`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`NETFLIX, INC.,
`
`Petitioner
`
`v.
`
`UNILOC 2017 LLC,
`
`Patent Owner
`
`IPR2020-00041
`
`PATENT 8,407,609
`
`PATENT OWNER SUR-REPLY TO PETITION
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`
`TABLE OF CONTENTS
`
`Exhibits ..................................................................................................................... iv
`
`I.
`
`II.
`
`INTRODUCTION ........................................................................................... 1
`
`NETFLIX FAILS TO PROVE UNPATENTABILITY FOR THE
`CHALLENGED CLAIMS .............................................................................. 1
`
`A.
`
`Netflix fails to prove that its proposed Davis-Choi
`combination renders obvious limitations directed to the “timer
`applet,” such as those recited in the “receiving” step (Ground
`1) ............................................................................................................ 1
`
`Netflix concedes that Davis fails to disclose the “timer applet,”
`1.
`such as those in the “receiving step” ..................................................... 1
`
`Choi fails to disclose the “time applet,” such as those in the
`2.
`“receiving step” ..................................................................................... 2
`
`Netflix’s belated new theories regarding the proposed Davis-
`3.
`Choi combination constitute an improper attempt to circumvent the
`Board’s rules and precedent .................................................................. 4
`
`4.
`
`5.
`
`Netflix’s belated new theory is factually and legally defective .. 8
`
`No motivation to combine Davis and Choi ..............................12
`
`Petitioner fails at least to prove its proposed Siler-Davis
`combination renders obvious limitations directed to the “timer
`applet,” such as those recited in the “receiving” step (Ground
`2) .......................................................................................................... 14
`
`The Petition fails at least to prove its proposed Siler-Davis
`combination renders obvious “wherein each provided
`webpage causes corresponding digital media presentation data
`to be streamed from a second computer system distinct from
`the first computer system directly to the user’s computer
`
`B.
`
`C.
`
`ii
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`independent of the first computer system” (Ground 2)....................... 17
`
`III. CONCLUSION .............................................................................................. 21
`
`CERTIFICATE OF COMPLIANCE .......................................................................... i
`
`CERTIFICATE OF SERVICE ................................................................................. ii
`
`
`
`iii
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`
`UNILOC’S EXHIBITS
`
`EXHIBITS
`
`2001
`
`File History of ’609 Patent
`
`2002
`
`Claim Construction Memorandum and Order, Uniloc 2017
`
`LC v. Google LLC, Case 2:18-CV-00502-JRG-RSP (E.D.
`
`Tex.), Dkt. 149 (Jan. 20, 2020)
`
`2003
`
`Claim Construction Ruling, Uniloc 2017 LLC v. Netflix, Inc.,
`
`SACV 18-2055-GW-DFMx, Dkt. 138 (C.D. Cal. Mar. 9,
`
`2020)
`
`
`
`
`
`iv
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`
`I.
`
`INTRODUCTION
`
`Uniloc 2017 LLC (“Uniloc”) submits this Sur-Reply in further response to
`
`Petition IPR2020-00041 for Inter Partes Review (“Petition”) of United States Patent
`
`No. 8,407,609 (“the ’609 patent”) filed by Netflix, Inc. (“Netflix”).1 As discussed
`
`below and in Uniloc’s Response, the Board should deny the Petition in its entirety,
`
`as Netflix failed to carry its burden of proof that Claims 1-3 of the ’609 Patent are
`
`unpatentable. 35 U.S.C. § 316(e).
`
`II. NETFLIX FAILS TO PROVE UNPATENTABILITY FOR THE
`CHALLENGED CLAIMS
`
`A. Netflix fails to prove that its proposed Davis-Choi combination
`renders obvious limitations directed to the “timer applet,” such as
`those recited in the “receiving” step (Ground 1)
`
`1.
