`
`
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`___________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`___________
`
`CLOUDFLARE, INC. and
`SPLUNK INC.,
`Petitioners,
`
`v.
`
`SABLE NETWORKS, INC.
`Patent Owner.
`
`___________
`
`IPR2021-009091
`Patent 8,243,593 B2
`___________
`
`
`
`PETITIONERS’ REPLY TO PATENT OWNER’S RESPONSE
`
`
`1 Splunk, Inc., which filed a petition in IPR2022-00228, has been joined as a
`petitioner in this proceeding.
`
`
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`TABLE OF CONTENTS
`
`I.
`II.
`
`INTRODUCTION ........................................................................................... 1
`COPELAND’S “CONCERN INDEX” CONSTITUTES A
`“BADNESS FACTOR” AS RECITED IN THE ’593 PATENT .................... 2
`Petitioners’ Unrebutted Evidence Establishes That Copeland’s
`A.
`“Concern Index” Constitutes a “Badness Factor” ................................. 2
`There Is No Indefiniteness Argument Before the Board ...................... 4
`Copeland’s “Concern Index” Indicates “Undesirable Behavior” ......... 7
`The Claims Do Not Require Calculating a “Badness Factor” for
`“Each” Flow .......................................................................................... 8
`III. PETITIONERS’ UNREBUTTED EVIDENCE ESTABLISHES
`MOTIVATION TO COMBINE YUNG AND COPELAND ......................... 9
`IV. THE FOUR-STEPS WHITEPAPER WAS PUBLICLY
`ACCESSIBLE ............................................................................................... 14
`Petitioners’ Supplemental Evidence Further Confirms That the
`A.
`Four-Steps Whitepaper Was Publicly Available ................................ 18
`The Four-Steps Whitepaper Was Publicly Accessible Via
`1.
`the Packeteer Website ............................................................... 19
`The Four-Steps Whitepaper and the Packeteer Website
`Were Cited in Prior Art Patents and Publications .................... 22
`The Board Should Not Ignore Petitioners’ Supplemental
`Evidence .................................................................................... 24
`Patent Owner’s Attorney Declaration Does Not Dispute
`Petitioners’ Evidence of Public Accessibility ..................................... 25
`CLAIMS 17, 18, 37, AND 38 ARE PART OF GROUND 2 ........................ 26
`V.
`VI. CONCLUSION .............................................................................................. 29
`
`
`
`B.
`C.
`D.
`
`B.
`
`2.
`
`3.
`
`
`
`i
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`TABLE OF AUTHORITIES
`
` Page(s)
`
`Federal Court Cases
`AstraZeneca AB v. Mylan Pharmaceuticals Inc.,
`19 F.4th 1325 (Fed. Cir. 2021) ............................................................................. 7
`Blue Calypso, LLC v. Groupon, Inc.,
`815 F.3d 1331 (Fed. Cir. 2016) .......................................................................... 23
`Cochlear Bone Anchored Sols. AB v. Oticon Medical LLC,
`958 F.3d 1348 (Fed. Cir. 2020) ............................................................................ 4
`Enzo Biochem Inc. v. Applera Corp.,
`599 F.3d 1325 (Fed. Cir. 2010). ........................................................................... 5
`Everstar Merch. Co. Ltd. v. Willis Elec. Co. Ltd.,
`No. 2021-1882, 2022 WL 1089909 (Fed. Cir. Apr. 12, 2022) ........................... 16
`GoPro, Inc. v. Contour IP Holding LLC,
`908 F.3d 690 (Fed. Cir. 2018) ...................................................................... 15, 17
`Hewlett-Packard Co. v. Mustek Sys., Inc.,
`340 F.3d. 1314 (Fed. Cir. 2003) ........................................................................... 9
`Intel Corp. v. Qualcomm Incorp.,
`21 F.4th 801 (Fed. Cir. 2021) ............................................................................... 4
`Jazz Pharm., Inc. v. Amneal Pharm., LLC,
`895 F.3d 1347 (Fed. Cir. 2018) .................................................................... 17, 26
`In re Carol. F. Klopfenstein,
`380 F.3d 1345 (Fed. Cir. 2004) .......................................................................... 26
`KSR Int’l. Co. v. Teleflex Inc.,
`550 U.S. 398 (2007) ............................................................................................ 13
`In re Magnum Oil Tools Int’l, Ltd.
