Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 1 of 20
`
`UNITED STATES DISTRICT COURT
`SOUTHERN DISTRICT OF NEW YORK
`
`
`
`
`NETWORK-1 TECHNOLOGIES, INC.,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`v.
`
`
`
`Plaintiff,
`
`GOOGLE LLC and YOUTUBE, LLC,
`
`
`
`
`
`
`
`
`Defendants.
`
`
`
`
`14 Civ. 2396 (PGG)
`
`14 Civ. 9558 (PGG)
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`PLAINTIFF NETWORK-1 TECHNOLOGIES, INC.’S
`REPLY CLAIM CONSTRUCTION BRIEF
`
`
`
`

`

`I.
`II.
`
`III.
`IV.
`
`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 2 of 20
`
`
`TABLE OF CONTENTS
`
`
`
`The Federal Circuit Decision Does Not Compel the Conclusion that “Non-
`
`Despite Having the Opportunity to Do So, Google Did Not Allege “Non-
`Exhaustive Search” Rendered the ’464 Patent Claims Indefinite and Used
`
`INTRODUCTION .............................................................................................................. 1
`“NON-EXHAUSTIVE [. . .] SEARCH” ............................................................................ 2
`A.
`The Federal Circuit’s BRI Construction Was Specific to the BRI Standard .......... 2
`B.
`Exhaustive Search” Is Indefinite ............................................................................. 4
`C.
`Network-1’s Construction in the CBM ................................................................... 5
`D.
`Comparing to All Records in the Reference Data Set ............................................ 6
`E.
`Circuit’s Broad Construction Under the Phillips Standard ..................................... 9
`F.
`the ’988 Patent ...................................................................................................... 11
`G. Fact Questions Preclude Summary Judgment of Indefiniteness ........................... 12
`“CORRELATION INFORMATION” .............................................................................. 13
`“EXTRACTED FEATURES” AND “EXTRACTING FEATURES” ............................. 14
`
`The Specification Makes Clear the Hallmark of Non-Exhaustive Search Is Not
`
`Defendants Fail to Identify Any Extrinsic Sources that Support the Federal
`
`Defendants Have a Complete Failure of Proof as to Indefiniteness for Claim 17 of
`
`i
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 3 of 20
`
`
`TABLE OF AUTHORITIES
`
`CASES
`
`Aristocrat Techs. Australia Pty Ltd. v. Int’l Game Tech.,
` 709 F.3d 1348 (Fed. Cir. 2013)................................................................................................. 15
`
`Blue Calypso, LLC v. Groupon, Inc.,
` 815 F.3d 1331 (Fed. Cir. 2016)................................................................................................... 6
`
`Convolve, Inc. v. Compaq Comput. Corp.,
` 812 F.3d 1313 (Fed. Cir. 2016)................................................................................................... 4
`
`Dow Chem. Co. v. Novo Chems. Corp. (Canada),
` 809 F.3d 1223 (Fed. Cir. 2015)................................................................................................. 13
`
`Dow Chem. Co. v. Nova Chems. Corp. (Canada),
` 803 F.3d 620 (Fed. Cir. 2015)..................................................................................................... 9
`
`Eli Lilly & Co. v. Teva Parenteral Meds., Inc.,
` 845 F.3d 1357 (Fed. Cir. 2017)................................................................................................. 12
`
`Enzo Biochem, Inc. v. Applera Corp.,
` 599 F.3d 1325 (Fed. Cir. 2010)................................................................................................... 8
`
`Ethicon Endo-Surgery, Inc. v. Covidien, Inc.,
`
` 796 F.3d 1312 (Fed. Cir. 2015).................................................................................................. 8
`
`Google LLC v. Network-1 Techs., Inc.,
` 726 F. App’x 779 (Fed. Cir. 2018) .................................................................................... passim
`