`
`Netflix concedes that Davis fails to disclose the “timer applet,”
`such as those in the “receiving step”
`
`Uniloc had noted that the claim language requires that the “timer applet” must
`
`be “operative by the user’s computer as a timer,” such that “at least a portion of the
`
`identifier data [is received] from the user’s computer responsively to the timer applet
`
`each time a predetermined temporal period elapses using the first computer system.”
`
`
`1 Roku, Inc. (“Roku”) was also an original Petitioner, but the Board has terminated
`
`with respect to Roku due to settlement. Order (Paper 14) at 14.
`
`
`
`1
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`POR at 14-15. In particular, Uniloc pointed out Netflix’s concession that Davis does
`
`“not disclose that the timer applet used on the client machine repeatedly reported the
`
`consumption of a content at the expiration of predetermined time intervals.” Id. at 15
`
`(quoting Pet. 16); see also Pet. at 29 (“Davis did not disclose that this reception occurs
`
`‘each time a predetermined temporal period elapses’”).
`
`In its Reply,2 Netflix further confirms this concession by acknowledging that
`
`Davis discloses a mere timer applet by acknowledging that it was relying solely on
`
`“teachings of Choi that rendered the use of a predetermined time interval an obvious
`
`modification of Davis.” Reply at 7. Accordingly, Netflix admits that Davis, by itself,
`
`does not disclose a “timer applet” where “at least a portion of the identifier data [is
`
`received] from the user’s computer responsively to the timer applet each time a
`
`predetermined temporal period elapses using the first computer system.”
`
`2.
`
`Choi fails to disclose the “time applet,” such as those in the
`“receiving step”
`
`Uniloc had noted that the language of claim 1 expressly requires that the
`
`provided “timer applet” must itself operate as a timer, such that the first computer
`
`
`2 Paper 18, which is entitled “Petitioner’s Reply to Patent Owner’s Preliminary
`
`Response to Petition,” is understood by Uniloc to be Netflix’s Reply to Patent
`
`Owner’s Response.
`
`
`
`2
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`system receives “at least a portion of the identifier data . . . from the user’s computer
`
`system responsively to the timer applet each time a predetermined temporal period
`
`elapses.” POR at 16 (quoting Ex. 1001, 14:25-32). Choi does not disclose this
`
`limitation. The Petition only relies on the following snippets of Choi for this
`
`limitation:
`
`[0047] The client 110 periodically transmits state data (e.g.,
`
`logging statistics) to the server 108 for storage.
`
`[0097] …the frequency of reporting set by the statistics reporting
`
`* * *
`
`interval parameter sent in the initial request.
`
`Pet. at 29 (quoting Choi, ¶47, Appendix C, ¶97). 3
`
`
`But as Uniloc noted, Choi discloses the timing of the periodic transmissions is
`
`set by the “statistics reporting interval parameter sent in the initial request.” POR at
`
`16 (citing Choi at ¶0097). Accordingly, in Choi the client’s initial request dictates
`
`the timing interval of the periodic transmissions. Pet. 29‒30 (citing Choi at ¶0097;
`
`Franz ¶¶103-109). But sending an “interval parameter” in an initial request of Choi
`
`is distinguishable from the claimed “timer applet” acting as a timer “each time a
`
`
`3 Both parties reference to “¶0097” is to ¶0097 of Appendix C, starting on page 22
`
`of Choi. There is another Paragraph ¶0097 in the body of Choi ate page 7 that is
`
`not the focus of this term.
`
`
`
`3
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`predetermined temporal period elapses.” There is simply no disclosure in Choi of any
`
`applet acting as a timer. See generally Choi. And there is no particular disclosure of
`
`a timer that acts “each time a predetermined period elapses.”
`
`In its Reply, Netflix completely misses the point by arguing “[c]laim 1 does
`
`not specify how the predetermined temporal period is determined.” Reply at 7.