`829 F.3d 1364 (Fed. Cir. 2016) .......................................................................... 27
`
`
`
`ii
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Meiresonne v. Google, Inc.,
`849 F.3d 1379 (Fed. Cir. 2017) .......................................................................... 13
`Nautilus, Inc. v. Biosig Instruments, Inc.,
`572 U.S. 898 (2014) .............................................................................................. 5
`Samsung Elecs. Am., Inc. v. Prisua Eng’g. Corp.,
`948 F.3d 1342 (Fed. Cir. 2020) ............................................................................ 4
`Sirona Dental Sys. GMBH v. Institut Straumann AG,
`892 F.3d 1349 (Fed. Cir. 2018) .......................................................................... 27
`SRI Int’l. Inc. v. Internet Sec. Sys., Inc.,
`511 F.3d 1186 (Fed. Cir. 2008) .......................................................................... 15
`Suffolk Techs., LLC v. AOL Inc.,
`752 F.3d 1358 (Fed. Cir. 2014) .......................................................................... 16
`Valve Corp. v. Ironburg Inventions Ltd.,
`8 F.4th 1364 (Fed. Cir. 2021) ........................................................... 16, 17, 22, 24
`VidStream LLC v. Twitter, Inc.,
`981 F.3d 1060 (Fed. Cir. 2020) .......................................................................... 24
`Patent Trial and Appeal Board Decisions
`Align Technology v. 3Shape,
`IPR2020-01087, Paper 38 (PTAB Jan. 19, 2022) ................................................ 5
`Amperex Technology v. LG Chem.,
`IPR2018-00783, Paper 8 (PTAB Sept. 19, 2018) ............................................... 28
`Axonics v. Medtronic,
`IPR2020-00713, Paper 42 (PTAB Sept. 22, 2021)............................................. 28
`Coalition for Affordable Drugs VIII v. Univ. Penn.,
`IPR2015-01836, Paper 58 (PTAB Mar. 6, 2017) ......................................... 21, 22
`DJI Europe v. Daedalus Blue,
`IPR2020-01472, Paper 31 (PTAB Feb. 14, 2022) ................................................ 5
`Dr. Reddy’s Laboratories v. Horizon Pharma USA,
`IPR2018-00272, Paper 35 (PTAB Mar. 28, 2019) ............................................... 7
`
`
`
`iii
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Dr. Reddy’s Laboratories v. Horizon Pharma USA,
`IPR2018-00272, Paper 74 (PTAB Sept. 6, 2019) ................................................. 5
`Facebook v. Sound View Innovations,
`IPR2017-00985, Paper 17 (PTAB Sept. 5, 2017) ................................................. 6
`Fanduel v. Cg Tech. Dev.,
`IPR2017-00902, Paper 45 (PTAB Oct. 4, 2018) ................................................ 24
`Grünenthal v. Antecip Bioventures II,
`PGR2018-00092, Paper 24 (PTAB Feb. 24, 2020) ............................................ 17
`IBM v. Rigetti & Co.,
`IPR2020-00494, Paper 13 (PTAB Aug. 11, 2020) ............................................. 28
`IBM v. Trusted Knight,
`IPR2020-00323, Paper 37 (June 30, 2021) ......................................................... 28
`Japan Radio v. Broadcom,
`IPR2019-00816, Paper 34 (PTAB Aug. 28, 2020) ............................................. 22
`Netapp v. KOM Software,
`IPR2019-00606, Paper 37 (PTAB Sept. 23, 2020)............................................. 28
`Nippon Suisan Kaisha v. Pronova Biopharma Norge,
`PGR2017-00033, Paper 37, and 12 (PTAB Jan. 16, 2019) ................................ 28
`Schlumberger Technology v. Integrated Drive Systems,
`IPR2018-00603, Paper 40 (PTAB Sept. 3, 2019) ............................................... 22
`Target v. Proxicom Wireless,
`IPR2020-00904, Paper 11 (PTAB Nov. 10, 2020) ............................................... 5
`Veeam Software v. Symantec,
`IPR2013-00142, Paper 11 (PTAB Aug. 7, 2013) ............................................... 15
`Zillow Group v. IBM,
`IPR2020-01655, Paper 8 (PTAB Mar. 15, 2021) ................................................. 5
`Federal Statutes
`35 U.S.C. § 102 .............................................................................................. 5, 17, 20
`35 U.S.C. § 112 .......................................................................................................... 6
`
`
`
`iv
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Federal Rules and Regulations
`37 C.F.R. § 42.104(b)(3) ............................................................................................ 6
`
`
`
`
`
`v
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`I.