`Halliburton Energy Servs., Inc. v. M-I LLC,
` 514 F.3d 1244 (Fed. Cir. 2008)................................................................................................. 11
`
`In re Am. Acad. of Sci. Tech. Ctr.,
` 367 F.3d 1359 (Fed. Cir. 2004)............................................................................................... 2, 3
`
`Nautilus, Inc. v. Biosig Instruments, Inc.,
` 572 U.S. 898 (2014) ................................................................................................................ 5, 7
`
`O2 Micro Int’l Ltd. v. Beyond Innovation Tech. Co.,
` 521 F.3d 1351 (Fed. Cir. 2008)................................................................................................. 14
`
`Otsuka Pharm. Co. v. Torrent Pharm. Ltd.,
` 151 F. Supp. 3d 525 (D.N.J. 2015) ........................................................................................... 10
`
`
`ii
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 4 of 20
`
`
`Paice LLC v. Ford Motor Co.,
` 881 F.3d 894 (Fed. Cir. 2018)..................................................................................................... 7
`
`Personal Audio, LLC v. Google LLC,
` No. 17-cv-1751, 2019 U.S. Dist. LEXIS 41238 (D. Del. Mar. 13, 2019) .................................. 5
`
`Phillips v. AWH Corp.,
` 415 F.3d 1303 (Fed. Cir. 2005)................................................................................................. 10
`
`PPC Broadband, Inc. v. Corning Optical Commc’ns RF, LLC,
` 815 F.3d 747 (Fed. Cir. 2016)..................................................................................................... 4
`
`Presidio Components, Inc. v. Am. Tech. Ceramics Corp.,
` 875 F.3d 1369, 1374 (Fed. Cir. 2017)................................................................................. 13, 14
`
`Rembrandt Data Techs., LP v. AOL, LLC,
` 641 F.3d 1331 (Fed. Cir. 2011)................................................................................................. 13
`
`Retractable Techs., Inc. v. Becton, Dickinson & Co.,
` 653 F.3d 1296 (Fed. Cir. 2011)................................................................................................... 6
`
`Signal IP v. Am. Honda Motor Co.,
` No. 14-cv-2454, 2015 U.S. Dist. LEXIS 137339 (C.D. Cal. Apr. 17, 2015) ........................... 11
`
`Sys. Mgmt. Arts Inc. v. Avesta Techs., Inc.,
` 137 F. Supp. 2d 382, 400-02 (S.D.N.Y. 2001) ......................................................................... 12
`
`Teva Pharm. USA, Inc. v. Sandoz, Inc.,
` 789 F.3d 1335 (Fed. Cir. 2015)................................................................................................... 9
`
`
`
`iii
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 5 of 20
`
`
`I.
`
`INTRODUCTION
`
`Defendants do not provide a proposed construction for “non-exhaustive search” or
`
`“correlation information,” only asserting these terms render the claims in which they appear
`
`indefinite. To prove a claim is indefinite, Defendants have a high burden of proof—clear and
`
`convincing evidence, and must also show there are no factual issues precluding summary
`
`judgment of invalidity. They fall far short of those burdens.
`
`“Non-exhaustive search” appears in the ’988 patent’s sole asserted claim and in each
`
`asserted claim of the ’464 patent. Claim 17 of the ’988 patent further constrains non-exhaustive
`
`search: “wherein the non-exhaustive search is sublinear.” Defendants fail to address a “non-
`
`exhaustive search” that is “sublinear,” putting forward no evidence one of skill in the art would
`
`not understand claim 17’s scope. Notably, Google could have raised the indefiniteness of this
`
`term in the covered business method (“CBM”) proceeding concerning the ’464 patent before the
`
`Patent Trial and Appeal Board (“PTAB”), but chose not to. Instead, Google took the position
`
`that no construction was necessary, and applied the same construction Network-1 proposes here.