`
`Netflix misapprehends Uniloc’s argument. The issue is not how the temporal period
`
`is determined but that Choi’s mere transmission of an interval parameter (e.g., a
`
`certain number of seconds) in the initial request is not a teaching of an applet acting
`
`as a timer. And the claimed timer must act “each time a predetermined temporal
`
`period elapses.” Just because the “interval parameter” is sent once, does not
`
`necessarily infer that the “interval parameter” is used by a “timer applet.” And Netflix
`
`fails to explain why the disclosure of an “interval parameter” would necessarily mean
`
`that such a parameter will be used “each time a predetermined temporal period
`
`elapses.”
`
`3.
`
`Netflix’s belated new theories regarding the proposed Davis-
`Choi combination constitute an
`improper attempt
`to
`circumvent the Board’s rules and precedent
`
`With respect to the proposed Davis-Choi combination, Netflix submits
`
`additional evidence that should have been presented in the Petition.
`
`A petitioner may not submit new evidence or argument in reply that it could
`
`have presented earlier. Patent Trial and Appeal Board Consolidated Trial Practice
`
`
`
`4
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`Guide November 2019 (“Consolidated Trial Practice Guide”), at 73-74. Moreover,
`
`the Board has warned:
`
`The opportunity to submit additional evidence does not allow a
`
`petitioner to completely reopen the record, by, for example,
`
`changing theories after filing a petition. See Intelligent Bio-Sys.,
`
`Inc. v. Illumina Cambridge, Ltd., 821 F.3d 1359, 1369–70 (Fed.
`
`Cir. 2016) (affirming Board discretion to deny entry of
`
`petitioner’s reply brief
`
`that contained an
`
`improper new
`
`unpatentability theory and evidence, citing, among other things, §
`
`312(a)(3)); see also CTPG at 74 (“‘Respond,’ in the context of 37
`
`C.F.R. § 42.23(b), does not mean proceed in a new direction with
`
`a new approach as compared to the positions taken in a prior
`
`filing.”).
`
`Hulu, LLC v. Sound View Innovations, LLC, Case IPR2018-01039, Paper 20, at 15-
`
`16 (Dec. 20, 2019) (Precedential).
`
`The Board should not consider the entire reply (as opposed to the objectionable
`
`portions of the reply) when it “raises a new issue or belated presents evidence.”
`
`Consolidated Trial Practice Guide at 74. This is because the “Board is not required to
`
`attempt to sort proper from improper portions of the reply.” Id.
`
`Here, the Reply and its accompanying Franz Supplement (Ex. 1009) belatedly
`
`introduce at least two new theories, both of which are improper as addressed below.
`
`
`
`5
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`First, the Petition relies on just two short snippets of Choi (portions of ¶47 and
`
`Appendix C, ¶97) for the disclosure of the “timer applet” term in context of the
`
`receiving step. In its Reply, Netflix now improperly provides the supplemental
`
`declaration of Dr. Franz (“Franz Supplement”) (Ex. 1009) in an effort to interject a
`
`new inherency argument that “identifier information would naturally be required in
`
`each message,” and therefore the identifier information could somehow be sent by
`
`the client server with each periodic message. Franz Supplement, ¶8; see generally id.,
`
`¶¶6-8; Reply at 8-10 (citing Franz Supplement, ¶¶8-11).
`
`For the Board’s convenience, the full paragraph of Paragraph 97 of Appendix
`
`C is reproduced in its entirety below:
`
`
`
`Pet. 30 (citing Choi, Appendix C at ¶97).
`
`The plain language of Paragraph 97 mandates a bifurcated and non-
`
`overlapping set of parameters. If a parameter is “constant,” it is “sent only once at the
`
`beginning of the session.” POR at 17 (quoting Choi, ¶97). On the other hand, only
`
`the non-constant, “dynamically changing parameters” in Choi “are sent regularly.”