`
`INTRODUCTION
`After institution, Patent Owner voluntarily disclaimed 17 challenged claims.
`
`The remaining claims are addressed in Grounds 2 and 3:
`
`Ground
`
`2
`
`Challenged Claims
`
`Prior Art
`
`Independent claim 9 with dependent claims
`10-13 and 17-24, and independent claim 29
`with dependent claims 30-33 and 37-442
`
`Yung and Copeland
`
`3
`
`Independent claim 3
`
`Yung and Four-Steps
`Whitepaper
`
`Patent Owner’s response largely recycles arguments the Board rejected in its
`
`Institution Decision attempting to distinguish Copeland’s “concern index” from the
`
`claimed “badness factor” and suggesting that there is no motivation to combine
`
`Yung and Copeland. Notably, Patent Owner offers no expert declaration to
`
`support its various assertions about what a POSA would or would not have known,
`
`understood, or been motivated to do. In contrast, Petitioners’ expert Dr. Jeffay
`
`provides detailed and well-supported testimony explaining (1) why Copeland’s
`
`
`2 As the Institution Decision recognized, due to a typographical error dependent
`
`claims 17, 18, 37, and 38 appeared in the petition’s section addressing Ground 1
`
`rather than Ground 2 where they logically and clearly belong—and as
`
`Cloudflare immediately confirmed to Patent Owner. See Section V, below.
`
`
`
`
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`“concern index” discloses the claimed “badness factor” to a POSA, and (2) why a
`
`POSA would have been motivated to combine Yung’s bandwidth management
`
`device with Copeland’s “concern index” teachings and how such a person would
`
`have done so. The record evidence is not even close.
`
`Regarding Ground 3, Patent Owner relies on flawed evidence in an attorney
`
`declaration to argue that the Four-Steps Whitepaper was not publicly accessible.
`
`In contrast, Petitioner provides clear evidence showing how a POSA, exercising no
`
`more than reasonable diligence, could have obtained the whitepaper from the
`
`Packeteer website in the relevant timeframe.
`
`II. COPELAND’S “CONCERN INDEX” CONSTITUTES A “BADNESS
`FACTOR” AS RECITED IN THE ’593 PATENT
`A.
`Petitioners’ Unrebutted Evidence Establishes That Copeland’s
`“Concern Index” Constitutes a “Badness Factor”
`Patent Owner argues that the petition cites “no evidence except the ipse dixit
`
`in the accompanying declaration” of Dr. Jeffay, quoting a single sentence from the
`
`declaration and ignoring Dr. Jeffay’s actual testimony. Paper 30 (“POR”) at 17.
`
`Dr. Jeffay noted that the ’593 Patent does not define the term “badness factor” but
`
`includes a description and examples in the specification. EX1003 ¶208.3 He
`
`
`3 One of the examples of a “badness factor” is a ratio based on the total byte
`
`count. EX1003 ¶208 (citing EX1001, Fig. 5). Similarly, Copeland discloses an
`
`
`
`2
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`explained that Copeland discloses computing a badness factor for the flow as
`
`claimed, because it describes analyzing flow statistics to determine if the flow
`
`appears to be “legitimate traffic or possible suspicious activity” and assigning a
`
`“concern index” value to flows that appear suspicious. EX1003 ¶209 (citing
`
`EX1007, 3:31-35, 18:16-17, 26:24-26). Dr. Jeffay then explained that “[j]ust as
`
`the badness factor in the ’593 patent is a calculated value that indicates a level of
`
`misbehavior attributable to a flow, Copeland’s CI value is assigned to the flow
`
`based on the behavior recognized by the system.” EX1003, ¶210. Further, he
`
`explained that the concern index provides an indication of whether the flow is
`
`exhibiting undesired behavior as claimed. EX1003, ¶211. Copeland explicitly
`
`states that the “concern index” indicates whether the flow exhibits “suspicious
`
`activity”—in other words, it is an indication of whether the flow is exhibiting
`
`undesirable behavior. EX1003, ¶211; see also id. (Copeland contrasts such
`
`“suspicious activity” with “legitimate traffic”). Dr. Jeffay’s opinion is well-
`
`reasoned and well-supported by the evidence. EX1003 ¶¶208-214; EX2007,
`
`35:17-38:2.