`
`Defendants posit that because the Federal Circuit held there are two possible reasonable
`
`constructions under a different standard than Phillips called Broadest Reasonable Interpretation
`
`(“BRI”), the term is indefinite. This is an incorrect statement of the law. No case so holds, and
`
`there are multiple cases finding different constructions under these different standards without
`
`holding the claims invalid for indefiniteness. Defendants also fail to prove both constructions
`
`are equally plausible under Phillips. Only Network-1’s proposed construction is correct under
`
`Phillips, in light of the intrinsic and extrinsic evidence (the latter of which was not before the
`
`Federal Circuit). The Court should reject Defendants’ indefiniteness challenge and adopt
`
`Network-1’s construction of “non-exhaustive search.”
`
`Similarly, the term “correlation information” does not render the claims in which it
`
`1
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 6 of 20
`
`
`appears indefinite. Again, Google chose not to raise this argument with the PTAB, even though
`
`it could have. Defendants fall far short of showing by clear and convincing evidence that a
`
`person of skill in the art would not have understood the scope of the claims containing this term,
`
`especially because the phrase is explained in the claims themselves. The Court should hold the
`
`term does not need construction, or adopt Network-1’s alternative construction.
`
`Finally, for “extracting features” and “extracted features,” Defendants argue against
`
`Network-1’s proposal to clarify that the parties’ agreed construction of “deriving electronic data
`
`from a work itself” can involve sampling data from or performing calculations on data from a
`
`media file. Their arguments illustrate precisely why a clarification is necessary—to avoid
`
`unnecessary disputes later in the litigation concerning the scope of these terms. Network-1’s
`
`construction is fully supported by the specification and should be adopted.
`
`II.
`
`“NON-EXHAUSTIVE [. . .] SEARCH”
`
`
`
`A.
`
`The Federal Circuit’s BRI Construction Was Specific to the BRI Standard
`
`The Federal Circuit arrived at a different construction because it applied a different
`
`standard than the standard that applies here. Defendants agree the Federal Circuit applied the
`
`BRI standard, not Phillips. See Dkt. No. 151 (Resp. Br.) at 10; see also Google LLC v. Network-
`
`1 Techs., Inc., 726 F. App’x 779, 782-86 (Fed. Cir. 2018). The Federal Circuit made no findings
`
`as to how one of skill would view “non-exhaustive search” under Phillips, and recognized its
`
`BRI construction is not necessarily the “correct construction under the framework laid out in
`
`Phillips.” Google, 726 F. App’x at 784. Under BRI, “the claims of the patent will not be read
`
`restrictively unless the patentee has demonstrated a clear intention to limit the claim scope using
`
`words or expressions of manifest exclusion or restriction.” In re Am. Acad. of Sci. Tech. Ctr.,
`
`367 F.3d 1359, 1369 (Fed. Cir. 2004) (internal quotations omitted). In other words, under BRI, a
`
`construction should not be narrowed “absent clear disclaimer in the specification.” Id.
`
`2
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 7 of 20
`
`
`In arriving at its BRI construction, the Federal Circuit first explained that “non-
`
`exhaustive searches” are searches that are not exhaustive. To find the “broadest” construction of
`
`“non-exhaustive search,” the Federal Circuit thus examined the narrowest way to view
`
`“exhaustive.” Google, 726 F. App’x at 782-84. The PTAB and Network-1 both viewed
`
`“exhaustive search,” consistent with the specification, as a search designed to compare to all
`
`possible matches (i.e., all records in a data set). Google proposed that a narrower view of
`
`exhaustive would require not only comparing to all records in a data set, but also to all data in
`
`every record, even if not necessary to determine whether a record is a match. See id. at 783.
`
`The Federal Circuit adopted Google’s suggestion as its BRI construction for two reasons.
`
`First, it was broader than Network-1’s. Id. at 784. Second, the specification did not “draw a
`
`clear line between ‘exhaustive’ and ‘non-exhaustive’ searching in terms of how much data
`
`within a record a search must consider in order to qualify as one or the other”—i.e., there was
`
`not a clear disclaimer indicating searches that compare all records but not all data within each
`
`record are exhaustive.1 Id. at 785; see In re Am. Acad. of Sci., 367 F.3d at 1369.
