`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`JUNIPER NETWORKS, INC.
`Petitioner
`
`v.
`
`IMPLICIT, LLC
`Patent Owner
`
`Case: IPR2020-00587
`Patent No. 9,591,104
`
`
`
`
`
`
`
`
`
`PRELIMINARY RESPONSE TO PETITION FOR INTER PARTES
`REVIEW OF U.S. PATENT NO. 9,591,104
`
`
`
`
`
`
`
`
`
`
`TABLE OF CONTENTS
`
`
`
`
`
`
`
`
`
`
`
`
`
`THE IMPLICIT PATENTS ............................................................................ 1
`
`
`
`THE PROSECUTION HISTORY OF THE IMPLICIT PATENTS .............. 4
`
`I.
`
`II.
`
`III. CLAIM CONSTRUCTION ............................................................................ 7
`
`IV. APPLICABLE LAW ...................................................................................... 7
`
`V.
`
`THE BOARD SHOULD DECLINE TO INSTITUTE PROCEEDINGS
`UNDER SECTION 325(D) ............................................................................. 8
`
`
`
`
`
`
`
`
`
`
`
`
`A.
`
`B.
`
`C.
`
`D.
`
`E.
`
`F.
`
`Factor (a): The Patent Office Has Already Considered Decasper ........ 9
`
`Factor (b): The “New” CheckPoint Reference and the Smith
`Reference Are, At Most, Cumulative of Other Examined Art ........... 10
`
`Factor (c): The Asserted Prior Art Was Highlighted During
`Examination of the Implicit Patents .................................................... 14
`
`Factor (d): The Patent Office Considered the Prior Art in the Same
`Light That Petitioner Asserts Here ..................................................... 16
`
`Factor (e): Petitioner Does Not Point to Any Examination Error ....... 18
`
`Factor (f): Petitioner Does Provide Sufficient Additional Evidence .. 19
`
`
`
`
`
`
`VI. CONCLUSION ............................................................................................. 19
`
`
`
`
`
`i
`
`
`
`
`
`TABLE OF AUTHORITIES
`
`Cases
`
`Becton, Dickinson & Co. v. B. Braun Melsungen AG
`
`IPR2017-01586, Paper 8 (PTAB Dec. 15, 2017) ................................... passim
`
`Invantis, Inc. v. Glaukos Corp.
`No. IPR2019-00483 (Paper No. 8), 2019 PAT. APP. LEXIS 10371
`(P.T.A.B. July 8, 2019) ................................................................................. 13
`
`
`Puma N. Am., Inc. v. Nike Inc.
`
`IPR2019-01042, Paper No. 10, 2019 Pat. App. LEXIS 12352
`
`(P.T.A.B. Oct. 31, 2019) ................................................................... 10, 18, 19
`
`Statutes
`
`35 U.S.C. § 325(d) ............................................................................................ passim
`
`
`
`
`
`ii
`
`
`
`
`
`
`
`
`
`
`
`
`Exhibit
`
`2001
`
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`2008
`2009
`2010
`
`2011
`
`2012
`
`2013
`
`EXHIBIT LIST
`
`
`Description
`Request for Inter Partes Reexamination of U.S. Patent No.
`6,629,163
`Decision Instituting Inter Partes Reexamination of U.S. Patent
`No. 6,629,163
`Action Closing Prosecution in Inter Partes Reexamination of
`U.S. Patent No. 6,629,163
`Decision Vacating Inter Partes Reexamination of U.S. Patent
`No. 6,629,163
`Request for Inter Partes Reexamination of U.S. Patent No.
`7,711,857
`Decision Instituting Inter Partes Reexamination of U.S. Patent
`No. 7,711,857
`Right of Appeal Notice in Inter Partes Reexamination of U.S.