`
`
`exemplary “concern index” based on total byte count. EX1007, Fig. 6; EX1003
`
`¶213.
`
`
`
`3
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Furthermore, Dr. Jeffay’s opinion is unrebutted by any contrary evidence.
`
`Patent Owner offers only attorney argument in response.
`
`B.
`There Is No Indefiniteness Argument Before the Board
`Patent Owner asserts that because Cloudflare has argued in district court that
`
`the term “badness factor” is indefinite, Petitioners cannot maintain obviousness
`
`arguments for challenged claims including that term. POR at 18. Patent Owner’s
`
`argument is legally incorrect. The Board can invalidate a claim as obvious even if
`
`it is also indefinite. See Samsung Elecs. Am. v. Prisua Eng’g, 948 F.3d 1342, 1355
`
`(Fed. Cir. 2020); Cochlear Bone Anchored Solutions v. Oticon Medical LLC, 958
`
`F.3d 1348 (Fed. Cir. 2020) (reversing Board’s refusal to conduct prior art
`
`invalidity analysis of an indefinite claim). Indeed, “[t]he indefiniteness of a
`
`limitation…precludes a patentability determination only when the indefiniteness
`
`renders it logically impossible for the Board to reach such a decision.” Intel v.
`
`Qualcomm, 21 F.4th 801, 813 (Fed. Cir. 2021).4 Patent Owner does not even
`
`suggest that it is logically impossible for the Board to adjudicate Petitioners’ prior
`
`
`4 All emphasis is added unless otherwise indicated.
`
`
`
`4
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`art challenges on the merits.5
`
`The Board regularly reaches the merits even where a petitioner argues that
`
`the challenged claims are indefinite in related district court litigation. See, e.g.,
`
`DJI Europe v. Daedalus Blue, IPR2020-01472, Paper 31 at 17, 19 (PTAB Feb. 14,
`
`2022); Align Technology v. 3Shape, IPR2020-01087, Paper 38 at 8, 40-41 (PTAB
`
`Jan. 19, 2022) (finding claim obvious even though it had already been held
`
`indefinite in district court); Zillow Group v. IBM, IPR2020-01655, Paper 8 at 12
`
`(PTAB Mar. 15, 2021) (rejecting patent owner’s argument that “we may not assess
`
`the patentability of claims under § 102 in an inter partes review merely because
`
`the Petitioner pled that the claims are indefinite in District Court”); Target v.
`
`Proxicom Wireless, IPR2020-00904, Paper 11 at 12 (PTAB Nov. 10, 2020); Dr.
`
`
`5 Patent Owner cites Enzo Biochem v. Applera, in which the court stated that the
`
`claims at issue could not be both indefinite and anticipated. 599 F.3d 1325,
`
`1332 (Fed. Cir. 2010). The Enzo court’s statement was based on the premise
`
`that “[i]f a claim is indefinite, the claim, by definition, cannot be construed,”
`
`which was the legal standard for indefiniteness later overruled in Nautilus v.
`
`Biosig Instruments, 572 U.S. 898, 901 (2014) (rejecting standard requiring
`
`claims to be “amenable to construction” and not “insolubly ambiguous” to be
`
`sufficiently definite).
`
`
`
`5
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Reddy’s Laboratories v. Horizon Pharma USA, IPR2018-00272, Paper 74 at 12-13
`
`(PTAB Sept. 6, 2019).