`
`Id. (quoting Choi, ¶97). Given this plain teaching of Choi, Netflix should have
`
`
`
`6
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`presented its new inherency argument in its Petition. Indeed, the issue seems to be
`
`compounded by Dr. Franz and Netflix cherry picking only a snippet of Paragraph 97
`
`of Appendix C, and not reading Paragraph 97 in its entirety. Pet. at 30 (citing portion
`
`Choi, ¶97); Franz Declaration, ¶¶ 103-109; accord POR at 16-17 (noting that Netflix
`
`only “partially quotes paragraph 97”). And Netflix extensively discussed other
`
`aspects of Choi in the Petition (Pet. at 16-18, 25-26, 29-42) and not once stated that
`
`the identifier information would “naturally be required in each message.”
`
`Second, Netflix and Dr. Franz attempt to interject what Dr. Franz opines as the
`
`“long … standard practice” of using “cookies” to “keep[] track of state between
`
`clients and servers on the web.” Franz Supplement, ¶10; Reply at 8-10 (citing Franz
`
`Supplement, ¶¶6-11). Dr. Franz also refers to an unauthenticated archive.org web
`
`page that purportedly shows that “this technology originated at Netscape and found
`
`its way even into competing browsers by 1995.” Franz Supplement, ¶10. And Dr.
`
`Franz points to a U.S. Patent issued to Netscape Communications Corp. for what he
`
`describes—without providing any explanation whatsoever—as relating to “web
`
`browser cookies.” Id. Neither the unauthenticated web page nor the Netscape patent
`
`were mentioned in the Petition and have not been introduced as exhibits. See
`
`generally Pet. Indeed, the phrase “cookies” appears only once in brief passing in the
`
`Petition and the original Franz Declaration, respectively. Pet. at 24 (“Davis also
`
`disclosed that Server B sent a client ID or “cookie” to the client.”);Franz Declaration,
`
`
`
`7
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`¶84 (“Davis explains that this registration and setup can be performed using a cookie,
`
`which the Server B can create and provide to the Client.). Netflix should have raised
`
`the new arguments and references relating to “cookies” earlier. Netflix cited to
`
`“cookies” (albeit only once) in its Petition, and accordingly should have understood
`
`that the term and accompanying expanded opinions of Dr. Franz—under its theory—
`
`were germane.
`
`For both new theories, Netflix’s tactics are improper. See Consolidated Trial
`
`Practice Guide at 74 (“It is also improper for a reply to present new evidence
`
`(including new expert testimony) that could have been presented in a prior filing, for
`
`example newly cited prior art references intended to ‘gap-fill’ by teaching a claim
`
`element that was not present in the prior art presented with the petition.”). Netflix
`
`should not be allowed to “gap-fill” deficiencies in its Petition when (as demonstrated
`
`above) it could have done so at the time it filed its Petition. Accordingly, the Board
`
`is respectfully requested not to consider Netflix’s Reply in its entirety.
`
`4.
`
`Netflix’s belated new theory is factually and legally defective
`
`Even if the Board does consider Netflix’s effort to reopen the record to
`
`introduce new theories, that effort is factually and legally defective.
`
`First, the Franz Supplement conflicts with the plain teaching of Choi. As
`
`noted, the plain language of Choi mandates that if a parameter is “constant,” it is “sent
`
`only once at the beginning of the session.” POR at 17 (quoting Choi, ¶97). On the
`
`
`
`8
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`other hand, only the “dynamically changing parameters” in Choi “are sent regularly.”
`
`Id. (quoting Choi, ¶97). The Franz Supplement alleges that Choi teaches a “identifier
`
`information would naturally be required in each reporting message so that the server
`
`is able to determine which client it is communicating with.” Franz Supplement, ¶8.
`
`But the Franz Supplement then fails to point to any statement in Choi that makes that
`
`statement. See generally Franz Supplement. Dr. Franz’s testimony contradicts the
`
`plain language of Choi and, accordingly, cannot constitute substantial evidence. See
`
`Ericsson Inc. v. Intellectual Ventures I LLC, 890 F.3d 1336, 1346 (Fed. Cir. 2018)
`
`(“To contradict a reference, an unsupported opinion is not substantial evidence.”);
`
`Homeland Housewares, LLC v. Whirlpool Corp., 865 F.3d 1372, 1378 (Fed. Cir.