`Patent No. 7,711,857
`Excerpts of File History of U.S. Patent No. 8,694,683
`September 1995 CheckPoint FireWall-1 White Paper
`July 1994 CheckPoint FireWall-1 Technical White Paper
`IBM Local Area Network Concepts and Products: Routers and
`Gateways
`Active Gateway: A Facility for Video Conferencing Traffic
`Control
`Excerpts of October 3, 2017 Deposition of Daniel Decasper
`
`
`iii
`
`
`
`
`
`Patent Owner Implicit, LLC (“Implicit”) opposes institution of Inter Partes
`
`Review on all grounds. The Patent Office has already thoroughly examined the
`
`Implicit Patents. And petitioner Juniper Networks, Inc. (“Petitioner” or “Juniper”)
`
`raises subject matter that the Patent Office extensively considered during
`
`examination of the Implicit Patents and their family.
`
`For that reason, the Board’s Becton, Dickinson factors weigh in favor of
`
`exercising discretion under Section 325(d) and denying this Petition. The purpose
`
`of Inter Parties Review proceedings is to review patents that may have lacked an
`
`adequate prior art record during prosecution or whose examination may not have
`
`provided a sufficient opportunity for a thorough review. The Implicit Patents are
`
`not those type of patents. For these reasons, and those below, Implicit respectfully
`
`requests that the Board decline to institute the proceedings.1
`
`I.
`
`THE IMPLICIT PATENTS
`
`The Petitions involve six related patents that are referred to collectively as the
`
`Demultiplexing Patents (or “the Implicit Patents”).2 Each Patent shares a common
`
`specification and is part of a large Demultiplexing Patent family. The Patents
`
`provide an architecture for demultiplexing data into different messages (or flows) to
`
`
`1 Should the Board institute proceedings, Implicit reserves its rights to challenge the
`Petition on any basis, including whether the cited references are prior art.
`2 The Implicit Patents here are U.S. Patent Nos. 8,694,683 (“the ’683 Patent”);
`9,270,790 (“the ’790 Patent”); 9,591,104 (“the ’104 Patent”); 10,027,780 (“the ’780
`Patent”); 10,033,839; (“the ’839 Patent”) and 10,225,378 (“the ’378 Patent”).
`
`
`
`1
`
`
`
`
`
`inspect and analyze each flow at the application layer. Traffic on a network (such
`
`as the Internet) is multiplexed together, which means that the traffic from different
`
`messages (such as video, image, or email) is combined and transmitted over a
`
`medium. Demultiplexing refers to the process of separating out that combined
`
`traffic back to identify and inspect packets from each message.
`
`Network traffic is transmitted in smaller units called “packets” that contain
`
`smaller pieces of the overall message. Computers communicate using a layered
`
`“stack” of protocols, and each packet has a number of headers that generally
`
`correspond to each protocol in the stack, e.g., TCP/IP is a common protocol stack
`
`for Internet communications that uses the TCP and IP protocols. The data itself (e.g.,
`
`the video, image, or email) is on top of the protocol stack in the packet. As the
`
`packets travel along a network like the Internet, packets from multiple messages are
`
`mixed together on the network, resulting in a multiplexed stream of packets. A
`
`multiplexed stream of packets for three different messages—video, email, and an
`
`image, is shown in the figure below:
`
`
`
`
`
`2
`
`
`
`
`
`
`
`The Implicit Patents provide a path-based architecture that inspects the
`
`packets at the application layer to demultiplex a received multiplexed stream. As a
`
`packet arrives, the software inspects information in that packet in the lower-level
`
`protocols. Packets in a message are typically associated together into a message. If
`
`the packet is for a message that has not yet been identified, the software creates the
`
`path for that message.
`
`The software routines then inspect the packet at the higher-level protocols and
`
`continue to create the “path” as the software inspects and identifies the application-
`
`level protocol for the message (e.g., HTTP) and the application data (e.g., Netflix).
`
`This example process is shown in the annotated version of Figure 1 of the Implicit
`
`Patents for Netflix traffic.
`
`The process to create the path requires more complicated programming and
`
`
`
`3
`
`
`
`
`
`
`
`
`
`analysis at the application protocol and application stages. The Implicit Patents
`
`overcome this speed problem by providing a way to make the complicated branching
`
`decisions for the initial packet or packets of the flow, store that path, and then use
`
`that path when the software processes subsequent packets of the flow.
