4001-4002.
`
`
`
`
`
`13
`
`
`
`
`
`
`Case: 16-2509 Document: 30 Page: 22 Filed: 02/08/2017Case 1:14-cv-02396-PGG-MHD Document 153-10 Filed 06/28/19 Page 23 of 74
`
`SUMMARY OF THE ARGUMENT
`
`This Court should affirm the Board’s determinations that Iwamura, Ghias,
`
`and Conwell do not disclose the claimed “non-exhaustive search.” The Board’s
`
`factual findings underlying its construction of “non-exhaustive search” are
`
`supported by substantial evidence, as are its findings regarding the teachings of
`
`Iwamura, Ghias, and Conwell.
`
`1.
`
`The Board’s construction of a “non-exhaustive search” should be
`
`affirmed because it was based on factual findings supported by substantial
`
`evidence. Google’s sole contention is that the Board’s construction of “non-
`
`exhaustive search” should include an additional clause: “and all data within all
`
`possible matches.” Google admits no intrinsic evidence supports its “all data”
`
`clause, and the Board found Google’s only extrinsic evidence (conclusory expert
`
`testimony) to be unpersuasive. Network-1 submitted contrary expert testimony
`
`supported by objective sources. This is substantial evidence for th