`
`2017) (“[W]e must disregard the testimony of an expert that is plainly inconsistent
`
`with the record, or based on an incorrect understanding of the claim[s].” (citations
`
`and internal quotation marks omitted)); accord MEHL/Biophile Intern. Corp. v.
`
`Milgraum, 192 F.3d 1362, 1367 (Fed. Cir. 1999) (noting the analogous summary
`
`judgment context that “expert testimony contradicting the plain language of the
`
`reference does not create a genuine issue of fact”).
`
`Second, Netflix contends that the “[o]ne of the main purposes of Choi is to
`
`track media streams provided to clients.” Franz Supplement, ¶8 (citing Choi, ¶0006).
`
`Even assuming arguendo that this is one of Choi’s “main purposes,” it does not mean
`
`that “identifier information would naturally be required in each reporting message so
`
`
`
`9
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`that the server is able to determine which client it is communicating with.” For
`
`instance, identifier information could be sent in some but not “each” reporting
`
`message, and still track media streams provided to clients. Dr. Franz fails to address
`
`this possibility. This only highlights that Dr. Franz is merely speculating on Choi
`
`might or might not teach. Netflix also overlooks that Choi expressly states that, under
`
`its disclosed scheme, a client should not send its own identifier, as this would violate
`
`privacy concerns. Rather, “due to privacy concerns the client sends up an empty value
`
`(denoted as a dash character).” Choi, ¶99, Table C1 (entry for the “c-dns” Logging
`
`Field). Moreover, as repeatedly emphasized above, to the extent a client ID is
`
`considered constant, then under the express teachings in Choi, such a parameter
`
`would be sent only once. Choi, Appendix C, ¶97
`
`Third, the Franz Supplement is based on an improper inherency theory—i.e.,
`
`that “identifier information would naturally be required in each reporting message
`
`so that the server is able to determine which client it is communicating with.” While
`
`Netflix does expressly invoke the inherency doctrine by name in its Petition, Netflix
`
`attempts to interject an inherency theory by relying on the Franz Supplement. But
`
`Netflix and the Franz Supplement fail to even articulate the legal standard for
`
`inherency. And accordingly, they do not establish inherency under the requisite legal
`
`standard. Toshiba Corp. v. Optical Devices, LLC, IPR2014-01446, Paper 31 (PTAB
`
`Mar. 9, 2016) (giving expert’s testimony regarding inherency persuasive weight
`
`
`
`10
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`when the expert’s testimony failed to establish inherency under the requisite legal
`
`standard); I-Blason LLC v. Aevoe Corp., IPR2016-00231, Paper 6 (PTAB May 18,
`
`2016) (finding no inherency when the evidence only showed that the claimed
`
`functionality may be present); In order to demonstrate that a reference inherently
`
`discloses a claim element, Netflix must show that the reference “‘necessarily
`
`functions in accordance with, or includes, the claimed limitations.’” In re Cruciferous
`
`Sprout Litig., 301 F.3d 1343, 1349 (Fed. Cir. 2002) (citing MEHL/Biophile Int’l
`
`Corp. v. Milgraum, 192 F.3d 1362, 1365 (Fed. Cir. 1999)). Netflix and Dr. Franz
`
`failed to articulate and apply the requisite standard.