`
`After the software identifies the path for a message, it stores that path for that
`
`message. Then, the software processes subsequent packets using the path that it has
`
`already built to handle that message. This processing is fast because it does not need
`
`to re-create the path. The example below from Figure 4 of the Patents shows the
`
`stored path data structure (typically a flow table) for three messages—a Netflix
`
`video, an email for Gmail, and an image for Instagram:
`
`
`II. THE PROSECUTION HISTORY OF THE IMPLICIT PATENTS
`
`
`
`The Implicit Patent family has a long examination history, and the Patents cite
`
`over 300 references on their face. One of those references is Decasper, which the
`
`
`
`4
`
`
`
`
`
`Petition relies on for both proposed grounds. Another reference is CheckPoint
`
`FireWall-1 (disclosed in two references).
`
`The relevant history begins with two Inter Partes Reexamination proceedings
`
`of two patents within the Implicit Patent family, U.S. Patent Nos. 6,629,163 and
`
`7,711,857. The ’163 Patent is the parent patent of each Implicit Patent, and the ’163
`
`Patent is the parent patent of the ’857 Patent. The Implicit Patents have the same
`
`disclosure as the ’163 and ’857 Patents.
`
`In 2012, Juniper petitioned to institute Inter Partes Reexamination of the ’163
`
`Patent on forty-six different grounds based on approximately thirty references,
`
`including Decasper and a Checkpoint FireWall-1 reference. Ex. 2001, at 10–133;
`
`Ex. 2002, at 4–22. The Patent Office granted review on two grounds related to
`
`Decasper (among other grounds, but not FireWall-1). Ex. 2002, at 57–81. The
`
`Patent Office considered Decasper through multiple Office Actions and an Action
`
`Closing Prosecution. Ex. 2003, at 12–45, 95–105, 120–139. After the ’163 Patent
`
`was found invalid in a final court decision, the Patent Office vacated the proceedings.
`
`Ex. 2004, at 4–5.
`
`Juniper also petitioned to institute Inter Partes Reexamination of the ’857
`
`Patent in 2012 based on sixty-one different grounds based on approximately thirty
`
`
`3 All citations are to the Exhibit page numbers for each exhibit in the bottom left-
`hand corner.
`
`
`
`5
`
`
`
`
`
`references, including Decasper and FireWall-1. Ex. 2005, at 12–15; Ex. 2006, at 4–
`
`5, 99. The Patent Office instituted proceedings based on Decasper, among other
`
`references (but not FireWall-1). Id. at 100–101. The Patent Office considered
`
`Decasper through multiple Office Actions and a Right of Appeal Notice Ex. 2007,
`
`at 6–15 (describing eight rejections based on Decasper). The ’857 Patent was found
`
`invalid in a court decision. See id. at 36–37.
`
`In June, 2013, the earliest of the Implicit Patents in this case (the ’683 Patent)
`
`in prosecution before the Patent Office. Based on the reexaminations, Implicit
`
`amended the claims. Ex. 2008, at 64. Specifically, Implicit amended the claims to
`
`add a limitation to expressly capture the inventions’ ability to inspect network traffic
`
`at the application layers and above: the claims required functionality to convert
`
`packets from a transport layer format (e.g., TCP or UDP) to another format (e.g., an
`
`application-layer format, such as HTTP). Id. at 59–63.
`
`Implicit then addressed Decasper in significant detail, including addressing
`
`why Decasper lacked the limitations in Implicit’s claims. Ex. 2008, at 69–77.
`
`Implicit specifically referenced the reexamination proceedings in its remarks. Id. at
`
`64. The Patent Office also considered the reexamination record. Id. at 13–14; ’683
`
`Patent, at [56] (listing filings in 95/000,659 and 95/000,660 proceedings). The
`
`Patent Office, after initially rejecting the claims based on other references (Amir and
`
`Taylor), allowed the claims. Ex. 2008, at 16, 46–56.
`
`
`
`6
`
`
`
`
`
`The remaining Implicit Patents (the ’790, ’104, ’780, ’839, and ’378 Patents)
`
`issued from a chain of continuation applications of the ’683 Patent and also cite the
`
`references that the ’683 Patent cites. The Implicit Patents each contain limitations
`
`directed to the ability to inspect network traffic at the application layers and above,
`
`similar to the ’683 Patent.