`
`Patent Owner’s reliance on Facebook v. Sound View Innovations, IPR2017-
`
`00985, Paper 17 (PTAB Sept. 5, 2017), is misplaced. In that pre-SAS proceeding,
`
`the Board denied institution of several challenged claims because the petition
`
`identified a means-plus-function claim term without also identifying any
`
`corresponding structure. Id. at 13-14 (“The analysis of Petitioner’s arguments
`
`regarding claims 1-3 begins and ends with Petitioner’s failure to provide
`
`constructions of the claim terms including ‘controller’ under 35 U.S.C. § 112,
`
`¶6….”); see also 37 C.F.R. § 42.104(b)(3). Contrary to Patent Owner’s argument
`
`here, the Board instituted review of several claims in Facebook even though the
`
`petitioner was simultaneously arguing in district court that those claims were
`
`indefinite. IPR2017-00985, Paper 17 at 14 fn.3.
`
`Dr. Jeffay explained during his deposition that while a POSA might not
`
`understand the “bounds” of the term, “Copeland’s recitation of what it refers to
`
`as…measures of suspicious activity would certainly qualify as a badness factor.”
`
`
`
`6
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`EX2007 at 36:2-15.6 Dr. Jeffay did not form an opinion as to the full scope of
`
`what might qualify as a “badness factor”—and Petitioners have not proposed a
`
`specific definition here—because it was unnecessary to do so. Dr. Reddy’s
`
`Laboratories v. Horizon Pharma USA, IPR2018-00272, Paper 35 at 6 (PTAB Mar.
`
`28, 2019) (“[W]e need not necessarily determine the outer boundaries of a claim
`
`in reaching a conclusion as to whether the asserted prior art anticipates a claim or
`
`renders it obvious.”); AstraZeneca v. Mylan Pharmaceuticals, 19 F.4th 1325, 1345
`
`(Fed. Cir. 2021) (“Claims must be construed ‘only to the extent necessary to
`
`resolve the controversy.’”; citation omitted). Copeland’s “concern index” easily
`
`qualifies as a “badness factor” based on its similarity to the descriptions and
`
`examples in the ’593 Patent, as shown by Petitioners’ unrebutted evidence.
`
`EX1003, ¶¶208-214; EX2007, 35:17-38:2.
`
`C. Copeland’s “Concern Index” Indicates “Undesirable Behavior”
`Patent Owner argues that Copeland does not teach the claimed “badness
`
`factor” because that term is “concerned with ‘undesirable’ behavior or
`
`‘misbehavior’ associated with P2P traffic, which, unlike the host activity assigned
`
`
`6 While the term “badness factor” is subjective, Copeland’s illustration of the
`
`disclosed invention explicitly identifies “‘Bad’ Flow(s) & Other Concern Index
`
`Events.” EX1007, Fig. 1.
`
`
`
`7
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`a concern index value by Copeland, is not necessarily suspicious or non-
`
`suspicious.” POR at 21 (emphasis in original). But Patent Owner has not argued
`
`here or in district court that the “badness factor” is somehow limited to P2P traffic.
`
`EX2009, 28 (arguing the term “badness factor” “do[es] not require construction”
`
`and should have its “plain and ordinary meaning”); EX1096, 23-26 (Patent
`
`Owner’s claim construction brief addressing the term “badness factor” without
`
`mentioning P2P traffic). Moreover, as the Institution Decision notes, Patent
`
`Owner’s argument “flips the analysis.” Paper 16 (“ID”) at 44. The relevant
`
`question is whether Copeland’s “concern index” discloses the claimed “badness
`
`factor”—i.e., whether Copeland’s suspicious activity indicates “undesirable”
`
`behavior—not whether the claimed “undesirable” activity associated with the
`
`“badness factor” is suspicious.
`
`D. The Claims Do Not Require Calculating a “Badness Factor” for
`“Each” Flow
`Patent Owner argues that Copeland’s “concern index” cannot constitute the
`
`claimed “badness factor” because it “is not computed on a per-flow basis.” POR at
`
`22. While Patent Owner cites passages in Copeland that suggest the “concern
`
`index” is only calculated for suspicious flows, it simply ignores Copeland’s
`
`explicit statement that “a value called the ‘concern index’ is calculated or
`
`otherwise determined for each flow….” EX1007, 18:15-19.
`
`
`
`8
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`In any event, as noted in the Institution Decision, Patent Owner’s argument
`
`that the claimed “badness factor” must be assigned to “each flow…is not
`
`commensurate with the scope of the claim.” ID at 44 (emphasis in original).