`
`Fourth, the Franz Supplement is based on assumptions unsupported by
`
`evidence. See Consolidated Trial Practice Guide at 40-41 (“Affidavits expressing an
`
`opinion of an expert must disclose the underlying facts or data upon which the opinion
`
`is based. . . . Opinions expressed without disclosing the underlying facts or data may
`
`be given little to no weight.” (citations omitted)). For instance, the Franz Supplement
`
`references an unauthenticated archive.org webpage and a Netscape patent, but Netflix
`
`has not introduced them as exhibits into the record.4 Thus, the Board should not afford
`
`Dr. Franz’s unsupported opinions persuasive weight. The Franz Supplement is also
`
`
`4 To be clear, Uniloc would oppose any attempt by Netflix to file these documents
`
`as exhibits at this stage.
`
`
`
`11
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`filled with conclusory statements without providing any evidentiary support or
`
`explanation. For instance, the Franz Supplement makes the following bald and
`
`conclusory assertions without providing any supporting evidence:
`
`• “That [tracking media streams to clients] would be impossible
`
`if the clients anonymously sent reporting messages to the
`
`server, i.e. without identifier data.”
`
`• “There would be no ability, or reason, to perform the stream
`
`tracking information described in Choi if the server cannot
`
`determine which client provides each reporting message.”
`
`• Referring to “widespread discussion of their [i.e., cookies]
`
`privacy implications” without referring to evidence or
`
`explaining what is meant by “wide spread discussion” or
`
`providing any support.
`
`Franz Supplement, ¶¶ 8, 10. In each of these instances, Dr. Franz makes no attempt
`
`to “disclose the underlying facts or data upon which” he bases these opinions. He just
`
`lobs unsupported statements. See Consolidated Trial Practice Guide at 40-41. The
`
`Board should give no weight to the Franz Supplement.
`
`5.
`
`No motivation to combine Davis and Choi
`
`Uniloc had noted the incompatibility of combing Davis and Choi, namely that
`
`Choi expressly states that a client should not send its own identifier, as this would
`
`violate privacy concerns. Rather, “due to privacy concerns the client sends up an
`
`empty value (denoted as a dash character).” Choi, Appendix C at ¶99, Table C1 (entry
`
`
`
`12
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`for the “c-dns” Logging Field). Moreover, as repeatedly emphasized above, to the
`
`extent a client ID is considered constant, then under the express teachings in Choi,
`
`such a parameter would be sent only once. Choi, Appendix C at ¶97.
`
`In its Reply, Netflix makes two arguments, neither of which has merit.
`
`First, Netflix argues that “Davis itself already disclosed sending the client ID
`
`in each reporting message from the client to the Server B” and that “nothing in Choi’s
`
`disclosure suggests that the Server B would somehow not need to keep track of the
`
`association between the tracking data and the client in the subsequent periodic
`
`reports.” Reply at 10. Netflix misses the mark. The concern is that Davis and Choi
`
`teach distinct and incompatible approaches. By its own admission, Netflix reads
`
`Davis as disclosing “sending the client ID in each reporting message.” Id. In contrast,
`
`Choi should not send its own identifier “due to privacy concerns” and further to the
`
`extent a client ID is considered a constant, it would be sent only once. Choi, Appendix
`
`C at ¶97. This inherent incompatibility would block any POSITA from being
`
`motivated to combine Davis and Choi.
`
`Second, Netflix argues that it did not rely on the “c-dns” value, and instead
`
`relied only on the “stream identifier” in its Petition. Reply at 10. In other words,
`
`Netflix argues that it can carve out just the “stream identifier” from other identifiers,
`
`including the “c-dns” value. Netflix, however, fails acknowledge the “privacy
`
`concern” noted in Choi teach away from sending the client ID regularly. Nor should
`
`
`
`13
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`Netflix be able to “stitch together an obviousness finding from discrete portions of
`
`prior art references without considering the references as a whole.” See In re
`
`Enhanced Sec. Research, LLC, 739 F.3d 1347, 1355 (Fed. Cir. 2014) (“Panduit
`
`explains that § 103 does not permit a court to stitch together an obviousness finding
`
`from discrete portions of prior art references without considering the references as a
`
`whole.”); Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561, 1568 (Fed. Cir. 1987)
`
`(citing W.L . Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1550
`
`(Fed.Cir.1983), cert. denied, 469 U.S. 851 (1984)) (“[P]rior [art] patent must be
`
`considered in its entirety, i.e., as a whole, including portions that would lead away
`
`from the invention in suit.”). The prior art “references as whole” suggest that that
`
`they would not be combined.