`
`III. CLAIM CONSTRUCTION
`
`The Board does not need to construe any claim elements to deny the Petition
`
`for the reasons in this Preliminary Patent Owner Response. While Patent Owner
`
`does not agree with the Petition’s proposed claim constructions, for the purposes of
`
`this Preliminary Response only, Patent Owner does not contest those claim
`
`constructions.4
`
`IV. THE APPLICABLE LAW
`
`The Board has discretion under Section 325(d) to deny institution. Section
`
`325(d) states that, in exercising this discretion, the Director may “take into account
`
`whether, and reject the petition or request because, the same or substantially the same
`
`prior art or arguments previously were presented to the Office.” 35 U.S.C. § 325(d).
`
`
`4 Patent Owner reserves the right to contest any of Petitioner’s proposed claim
`constructions and propose its own constructions in this proceeding, should review
`be instituted, or in any other proceeding.
`
`
`
`7
`
`
`
`
`
`The Board has outlined six factors to consider in deciding whether to apply
`
`Section 325(d):
`
`(a) the similarities and material differences between the
`asserted art and the prior art involved during
`examination;
`
`(b) the cumulative nature of the asserted art and the prior
`art evaluated during examination;
`
`(c) the extent to which the asserted art was evaluated
`during examination, including whether the prior art
`was the basis for rejection;
`
`(d) the extent of the overlap between the arguments made
`during examination and the manner in which Petitioner
`relies on the prior art or Patent Owner distinguishes the
`prior art;
`
`(e) whether Petitioner has pointed out sufficiently how the
`Examiner erred in its evaluation of the asserted prior
`art; and
`
`(f) the extent to which additional evidence and facts
`presented in the Petition warrant reconsideration of the
`prior art or arguments.
`
`
`
`
`
`
`
`
`
`
`Becton, Dickinson & Co. v. B. Braun Melsungen AG, IPR2017-01586, Paper 8 at
`
`17-18 (P.T.A.B. Dec. 15, 2017) (precedential in relevant part).
`
`V. THE BOARD SHOULD DECLINE TO INSTITUTE PROCEEDINGS
`UNDER SECTION 325(D)
`
`Implicit respectfully requests that the Board decline to institute proceedings
`
`under Section 325(d). The Patent Office has already expended significant resources
`
`examining the family of the Implicit Patents and allowed these claims. The Petition
`
`
`
`8
`
`
`
`
`
`asserts two grounds, both of which retread this prior work: (1) obviousness based on
`
`Decasper and CheckPoint; and (2) obviousness based on Decasper and Smith.
`
`The Becton-Dickinson factors, outlined below, weigh against expending
`
`additional resources to re-review these Patents based on these grounds. Juniper
`
`knew that Section 325(d) would be a significant issue—Juniper filed both prior
`
`reexamination requests and acknowledges that the Patent Office “substantially”
`
`considered the Petition’s central reference, Decasper. Petition, at 15.5
`
`Despite that, Juniper addressed Section 325(d) in only two-and-a-half pages.
`
`And, tellingly, Juniper sidesteps the role that Decasper played during reexamination
`
`and examination. And it omits that the Patent Office considered the CheckPoint
`
`FireWall-1 reference that Juniper itself raised in prior proceedings. It now asserts
`
`that the Patent Office did not consider FireWall-1. Petition, at 15. But peeling back
`
`the actual examination record reveals that the Petition seeks to reopen prior art
`
`teachings that the Patent Office has already considered. Subjecting this family of
`
`thoroughly-examined patents to further review is not appropriate in this case.
`
`A.
`
`Factor (a): The Patent Office Already Considered Decasper
`
`The first Becton, Dickinson factor weighs against institution. Both grounds
`
`in the Petition hinge on Decasper (in combination with FireWall-1 or Smith). It is
`
`
`5 Because each of the Petitions contains the same statements cited to in this
`Response, Implicit cited only the Petition in the IPR2020-00585 proceedings for
`consistency.