`
`Comparing what is “exhaustive” under Federal Circuit’s and Network-1’s construction:
`
`
`
`Comparison of all possible matches using
`sufficient data to determine whether a given
`record is a match;
`AND any additional data in a record, even
`though not necessary to determine if a record
`is a match.
`
`Federal Circuit’s
`BRI Construction
`ü
`
`Network-1’s
`Proposed Construction
`ü
`
`ü
`
`NO
`
`
`Both the Federal Circuit’s BRI construction and Network-1’s proposed construction define an
`
`search would qualify as exhaustive as long as it considers “any portion of each potential match
`—even a single bit of a long string.” Google, 726 F. App’x at 783. Respectfully, this is not what
`
`
`1 The Federal Circuit accepted Google’s characterization of the PTAB’s construction that a
`
`the PTAB meant, and certainly not what Network-1 intended. To be exhaustive, a search must
`compare to all possible records, and that comparison must be sufficient to determine whether a
`record is a match. See Mitzenmacher Reply Decl. ¶ 7.
`
`3
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 8 of 20
`
`
`exhaustive search as one that is designed to compare to all records. Under both, any search that
`
`does not compare to all records is “non-exhaustive.” See Mitzenmacher Reply Decl. ¶ 6. The
`
`only difference is that the Federal Circuit’s BRI (looking for the narrowest meaning of
`
`“exhaustive”) would also require comparing any additional data, if any, in each record, even
`
`though not necessary to determine if the record is a match to the query. See id. If there is no
`
`extraneous data in the records used in a search, the definitions merge and there is no difference
`
`between the Federal Circuit’s BRI construction and Network-1’s construction. See id.
`
`But there is no support in the record that one of skill, to a consider a search exhaustive,
`
`would require looking at extraneous data in a record, even though not necessary to determine if
`
`the record is a match. Said differently, there is no evidence one of skill in the art would consider
`
`a search that compares to all possible records to be non-exhaustive, simply because it does not
`
`compare to data not necessary to determine if the record matches. See infra §§ II.D-II.E.
`
`B.
`
`The Federal Circuit Decision Does Not Compel the Conclusion that
`“Non-Exhaustive Search” Is Indefinite
`
`Defendants urge this Court to apply a novel, unsupported indefiniteness standard—if
`
`there are two possible “reasonable” constructions, then the term is indefinite. See Resp. Br. at 2.
`
`Neither the Supreme Court nor the Federal Circuit has ever accepted the absurd standard
`
`Defendants advance. If a reasonable dispute about a term’s meaning necessitated an
`
`indefiniteness finding, there would be no reason to ever conduct claim construction. Courts have
`
`of course repeatedly held different constructions are possible under BRI and Phillips. See, e.g.,
`
`PPC Broadband, Inc. v. Corning Optical Commc’ns RF, LLC, 815 F.3d 747, 756 (Fed. Cir.
`
`2016) (different constructions are possible under BRI and Phillips); Convolve, Inc. v. Compaq
`
`Comput. Corp., 812 F.3d 1313, 1325 (Fed. Cir. 2016) (district court erred because it adopted the
`
`examiner’s “reasoning wholesale, without accounting for the differences between the broadest
`
`reasonable interpretation standard and Phillips”); Personal Audio, LLC v. Google LLC, No. 17-
`4
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 9 of 20
`
`
`cv-1751, 2019 U.S. Dist. LEXIS 41238, at *30-41 (D. Del. Mar. 13, 2019) (construing term
`
`under Phillips as limited to certain conditional algorithms, and noting that “while construing the
`
`algorithm at issue to encompass any ‘conditional’ algorithm might be acceptable under the BRI
`
`standard, it may not be appropriate under the Phillips standard”).
`
`Rather, to prove invalidity for indefiniteness, Defendants must show by clear and
`
`convincing evidence that the claim language, read in light of the intrinsic and extrinsic evidence,
`
`fails to inform a person of skill in the art with reasonable certainty of the scope of the invention.