`
`B.
`
`Petitioner fails at least to prove its proposed Siler-Davis
`combination renders obvious limitations directed to the “timer
`applet,” such as those recited in the “receiving” step (Ground 2)
`Uniloc had noted Petitioner points only to a passing and unexplained reference
`
`in Siler that Player application 122 “may be implemented as a web page.” POR at 21-
`
`22; Pet. 44 (citing Siler, ¶32). Uniloc also noted that neither the Petition nor its
`
`attached declaration explain how Siler’s passing reference to a webpage
`
`implementation expressly or even inherently relates to the provision of a “timer
`
`applet” by a remote computer system. POR at 22. Clearly, it is not necessary that all
`
`web pages contain applets, much less the particular “timer applet” claimed. Id.
`
`
`
`14
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`In its Reply, Netflix points to the “extensive explanation” describing how the
`
`disclosure of Siler would purportedly be modified. Reply at 17-18. The “extensive
`
`explanation” in the Petition is as follows:
`
`A POSA would have looked to and borrowed from the disclosure
`
`of Davis to modify the system of Siler for at least one prominent
`
`reason. Siler disclosed that streaming content could be displayed
`
`at a client using a software application 122 installed on the client.
`
`Siler at ¶¶0025, 0027. After describing the operation of such a
`
`system, Siler explained various alterations that could be made to
`
`the system, including that the software application could be
`
`implemented as a web page instead of the installed software
`
`application. Id. at ¶0032. Siler did not then repeat the preceding
`
`disclosure to explain how this alternative implementation would
`
`impact the system as disclosed. Id. But a POSA would have
`
`recognized that a change from an installed application to a web
`
`page would have motivated certain other changes. Franz ¶¶206-
`
`218.
`
`One such change that a POSA would have recognized as being
`
`advantageous with the web page implementation in Siler would
`
`be that certain application logic disclosed as being contained in
`
`the software application 122 could actually be included as
`
`downloadable logic embedded in the web page. Id. This was well
`
`known in the art, of which Davis is used as one example. Davis at
`
`12:13-50; see also id. at 9:20-23, 18:1-5; Franz ¶¶206-218. This
`
`downloadable logic could be an applet or any other script that
`
`
`
`15
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`
`would run in the context of the web page and web 15 browser.
`
`Franz ¶¶206-218. A POSA would have been motivated to use an
`
`applet for delivering the timer functionality for numerous reasons:
`
`because it was one of only a finite number of ways to provide the
`
`functionality, because it was a well known and preferable way to
`
`implement custom functionality in a web page, and because the
`
`applet implementation would allow the system to require less
`
`software 20 to be stored permanently on the user client 101. Id.
`
`Pet. at 43-44. The length and convoluted nature of Netflix’s “extensive
`
`explanation” only demonstrates how Netflix and Dr. Franz improperly apply
`
`hindsight reconstruction. Indeed, the number of modifications and steps explained
`
`above could only be made with the benefit of the claim as a template. Such hindsight
`
`reconstruction does not establish that it would be obvious for one of ordinary skill to
`
`combine features of disparate references to obtain the claims at issue.
`
`As the Court of Appeals for the Federal Circuit has warned:
`
`Not only must the claimed invention as a whole be evaluated, but so also
`
`must the references as a whole, so that their teachings are applied in the
`
`context of their significance to a technician at the time — a technician
`
`without our knowledge of the solution. The defendants propounded and
`
`the district court appears to have followed an analytical method that well
`
`illustrates the "mosaic" analogy discussed in W.L. Gore Assocs., 721
`
`F.2d at 1552, 220 USPQ at 312, where this court said: [T]he claims were
`
`used as a frame, and individual, naked parts of separate prior art
`
`references were employed as a mosaic to recreate a facsimile of the
`
`
`
`16
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`
`claimed invention.