`
`
`
`9
`
`
`
`
`
`undisputed, however, that the Patent Office “substantially” considered Decasper
`
`when it examined the Implicit Patents. See Petition, at 15. And the reference played
`
`an outsized role during examination and reexamination of the Implicit Patent family.
`
`Multiple reexamination filings and Implicit’s extensive remarks during prosecution
`
`put the reference front-and-center.
`
`Because it is undisputed that the Patent Office considered Decasper—the
`
`reference on which both grounds hinge—this factor weighs in favor denying
`
`institution of the entire Petition. See, e.g., Puma N. Am., Inc. v. Nike Inc., IPR2019-
`
`01042, Paper No. 10, at 10, 2019 Pat. App. LEXIS 12352 (P.T.A.B. Oct. 31, 2019)
`
`(concluding, where parties agreed that the prior art in the Petition was identical to
`
`that involved during examination that “this factor weighs in favor of exercising
`
`discretion to deny institution”).
`
`B.
`
`Factor (b): The “New” FireWall-1 Reference and the Smith
`Reference Are, At Most, Cumulative of Other Examined Art
`
`The second factor also weighs against institution. Juniper asserts that the
`
`Patent Office did not consider CheckPoint FireWall-1. Petition, at 14. That is
`
`incorrect. The Implicit Patents list two FireWall-1 references on their face. See,
`
`e.g., ’683 Patent, at [57] (disclosing CheckPoint FireWall-1 Technical White Paper,
`
`Jul. 18, 1994, CheckPoint Software Technologies, Ltd. and CheckPoint FireWall-1
`
`White Paper, Sep. 1995, Version 2.0, CheckPoint Software Technologies, Ltd.); Ex.
`
`2008, at 3. Juniper is aware of that fact: Juniper sought reexamination of two
`
`
`
`10
`
`
`
`
`
`(broader) patents in the Implicit Patent family based on 1995 FireWall-1 White
`
`Paper. Ex. 2001 at 12, 165–167; Ex. 2005 at 14, 273–275.
`
`For this Petition, Juniper created a new Checkpoint reference. It combined a
`
`smattering of webpages from the Wayback Machine to create the reference. See
`
`Petition, at 16–17. But those webpages are, at most, cumulative of the reference that
`
`Juniper submitted during reexamination and the references before the Patent Office
`
`during examination. In particular, Juniper’s Petition relies on the alleged disclosure
`
`of application-layer processing in the new FireWall-1 reference, such as for
`
`protocols like HTTP, FTP, and SMTP and for anti-virus and security.
`
`The FireWall-1 references the Patent Office considered and Juniper omitted
`
`from its Petition disclose the same subject matter, including, for example:
`
`•
`
`•
`
`•
`
`•
`
`“Firewall-1 secures complete applications more effectively than
`technologies that only have data in some of the layers available to them.
`For example, while application gateways have access only to the
`application layer and routers have access only to the lower layers,
`Firewall-1 integrates the information gathered from all layers into a
`single comprehensive inspection point.” Ex. 2009, at 3.
`
`“Because the Inspection Module has access to the ‘raw message,’ so to
`speak, it can inspect all the information in the message, including
`information relating to all the higher layers, as well as the message data
`itself (the application content of the packet).” Ex. 2009, at 2.
`
`security, user
`application-layer
`capable of
`is
`“FireWall-1
`authentication, unified support and handling of all protocols, and
`auditing and alerting.” Ex. 2009, at 2.
`
`“An efficient inspection module – applicable to all protocols – with
`Stateful Multi-Layer Inspection (SMLI) technology, understands data
`
`
`
`11
`
`
`
`
`
`in the packet intended for all other layers, from the network layer (IP
`headers) up to the application layer, and provides stateful context.” Ex.
`Ex. 2009, at 3.
`
`“FireWall-1 is able to extract data from a packet’s application content
`and store it to provide context in those cases where the application does
`not provide it . . . . FireWall-1’s ability to look inside a packet enables
`it to allow certain commands within an application while disallowing
`others.” Ex. 2009, at 4.