`
`See Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898, 901, 908 (2014).
`
`As discussed above, the Federal Circuit did not make any findings as to how one of skill
`
`in the art would construe “non-exhaustive search” under Phillips. Nor did it make any findings
`
`concerning indefiniteness. Google, 726 F. App’x at 782 n.3. The Federal Circuit also noted the
`
`record before it had very little extrinsic evidence, limited to a Wikipedia entry not before this
`
`Court. Id. at 785-86. In contrast, the extrinsic evidence before the Court here overwhelmingly
`
`supports Network-1’s proposed construction. See infra § II.E.
`
`C.
`
`Despite Having the Opportunity to Do So, Google Did Not Allege
`“Non-Exhaustive Search” Rendered the ’464 Patent Claims Indefinite
`and Used Network-1’s Construction in the CBM
`
`Defendants imply Google could not raise the “non-exhaustive search” indefiniteness
`
`issue in the proceedings it initiated before the PTAB. Resp. Br. at 9. Not so. In the CBM,
`
`Google could have asserted “non-exhaustive . . . search” rendered the claims of the ’464 patent
`
`indefinite, but chose not to do so, and did not assert the term even required construction. See
`
`Hayden Reply Decl., Ex. 2 (CBM Petition). In fact, Google advanced other indefiniteness
`
`arguments, which were summarily rejected. See id., Ex. 3 (CBM Institution Decision).
`
`Moreover, in comparing prior art references to the patent claims, Google consistently
`
`stated this term had a meaning analogous meaning to the one Network-1 now advocates. Google
`
`5
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 10 of 20
`
`
`argued that a certain “algorithm is ‘non-exhaustive’ in that it does not require the query to be
`
`compared to every record in the reference data set being searched until a match is identified.”
`
`CBM Petition at 33; see also id. at 34. Google’s expert witness in the CBM, Pierre Moulin,
`
`agreed. Hayden Reply Decl., Ex. 4 (Moulin CBM Decl.) ¶ 18 (“[C]ontent recognition schemes
`
`typically employed search algorithms that increased efficiency by intelligently searching only a
`
`subset of potential matches (i.e., ‘non-exhaustive algorithms’).”); see also id. at ¶¶ 27, 67, 68, 72.
`
`PTAB proceedings were designed to streamline validity determinations, not to allow
`
`Defendants to delay this litigation and make multiple, inconsistent, alternative, serial attacks.
`
`Their advancement of arguments deemed too weak to bring to the PTAB, and inconsistent with
`
`those made at the PTAB, should be viewed with skepticism.
`
`D.
`
`The Specification Makes Clear the Hallmark of Non-Exhaustive
`Search Is Not Comparing to All Records in the Reference Data Set
`
`Under Phillips, a construction should “strive to capture the scope of the actual invention,”
`
`and a narrower construction is properly adopted when required to “tether the claims to what the
`
`specifications indicate the inventor actually invented.” Retractable Techs., Inc. v. Becton,
`
`Dickinson & Co., 653 F.3d 1296, 1305 (Fed. Cir. 2011). Here, the hallmark of the patents’
`
`searches is they limit the number of records within the set of N entries that must be compared.
`
`The construction of “non-exhaustive search” should reflect this characteristic.
`
`According to the specification, many multimedia search applications at the time of the
`
`invention performed an exhaustive search: “a linear search of all N entries [in a data set] may
`
`be performed, perhaps halting the search when the first match is found,” which “will require N/2
`
`comparisons” on average (or 45,000,000 comparisons when there are 90,000,000 entries).2 ’988
`
`
`2 The specification does not explicitly use the word “non-exhaustive.” See Resp. Br. at 13. But
`“the exact terms appearing in the claim need not be used in haec verba” in the
`specification. Blue Calypso, LLC v. Groupon, Inc., 815 F.3d 1331, 1345 (Fed. Cir. 2016)
`(internal quotations omitted).