`
`Patent Owner argues that claims 9 and 29 recite such a limitation but identifies no
`
`particular claim language and instead quotes the patent specification stating that a
`
`“badness factor” is computed for “the flow.” POR at 23. Nowhere does the
`
`’593 Patent state that a “badness factor” is calculated for “each flow” as Patent
`
`Owner asserts.7
`
`III. PETITIONERS’ UNREBUTTED EVIDENCE ESTABLISHES
`MOTIVATION TO COMBINE YUNG AND COPELAND
`Petitioners explained, based on specific evidence, the motivation to
`
`incorporate Copeland’s flow-based “concern index” teachings into Yung’s
`
`bandwidth management device. Pet. at 45-48; EX1003 ¶¶196-205. Yung
`
`expressly contemplates incorporating “other factors” into its process, and a POSA
`
`would have sought to classify traffic behaving suspiciously or disruptively. Pet. at
`
`45; EX1003 ¶¶196-198. A POSA would have recognized that incorporating
`
`
`7 As also noted in the Institution Decision, prior art “that sometimes, but not
`
`always, embodies a claimed method nonetheless teaches that aspect of the
`
`invention.” Hewlett-Packard v. Mustek Systems, 340 F.3d. 1314, 1326 (Fed.
`
`Cir. 2003).
`
`
`
`9
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Copeland’s “concern index” would enhance Yung’s ability to identify and control
`
`unauthorized traffic including network attacks and probes, which were well-known
`
`concerns, while obviating the need for an administrator to understand the complex
`
`nature of network attacks. Pet. at 45-47; EX1003 ¶¶199-202. Further, Petitioners
`
`supplied detailed evidence explaining how a POSA would have combined
`
`Copeland and Yung with a reasonable expectation of success. Pet. at 47-48;
`
`EX1003 ¶¶203-204.
`
`Patent Owner argues that the motivation to combine Yung and Copeland
`
`“stems (entirely) from Yung’s statement that its ‘application behavior pattern can
`
`incorporate other factors as well.’” POR at 25; see also ID at 46 (rejecting the
`
`same argument). Patent Owner claims that Petitioners’ expert witness confirmed
`
`this in deposition. But Patent Owner selectively cites only part of Dr. Jeffay’s
`
`answer concerning a single paragraph in his declaration, while omitting the rest of
`
`his answer stating that “the full picture of the motivation to combine is developed
`
`over the following several pages.” EX2007, 38:11-20; see also EX1003 ¶¶197-
`
`205. Patent Owner’s counsel then elicited from Dr. Jeffay even more support for
`
`his opinion that a POSA would have been motivated to combine Yung and
`
`Copeland. EX2007, 38:21-40:11; see also id., 14:10-19:18. The motivation to
`
`combine stretches considerably beyond Yung’s statement, as Petitioners’ evidence
`
`shows.
`
`
`
`10
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Patent Owner also argues that there is no motivation to combine because
`
`“Copeland is directed to collecting self-reported flow information from packet
`
`headers” rather than behavioral statistics. POR at 26-27. This argument fails, as the
`
`Board noted in the Institution Decision, because Copeland describes storing flow
`
`“statistics” including the number of bytes and the number of packets in a flow, both
`
`of which are examples of “behavioral statistics” identified in the ’593 Patent.
`
`EX1007, 3:18-35, 16:65-17:2; see also EX1001, 6:11-14 (stating that “total byte
`
`count” is a behavioral statistic), Fig. 4 (identifying number of packets as a
`
`behavioral statistic).
`
`Patent Owner asserts that Dr. Jeffay confirmed in his deposition that
`
`“Copeland does not disclose collecting flow data from any source other than packet
`
`headers” (POR at 27), which is contradicted by his actual testimony. In the cited
`
`testimony, Dr. Jeffay confirmed that “use of information in the headers is primarily
`
`what I’m relying on” but added that Copeland also “talk[s] about packet length,”
`
`which is not included in certain types of packet headers. EX2007, 40:19-42:11. He
`
`also explained, however, that information contained in packet headers can provide
`
`“behavioral statistics,” as a POSA would have recognized, and that, “at the end of
`
`the day, Copeland is really about the concept of extracting data from packets, either
`
`in headers or measures of packets, and computing an index that is some
`
`
`
`11
`
`

`

`representation of the degree to which the packet is part of an undesirable flow.” Id.,
`
`Case IPR2021-00909—Petitioners’ Reply
`
`42:12-43:8.