`
`“The Firewall Module can inspect all the information in the message,
`and understands the internal structures of the IP protocol family and
`applications built on top of them. FireWall-1 is able to extract data
`from the packet’s application content and store it to provide context in
`those cases where the application does not provide it. FireWall-1 can
`perform any logical or arithmetic operation required to implement the
`security policy.” Ex. 2009, at 20.
`
`“Full packet context is available for screening, from lower layers (like
`interface and addresses’ data) up to the higher application layers (layers
`2 to 7 data in the ISO model).” Ex. 2010, at 6.
`
`“[T]he application information is extracted from the packet in order to
`identify the program used, overcoming the packet’s varying structure.”
`Ex. 2010, at 7.
`
`•
`
`•
`
`•
`
`•
`
`The references also discuss examples uses relating to SMTP, FTP, HTTP,
`
`User Authentication, and Security. E.g., Ex. 2009, at 5–10, 12; see also id. at 14
`
`(disclosing that FireWall-1 can “allow SMTP gets while disallowing sets, and so
`
`on”); id. at 15 (disclosing HTTP Authenticating Proxy); id. at 16 (“FireWall-1
`
`employs a single integrated security policy to establish enterprise-wide security,
`
`authenticate clients, and encrypt communications.”); id. at 19 (“FireWall-1 tracks
`
`the FTP session, examining FTP application layer data”); Ex. 2010, at 7 (same); id.
`
`
`
`12
`
`
`
`
`
`at 8 (discussing “Trojan Horses and Viruses”); id. at 10–11 (discussing Services
`
`Manager, which supports FTP, SMTP, HTTP, and Telnet, among others); id. at 13
`
`(explaining that CheckPoint’s “innovative packet filtering technology brings
`
`application level security capabilities into an efficient packet filtering engine”).
`
`The old FireWall-1 references that the Patent Office considered thus have the
`
`same relevant disclosure as Juniper’s new one. This makes sense: they are
`
`describing the same firewall. This factor weighs against instituting proceedings.
`
`See, e.g., Invantis, Inc. v. Glaukos Corp., No. IPR2019-00483 (Paper No. 8), at 18–
`
`20, 2019 PAT. APP. LEXIS 10371 (P.T.A.B. July 8, 2019) (concluding that this
`
`factor “weighs strongly in Patent Owner’s favor” where the references considered
`
`had the same disclosure relied on in the Petition and “the prosecution history also
`
`cites other disclosures that are similar” to the prior art in the Petition).
`
`The Smith reference is also cumulative of references the Patent Office
`
`considered. Juniper is correct that the Patent Office did not have the Smith reference
`
`before it during examination. Petition, at 14. But Juniper relies on Smith for a
`
`disclosure cumulative of multiple references the Patent Office did consider: firewall
`
`gateway functionality “to process various application-layer protocols,” id. at 21,
`
`including HTTP, SMTP, FTP, and Telnet.
`
`This disclosure is cumulative of the FireWall-1 references, which discuss
`
`HTTP, SMTP, FTP, and other protocols (such as TELNET) and using encryption.
`
`
`
`13
`
`
`
`
`
`And the references disclose that functionality as part of the “CheckPoint FireWall-1
`
`Internet Gateway.” See Ex. 2010, at 11 (“This document describes the architecture
`
`and unique characteristics of CheckPoint’s FireWall-1 Internet gateway security
`
`system.”); id. at 4 (“The CheckPoint FireWall-1 Internet Gateway”).
`
`Beyond FireWall-1, more references that were before the Patent Office show
`
`that Smith is cumulative. See, e.g., Ex. 2011, at 14 (IBM96) (disclosing that
`
`gateways “are considered application layer relays between network environments”
`
`and “must implement all seven layers of the OSI Reference Model”); Ex. 2012 at 1
`
`(Li) (“We present the architecture and the design of an application-level facility,
`
`called active gateway, for video conferencing traffic and quality of service (QoS)
`
`control.”). This factor weighs against instituting proceedings based on Smith.
`
`C.
`
`Factor (c): The Asserted Prior Art Was Highlighted During
`Examination of the Implicit Patents
`
`The Patent Office evaluated the subject matter of the Petition during
`
`examination. Both grounds hinge on Decasper. The Patent Office extensively
`
`considered that reference. That includes two through reexaminations of the Implicit
`
`Patent family, which contained significant discussion of Decasper through multiple
`
`office actions and responses.