`
`
`6
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 11 of 20
`
`
`patent at 21:23-39 (emphasis added); see Dkt. No. 148-1 (Mitzenmacher Decl.)3 ¶ 42; see also
`
`Dkt. No. 153-12 (Mitzenmacher Depo.) at 172:12-173:17. According to the patent, the claimed
`
`methods do not rely on a search of all entries (i.e., an exhaustive search) because with large
`
`databases, “efficient search to access entries” is critical. ’988 patent at 21:35-37.
`
`The specification provides several examples of non-exhaustive searches that compare to
`
`less than all of the N entries in the database. It describes binary search as an exemplary non-
`
`exhaustive method, which allows for around twenty comparisons for a database of 90,000,000
`
`entries. Id. at 21:37-39; see Mitzenmacher Decl. ¶ 47. The specification’s other examples of
`
`non-exhaustive searches all focus on improving efficiency by comparing to less than all entries.
`
`See Dkt. No. 148 (Opening Br.) at 11-13; ’988 patent at 9:23-39;4 Mitzenmacher Decl. ¶¶ 43-46;
`
`Mitzenmacher Depo. at 189:2-15. The dependent claims explicitly state these various methods
`
`are non-exhaustive searches. See, e.g., ’988 patent at claims 18-20 (“The method of claim 15,
`
`wherein the non-exhaustive search is based on [kd-trees]/[vantage point trees]/[excluded middle
`
`vantage point forest].”). Each of these techniques reduces the number of records compared, not
`
`the amount of data in each record compared. Mitzenmacher Reply Decl. ¶ 10.
`
`“One must bear in mind . . . that patents are not addressed to lawyers, or even to the
`
`public generally, but rather to those skilled in the relevant art.” Nautilus, 572 U.S. at 909
`
`
`3 Defendants criticize similarities between the Karypis and Mitzenmacher declarations. Resp.
`Br. at 16-17. That the declarations bear similarities reflect there is a common understanding of
`“non-exhaustive search” in the art, and that Network-1 has not waivered on how “non-exhaustive
`search” should be construed under Phillips, despite the protracted nature of this litigation.
`4 The patents incorporate several books and articles by reference. In reference to Duda & Hart
`and Fukunaga: “These references are incorporated herein by reference.” ’988 patent at 7:38-43.
`In reference to Yianilos articles: “Each of these references is incorporated herein by reference.”
`Id. at 9:35-39. “Such language is plainly sufficient to incorporate [each] in its entirety.” Paice
`LLC v. Ford Motor Co., 881 F.3d 894, 907 (Fed. Cir. 2018) (finding reference incorporated in its
`entirety because the patent “states that [the reference] ‘is,’ without qualification, incorporated
`into the [patent] application ‘by this reference’—i.e., the reference contained in the sentence”).
`But even if these references are viewed as extrinsic evidence, they still support Network-1’s
`proposed construction of “non-exhaustive search.” Notably, the Federal Circuit did not consider
`any of these materials in its analysis. See generally Google, 726 F. App’x 779.
`
`7
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 12 of 20
`
`
`(internal quotations omitted). In addition, “the definiteness requirement must take into account
`
`the inherent limitations of language. Some modicum of uncertainty, the [Supreme] Court has
`
`recognized, is the price of ensuring the appropriate incentives for innovation.” Id. (internal
`
`citation and quotations omitted). A skilled artisan would understand with reasonable certainty
`
`what searches are within the claim scope from the specification’s focus on efficient searching
`
`and its exemplary searches—all of which compare a query to less than all records in the data set.
`
`Mitzenmacher Reply Decl. ¶¶ 8-10. “Because the intrinsic evidence here provides a general
`
`guideline and examples sufficient to enable a person of ordinary skill in the art to determine [the
`
`scope of the claims], the claims are not indefinite . . . .” Enzo Biochem, Inc. v. Applera Corp.,
`
`599 F.3d 1325, 1335 (Fed. Cir. 2010) (internal citation and quotations omitted).