`
`Copeland discloses collecting and using behavioral statistics based on non-
`
`header information (e.g., byte count, packet count) as well as statistics based on
`
`header information to identify suspicious behavior. As an example of the latter,
`
`Copeland’s Figure 6 explains that a flow with “Multiple Packets from Same Source
`
`Port to Different Destination Ports” may indicate a “TCP Stealth Port Scan.”
`
`EX1007, Fig. 6. Because port numbers are in packet headers (id., 9:12-15), the
`
`behavior of a flow originating from one port and targeting multiple ports is a
`
`“behavioral statistic” based on header information.
`
`At bottom, the distinction between header and non-header information is
`
`immaterial because—contrary to Patent Owner’s implicit assertion—the term
`
`“behavioral statistics” does not exclude header information. Patent Owner certainly
`
`has not proposed such a claim construction here or in district court. EX1096 at 27-
`
`28. Thus, Patent Owner’s attempt to draw a distinction between Yung, which itself
`
`collects flow data based on header information (e.g., EX1005, 12:12-24), and
`
`Copeland merely because Copeland collects certain flow data from packet headers is
`
`not tied to the scope of the challenged claims. Further, Patent Owner’s response
`
`itself notes that Yung collects “behavioral data” that includes “packet count [and]
`
`byte count” (POR at 27), which are also collected by Copeland (EX1007, 3:18-35,
`
`
`
`12
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`16:65-17:2, Fig. 6), confirming that both references disclose collecting the same
`
`types of “behavioral statistics” without regard to the specific source.
`
`Finally, Patent Owner argues that Copeland and Yung are directed to different
`
`inventions. POR at 27-28. Patent Owner does not specifically argue that either
`
`reference teaches away from the other. See Meiresonne v. Google, 849 F.3d 1379,
`
`1382 (Fed. Cir. 2017) (“A reference teaches away ‘when a person of ordinary skill,
`
`upon reading the reference, would be discouraged from following the path set out in
`
`the reference, or would be led in a direction divergent from the path that was taken’
`
`in the claim.”). Indeed, Patent Owner does not even explain how the purported
`
`differences would be relevant to a POSA.8 In contrast, Petitioners’ expert Dr. Jeffay
`
`explained in detail why and how a POSA would have modified Yung’s bandwidth
`
`management device to incorporate Copeland’s “concern index” teachings. EX1003
`
`¶¶196-205. His detailed and well-supported expert opinion that a POSA would have
`
`been motivated to combine Yung and Copeland stands unrebutted.
`
`
`8 Patent Owner’s implicit argument that references addressing different problems
`
`cannot be combined was rejected in KSR Int’l v. Teleflex, 550 U.S. 398, 420
`
`(2007).
`
`
`
`13
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`IV. THE FOUR-STEPS WHITEPAPER WAS PUBLICLY ACCESSIBLE
`Petitioner cited the Four-Steps Whitepaper (EX1006) for one limitation of
`
`claim 3. Pet. at 61-62. Patent Owner does not dispute that the whitepaper
`
`discloses the claim limitation or that a POSA would have been motivated to
`
`combine it with Yung. Instead, Patent Owner argues only that the Four-Steps
`
`Whitepaper was not publicly accessible. The Board found at institution that the
`
`whitepaper was on Packeteer’s website as of March 17, 2003 (which Patent Owner
`
`does not dispute), but questioned whether it was publicly accessible and stated that
`
`the determination would be made “with the benefit of the full record.” ID at 49,
`
`52.