`
`Implicit then placed Decasper front-and-center during prosecution of the
`
`Implicit Patents (specifically, the ’683 Patent). Implicit amended the claims to
`
`require functionality to convert packets from a transport layer format (e.g., TCP or
`
`
`
`14
`
`
`
`
`
`UDP) to another format (e.g., an application layer format, such as HTTP or FTP).
`
`Each Implicit Patent claim recites that functionality. Implicit then addressed why
`
`Decasper did not render unpatentable Implicit’s claims. Implicit specifically
`
`referenced the reexamination proceedings. And the Patent Office considered
`
`reexamination file history during prosecution. See Ex. 2008, at 13–14, 69–77.
`
`The Patent Office reviewed that record and accepted Implicit’s explanation.
`
`While it initially rejected the claims, it did so based on other references (Amir and
`
`Taylor). Ex. 2008, at 46–56. The Patent Office then allowed the claims. Id. at 16.
`
`The Office then allowed the claims of each of the remaining Implicit Patents. That
`
`put Decasper to bed. This history weighs against resurrecting Decasper. It weighs
`
`in favor of denying both grounds in the Petition.
`
`The Patent Office also has experience with FireWall-1. Juniper raised
`
`FireWall-1 in its reexamination requests. The Patent Office rejected Juniper’s
`
`arguments. Ex. 2002, at 75–76; Ex. 2006, at 94–96. The Patent Office also
`
`considered the FireWall-1 references during prosecution of the Implicit Patents. See
`
`Ex. 2008, at 3. And it considered the reexamination histories. Id. at 13–14.
`
`The Patent Office (correctly) did not focus on FireWall-1 when examining the
`
`Implicit Patents: the Office already cast those references aside during reexamination.
`
`There was no need to later re-plow that ground during subsequent examination of
`
`the Implicit Patents (which are narrower). And there is no reason to re-plow that
`
`
`
`15
`
`
`
`
`
`ground here based on Juniper’s newly-created FireWall-1 reference. This weighs
`
`against instituting Juniper’s proposed ground based on FireWall-1.
`
`Lastly, Smith is cumulative of references that the Patent Office considered in
`
`detail. Smith is cumulative of at least the four references above: the two CheckPoint
`
`FireWall-1 references, IBM96, and Li. See, e.g., Ex. 2001, at 12 (showing IBM96
`
`and Checkpoint in Juniper reexamination request); Ex. 2005 (showing Li in Juniper
`
`reexamination request). Aware of those references and the reexamination history,
`
`the Patent Office rightly did not focus on them during prosecution of the Implicit
`
`Patents and allowed the claims. This weighs against instituting Juniper’s proposed
`
`ground based on Smith.
`
`D.
`
`Factor (d): The Patent Office Considered the Prior Art in the Same
`Light That Petitioner Asserts Here
`
`This factor also weighs against instituting proceedings in this case. Juniper
`
`presents Decasper in a similar light here that it did in the two reexaminations. Like
`
`its Petition here, Juniper specifically asserted that Decasper rendered obvious claims
`
`of the Implicit Patent family because it could be implemented in a firewall.
`
`Compare, e.g., Ex. 2001, at 202 (“Decasper98 teaches ‘a firewall plugin.’”) (citation
`
`omitted); id. at 243–244 (“Decasper98 was an especially obvious candidate for this
`
`technique, because Fraser uses the technique to control the policies of ‘application
`
`gateway firewall[s],’ and Decasper98 teaches an architecture that is ‘very well suited
`
`to Application Layer Gateways . . . and to security devices like Firewalls.’”)
`
`
`
`16
`
`
`
`
`
`(citations omitted); Ex. 2005, at 83 (same) with Petition, at 20 (“As detailed below,
`
`it was obvious to apply Decasper to Smith, particularly since Decasper explains that
`
`its “framework is also very well suited to Application Layer Gateways (ALGs), and
`
`to security devices like Firewalls.”) (emphases in original); id. at 50 (“Since
`
`Decasper explains its ‘framework is also very well suited to . . . security devices like
`
`Firewalls,’ it was obvious to apply Decasper’s framework to CheckPoint’s
`
`FireWall-1.”) (emphasis in original).