`
`Defendants make much of the Federal Circuit’s finding that the specification does not
`
`explicitly address how much data within each record must be compared in any search it
`
`describes. But a skilled artisan would recognize there is no reason the specification should do
`
`so—for any search, the amount of data in a given record to be compared is the amount needed to
`
`determine if the record matches the query conditions or not. Mitzenmacher Reply Decl. ¶ 9; see
`
`Ethicon Endo-Surgery, Inc. v. Covidien, Inc., 796 F.3d 1312, 1316-22 (Fed. Cir. 2015) (reversing
`
`indefiniteness finding because court failed to take into account a skilled artisan’s understanding
`
`of the specification). A skilled artisan would know searches, including those that compare a
`
`query to all records, need only compare sufficient data in the record to determine whether the
`
`record is a match. Mitzenmacher Reply Decl. ¶ 9. He would thus know the patent is drawing a
`
`distinction between searches that compare to all records (exhaustive) and those that do not (non-
`
`exhaustive). Id. The amount of data in each record compared is not the relevant distinction; in
`
`each case, the comparison is to sufficient data to determine if the record is a match. Id. The
`
`relevant distinction as described in the specification is the number of records compared. Id.
`
`8
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 13 of 20
`
`
`Finally, the cases Defendants cite are inapposite. In Teva Pharmaceuticals USA, Inc. v.
`
`Sandoz, Inc., the claims recited “molecular weight,” but the parties agreed there were three
`
`entirely different ways to measure molecular weight and the specification provided no guidance
`
`on which method to use. 789 F.3d 1335, 1344-45 (Fed. Cir. 2015); see Resp. Br. at 15-16.
`
`Similarly, Defendants also cite Dow Chemical Co. v. Nova Chemicals Corp. (Canada), 803 F.3d
`
`620, 630 (Fed. Cir. 2015) for the premise that “the patent and prosecution history must disclose a
`
`single known approach or establish that, where multiple known approaches exist, a person
`
`having ordinary skill in the art would know which approach to select.” Resp. Br. at 16. But
`
`here, all of the specification’s exemplary non-exhaustive searches improve search efficiency by
`
`not comparing a query to every record in the data set. See Mitzenmacher Reply Decl. ¶¶ 8-10.
`
`As explained here, all discussion in the specification is about distinguishing a “search of all N
`
`entries” from more efficient searches that do not require comparing to all N entries. The same is
`
`true of the extrinsic references before the Court. See infra § II.E.
`
`E.
`
`Defendants Fail to Identify Any Extrinsic Sources that Support the
`Federal Circuit’s Broad Construction Under the Phillips Standard
`
`The extrinsic evidence before this Court (none of which was before the Federal Circuit)
`
`supports Network-1’s construction, and not the Federal Circuit’s BRI. The extrinsic references
`
`define exhaustive searches as those that compare to all records in the data set. See Opening Br.
`
`at 13-15. None (1) indicate exhaustive search requires comparison to all data (including
`
`extraneous data) in each record, or (2) describe as non-exhaustive any search that compares to
`
`every record in the data set. See Resp. Br. at 16-20; Hayden Reply Decl., Ex. 1 (Storer Depo.) at
`
`89:16-93:14, 131:24-132:22 (unable to identify any such reference); id. at 125:8-129:2
`
`(dictionary definition “does not explicitly say” whether an exhaustive search must “consider all
`
`data within all possible matches”). Defendants also offer no expert testimony that the Federal
`
`Circuit’s BRI is the ordinary meaning under Philips. See Mitzenmacher Reply Decl. ¶ 11. The
`9
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 14 of 20
`
`
`Federal Circuit considered limited extrinsic evidence (a lone Wikipedia entry), and none
`
`concerning treatment of “extraneous” data. See Google, 726 F. App’x at 785-86.