`
`The full record confirms that the Four-Steps Whitepaper was publicly
`
`accessible—indeed, the whitepaper itself suggests as much. Contrary to Patent
`
`Owner’s incorrect assertion, the Four-Steps Whitepaper does not state on the first
`
`page “that it may not be copied or downloaded without Packeteer’s ‘express
`
`written consent’….” POR at 42. The footer instead states that the document is
`
`copyrighted and prohibits such things as “photocopying” or “storing on a retrieval
`
`system”—common limitations for copyrighted works that are freely available to
`
`the public—and says nothing to prohibit “downloading” it. EX1006, 1. The
`
`copyright notice suggests that the document was publicly accessible to people who
`
`could otherwise make photocopies, store the document in a retrieval system, etc.;
`
`
`
`14
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`otherwise the notice would be unnecessary. See Veeam Software v. Symantec,
`
`IPR2013-00142, Paper 11 at 11-12 (PTAB Aug. 7, 2013) (“The prohibition against
`
`unauthorized copying (like a copyright notice) does not prove the document was
`
`not publicly available. It indicates the opposite, as there would be no need for such
`
`a warning if the document were not disseminated.”).9
`
`Further, the Four-Steps Whitepaper is a promotional document describing
`
`Packeteer’s PacketShaper product. EX1006, 3 (“This paper … serves as a
`
`technical product overview for the PacketShaper product line.”); id. (“Today’s
`
`enterprises require performance, predictability, and consistency from their
`
`networks and the applications that traverse them. And that’s precisely what
`
`PacketShaper® from Packeteer delivers.”); id. (“PacketShaper is available in a
`
`variety of models based on capacity and features. For more detail, see
`
`Deployment – When and Where later in this paper.”); see also SRI Int’l v. Internet
`
`Sec. Sys., 511 F.3d 1186, 1196 (Fed. Cir. 2008) (an “intent to publicize” is a factor
`
`“involved in the public accessibility determination”); GoPro v. Contour IP
`
`Holding, 908 F.3d 690, 694 (Fed. Cir. 2018) (product catalog distributed at trade
`
`show was prior art because “a trade show is directed to individuals interested in the
`
`
`9 In contrast to several cases cited by Patent Owner, Petitioner is not relying on
`
`the copyright date alone to establish that the whitepaper is prior art.
`
`
`
`15
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`commercial and developmental aspects of products”). The Four-Steps Whitepaper
`
`“is intended for system and network administrators in organizations that manage
`
`their own applications and network performance.” EX1006 at 3. And the
`
`whitepaper concludes with the following invitation, further confirming the public,
`
`promotional nature of the document:
`
`If you’d like more information about PacketShaper or to order
`PacketShaper, consult Packeteer’s web site or call [Packeteer’s local
`or toll free telephone number].
`
`EX1006, 30. “[W]here a publication’s purpose is ‘dialogue with the intended
`
`audience,’ that purpose indicates public accessibility.” Valve v. Ironburg
`
`Inventions, 8 F.4th 1364, 1374 (Fed. Cir. 2021) (citing Suffolk Techs. v. AOL, 752
`
`F.3d 1358, 1365 (Fed. Cir. 2014)).10
`
`
`10 The author of the Four-Steps Whitepaper confirms that it was a promotional
`
`document that was on Packeteer’s website as of October 2002 and was broadly
`
`distributed to potential PacketShaper customers. EX1098, ¶¶6-10. The Board
`
`should consider the author’s declaration because it presents evidence that
`
`directly responds to evidence and arguments in Patent Owner’s response. See,
`
`e.g., POR at 42; see also Everstar Merch. v. Willis Elec., 2022 WL 1089909, at
`
`*4 (Fed. Cir. Apr. 12, 2022) (“Any ambiguity as to whether a reply constitutes a
`
`new argument is eliminated when the reply is a legitimate reply to arguments
`
`
`
`16
`
`

`

`Case IPR2021-00909—Petitioners’ Reply
`
`Packeteer’s PacketShaper product would have been of interest to a POSA
`
`implementing Yung’s bandwidth management device because Yung explicitly
`
`identifies PacketShaper as an existing bandwidth management device capable of
`
`identifying different types of network traffic. EX1005, 4:47-51. A POSA would
`
`have been aware of Packeteer and its products. EX1003, ¶¶41-43; EX2007, 45:21-
`
`48:19, 52:17-54:7. Given Yung’s suggestion, a POSA could simply visit the
`
`Packeteer website, locate product information about PacketShaper, and access the
`
`Four-Steps Whitepaper describing it. See GoPro, 908 F.3d at 693 (“We have
`
`interpreted § 102 broadly, finding that even relatively