`
`Juniper also presented FireWall-1 in a similar light. As it does here, Juniper
`
`highlighted the reference’s application-level disclosure as rendering obvious the
`
`claims: “Checkpoint explains that Firewall-1 is not limited to examining header data;
`
`it can ‘extract data from the packet’s application content and store it to provide
`
`context.’” Ex. 2001, at 166 (citation omitted, emphasis in original); Ex. 2005, at 274
`
`(same). The Patent Office did not institute proceedings on the reference.
`
`Implicit then pointed the Patent Office to the reexamination proceedings
`
`during examination of the Implicit Patents. It submitted Decasper and FireWall-1,
`
`among dozens of others. Implicit specifically addressed Decasper. It explained that
`
`the Decasper system did not contain functionality at the TCP layer or above because
`
`it only operated at the IP layer, Ex. 2008, at 69–76—firewall plugin or not. After
`
`this history, the Office allowed the claims of the Implicit Patents (all six of them).
`
`Juniper’s Petition cobbles together disclosures the Patent Office had before it
`
`
`
`17
`
`
`
`
`
`during examination, directly and through the reexamination records. The Patent
`
`Office considered these disclosures in connection with the Implicit Patents and
`
`allowed the claims. While the Office did not issue the exact Section 103(a)
`
`rejections on the grounds that Juniper now proposes, it had the relevant disclosed
`
`matter to do so.6 Implicit directed the Office to the reexamination proceedings. It
`
`addressed Decasper in detail. It submitted the FireWall-1 references (and other
`
`firewall and gateway references, of which Smith is cumulative). The Office
`
`considered that record and allowed the claims. This factor weighs against institution.
`
`E.
`
`Factor (e): Petitioner Does Not Point to Any Examination Error
`
`Juniper does not point to any examination error in its Petition. Petition, at 15.
`
`It is undisputed that the Patent Office reviewed Decasper during examination. And
`
`the Patent Office also considered the subject matter from Juniper’s “new” FireWall-
`
`1 reference and Smith for the points Juniper relies on in its Petition. Because Juniper
`
`“has not explained how the examiner may have erred in the evaluation of the prior
`
`art during prosecution,” this factor weighs in favor of denying institution. Puma,
`
`
`6 The FireWall-1 reference that Juniper omitted from its Petition shows why the
`Patent Office did not issue Juniper’s new Section 103(a) rejections. “Decasper is
`directed to ‘a high performance, modular, extended integrated services router
`software architecture.’” Ex. 2008, at 69 (citation omitted). FireWall-1 itself
`explains that “[r]outers do not implement many of the essential capabilities required
`to provide secure packet filtering (e.g. they are stateless, look only at TCP/UDP port
`numbers; see previous sections for details).” Ex. 2010, at 12–13; see also Ex. 2013,
`at 10 (deposition testimony of Dr. Decasper that “I don’t know why you would
`[implement the TCP protocol] in a router”).
`
`
`
`18
`
`
`
`
`
`IPR2019-01042, Paper No. 10, at 23–24, 2019 Pat. App. LEXIS 12352.
`
`F.
`
`Factor (f): Petitioner Does Provide Sufficient Additional Evidence
`
` The Petition hinges on Decasper, which undisputedly was considered in
`
`depth during prosecution. Juniper claims that its Petition contains additional
`
`evidence in the form of “other references” besides Decasper “for the allegedly
`
`missing limitations.” Petition, at 15. But those “other references” were before the
`
`Patent Office in substance—and FireWall-1 was also before the Office by name. At
`
`a minimum, there is substantial overlap between what was before Patent Office
`
`during examination of the Implicit Patents and the Petition. Without any subject
`
`matter beyond what the Patent Office has already considered in detail, this factor
`
`weighs against institution. See, e.g., Puma, IPR2019-01042, Paper No. 10, at 24,
`
`2019 Pat. App. LEXIS 12352 (“Petitioner does not make any argument that the
`
`infor