`
`Defendants’ attacks on the extrinsic references fall flat. Defendants make much of
`
`Orwant’s statement that “the definition of exhaustive search is vague.” See Resp. Br. at 18
`
`(quoting Dkt. No. 148-20 (Orwant) at 183). But the “vagueness” Orwant refers to is there are
`
`different ways to perform exhaustive searches—there are numerous exhaustive search
`
`algorithms. See Orwant at 183; Mitzenmacher Reply Decl. ¶ 12. And Orwant’s colloquial
`
`example, finding the most interesting item in a large museum, does not indicate one would
`
`examine all “data” in each item—there is no reason to examine every brushstroke of a painting if
`
`one is not interested in paintings as a whole. See Orwant at 180. Similarly, a skilled artisan
`
`would not design an exhaustive search to do the unnecessary work of continuing to examine data
`
`in a record if it already determined the record was not a match. Mitzenmacher Reply Decl. ¶ 13.5
`
`Google’s own patents also use “exhaustive search” consistent with Network-1’s proposed
`
`construction. See Opening Br. at 14-15. Defendants posit these patents are not relevant because
`
`the terms are used in the specifications and not the claims. Resp Br. at 20. But just like journal
`
`articles, patents using a claim term in their specifications are “evidence external to the patent and
`
`prosecution history.” Phillips v. AWH Corp., 415 F.3d 1303, 1317 (Fed. Cir. 2005).6
`
`And as with the specification, a skilled artisan would have understood why these extrinsic
`
`references do not state how much data within each record is compared—search algorithms are
`
`not designed to examine more data in a record than necessary. Mitzenmacher Reply Decl. ¶ 16.
`
`
`5 Dr. Storer makes several arguments that are not raised in Defendants’ brief. These arguments
`are not part of the record and therefore are not addressed in this brief. However, Dr.
`Mitzenmacher does address some of those arguments in his reply declaration. See Mitzenmacher
`Reply Decl. ¶¶ 14-17.
`6 The case Defendants cite to imply otherwise does not support this position. There, the Court
`rejected that a term should be found definite because other courts have been able to construe that
`term in other unrelated patents. See Otsuka Pharm. Co. v. Torrent Pharm. Ltd., 151 F. Supp. 3d
`525, 545, n.36 (D.N.J. 2015). That is not a position Network-1 takes here.
`10
`
`

`

`Case 1:14-cv-02396-PGG-MHD Document 158 Filed 07/19/19 Page 15 of 20
`
`
`Defendants fail to prove the extrinsic evidence supports the Federal Circuit’s BRI
`
`construction, such that one of skill in the art would not be able to ascertain with reasonable
`
`certainty the claim scope. The extrinsic references, consistent with the specification, describe
`
`searches in terms of whether they compare to all records (or “all N entries”) in the data set.
`
`There is also expert testimony that a skilled artisan would understand the extrinsic references to
`
`be referring to comparing to sufficient data to determine whether the record is a match. See, e.g.,
`
`id. ¶¶ 11-16. Under Phillips, Network-1’s proposed construction is how one of skill would
`
`understand “non-exhaustive search,” in light of both the intrinsic and extrinsic evidence.
`
`F.
`
`Defendants Have a Complete Failure of Proof as to Indefiniteness for
`Claim 17 of the ’988 Patent
`
`As a matter of law, Defendants have not met their burden of proving the sole asserted
`
`claim of the ’988 patent is indefinite. Even if the term “non-exhaustive search” did render
`
`unasserted claim 15 indefinite (it does not), it does not follow that dependent claim 17 is
`
`indefinite; claim 17 narrows “non-exhaustive search” by specifying its “non-exhaustive search”
`
`is “sublinear.” See ’988 patent at claim 17; see also Signal IP v. Am. Honda Motor Co., No. 14-
`
`cv-2454, 2015 U.S. Dist. LEXIS 137339, at *101-106 & n.15 (C.D. Cal. Apr. 17, 2015) (finding
`
`dependent cla