`
`Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1143 (Fed. Cir. 1985).
`
`In the instant case, Netflix and Dr. Franz adopt a convoluted, multi-step
`
`modification theory that is untethered from what the references themselves teach to
`
`an ordinary technician. Clearly, Netflix has failed to meet their burden of proof.
`
`C. The Petition fails at least to prove its proposed Siler-Davis
`combination renders obvious “wherein each provided webpage
`causes corresponding digital media presentation data to be
`streamed from a second computer system distinct from the first
`computer system directly to the user’s computer independent of
`the first computer system” (Ground 2)
`
`Uniloc has explained how the respective web servers and streaming server of
`
`Siler are not “distinct” from one another, and do not operate the claimed streaming
`
`“independent” of one another, as recited in claim 1. POR at 23. Uniloc specifically
`
`pointed out that Siler’s streaming server 105, web server 113 and web server 117 are
`
`each part of “distinct” computer systems, as opposed to being part of the same
`
`computer system. POR at 23-24. As another example, Petitioner also fails to prove
`
`that the streaming of content from streaming server 105 is “independent” of the web
`
`servers 113 and 117. POR at 24. Tellingly, the Petition does not even mention the
`
`claim term “independent” in its argument alleged addressing the claim language in
`
`which this term appears. See generally Pet.
`
`
`
`17
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`Uniloc explained various aspects of Siler in greater detail. POR at 24-25.
`
`Uniloc noted that the Petition recognizes that, under the construction applied therein
`
`for “computer system,” the obviousness theory depends on further establishing that
`
`Siler’s streaming server and web servers are “under separate operation or control.”
`
`Pet. at 62. In particular, if the “computer system” term recited in claim 1 means “one
`
`or more computing devices having a common operator or under common control,” as
`
`Petitioner argues, then it follows that “a second computer system” cannot be “distinct
`
`from the first computer system” if both have a common operator or are under common
`
`control. Otherwise, these distinct terms would be impermissibly conflated together.
`
`To satisfy this the application of these constructions, the Petition and its
`
`attached declaration by Dr. Franz merely speculate—without citing any
`
`documentary evidence—that Siler’s streaming server and web servers “could be
`
`under separate operation or control.” Pet. at 62. Dr. Franz speculates that a television
`
`station or an owner of a television station may control the streaming server. Ex. 1002,
`
`¶236. Dr. Franz further provides the unsupported speculation that each content
`
`owner could operate their own streaming server 105. Id. No documentation is cited
`
`as alleged support for such speculation that an individual source might control such
`
`a streaming server. Netflix continues to advocate this position in its Reply. Reply at
`
`19-26. The problem with this approach is that Netflix argues inherency (without
`
`invoking the doctrine by name). Just as noted above in Section II.A.4., Netflix has
`
`
`
`18
`
`
`
`IPR2020-00041
`U.S. Patent No. 8,407,609
`
`failed here to articulate and meet the requisite showing under the law of inherency.
`
`See I-Blason LLC v. Aevoe Corp., IPR2016-00231, Paper 6, at (PTAB May 18,
`
`2016) (ruling that Petitioner had not demonstrated that the claimed functional
`
`limitation would be inherent when “the evidence, as presented by petitioner, only
`
`indicates that, at best, the functionality may be present”).
`
`In its Reply, Netflix submits additional evidence. Reply at 24-25 (citing Franz
`
`Supplement, ¶¶12-13). In particular, Dr. Franz suggests that the Siler web and
`
`streamer servers “are not ‘distinct’ from one another, and do not operate the claimed
`
`streaming ‘independent[ly]’ of one another” by raising the following new
`
`arguments:
`
`12. …. However, the descrip