`
`
`
`
`
`
`
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SLING TV, L.L.C., et al.,
`Petitioner
`
`v.
`
`REALTIME ADAPTIVE STREAMING LLC,
`Patent Owner
`
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2018-01342
`Patent 8,934,535
`
`
`
`
`
`
`
`
`
`
`
`Before KEVIN W. CHERRY, GARTH D. BAER, and NABEEL U. KHAN,
`Administrative Patent Judges.
`
`
`
`
`PETITIONER’S MOTION TO FILE SUPPLEMENTAL INFORMATION
`PURSUANT TO 37 C.F.R. § 42.123
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`EXHIBITS
`
`DISH1029
`
`Supplemental Declaration of Dr. Scott Acton
`
`i
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`I. BOARD AUTHORIZATION AND LEGAL STANDARD
`
`Pursuant to Petitioner’s request under 37 C.F.R § 42.123(a), made on
`
`February 27, 2019 within a month of institution, and further pursuant to the
`
`Board’s authorization granted on March 11, 2019, Petitioner moves to submit
`
`DISH1029 (“Supplemental Declaration of Dr. Scott Acton”) as supplemental
`
`information. The Board recognizes that “supplemental information may prove
`
`beneficial to the Board in reaching a decision with respect to the trial.” Pacific
`
`Market Int’l v. Ignite USA, LLC, IPR2014-00561, Paper 23 at 3 (PTAB Dec. 2,
`
`2014). The Board may grant a motion to submit supplemental information where,
`
`as here, limited expert testimony is provided to further support an instituted ground
`
`of unpatentability. See id.; see also, e.g., The Boeing Co. v. Seymour Levine,
`
`IPR2015-01341, Paper 30 at 4 (PTAB Apr. 15, 2016) (granting motion to submit
`
`supplemental information where the supplemental information was limited in
`
`scope and did not alter the ground in the petition but merely provided additional
`
`evidence of obviousness).
`
`The supplemental information meets all legal requirements for such material
`
`because it is timely and relevant to a claim in this trial. The supplemental
`
`information relates to instituted Ground 3 and its applicability under the Board’s
`
`preliminary claim construction of “access profile” in the Netflix proceeding,
`
`IPR2018-01169, institution decision. See 37 C.F.R. 42.123(a). The material
`
`provides additional evidence, presented in the alternative, regarding the narrower,
`
`1
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`preliminary claim construction from the Netflix IPR, and confirms the invalidity of
`
`the challenged claims under Ground 3 in this matter, even under that narrower,
`
`preliminary construction from the Netflix IPR. The supplemental information will
`
`cause no prejudice to Patent Owner or delay in the proceedings. Petitioner
`
`respectfully requests the Board grant Petitioner’s request.
`II. STATEMENT OF FACTS
`
`On July 3, 2018, Petitioner sought review of claims 1–6, 8–12, and 14 (the
`
`“challenged claims”) of U.S. Pat. No. 8,934,535 B2 (DISH1001, “the ’535
`
`Patent”). IPR2018-01342, Paper No. 9 at 1. Petitioner submitted the declaration
`
`of Dr. Scott Acton (DISH1003) in support of its Petition.
`
`Petitioner challenges claims 1–6, 8–12, and 14 on the following grounds:
`Ground
`Claims
`Basis for Rejection
`Ground 1 1, 2, 9, 10, 14 Anticipated by Dvir1, 35 U.S.C. § 102
`Ground 2 1, 2, 9, 10, 14 Obvious over Dvir, 35 U.S.C. § 103
`Ground 3 3-6, 8, 11, 12 Obvious over Dvir in view of Ishii2, 35 U.S.C. § 103
`
`IPR2018-01342, Paper No. 9 at 8. Petitioner argued that the specification-
`
`supported, broadest reasonable interpretation (“BRI”) of the claim term “access
`
`profile” is “information that enables a controller to determine a compression
`
`routine that is associated with a data type of the data to be compressed.” Id. at 19-
`
`21 (citing certain portions of the ’535 Patent specification).
`
`
`
`1 DISH1004 – Dvir, U.S. Pat. No. 6,557,001
`
` 2
`
` DISH1005 – Ishii, U.S. Pat. No. 5,675,789
`
`2
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`On January 17, 2019, six months after Petitioner filed its petition, the Board
`
`granted a separate request for inter partes review of claims 1-14 of the ’535 Patent
`
`submitted by Netflix, Inc. et. al. (“Netflix”). IPR2018-01169, Paper No. 20 at 2.
`
`Netflix challenged claims 1-14 on the following grounds:
`Ground
`Claims
`Basis for Rejection
`Ground 1
`1-14
`Obvious over Imai3 and Ishii
`
`Id. at 8. Netflix’s Ground 1 and Petitioner’s Ground 3 rely on the Ishii reference.
`
`On January 31, 2019, the Board instituted a trial in this proceeding,
`
`IPR2018-01342, on all challenged claims, determining that it is more likely than
`
`not that Petitioner will prevail in showing unpatentability of the challenged claims.
`
`Id. at 2, 20-21. The Board did not construe the term “access profile” because it
`
`determined that no claim construction was necessary to institute. Id. at 10 (noting
`
`that the parties are still “expected to assert all their claim construction arguments
`
`and evidence . . . as permitted by our rules.” (emphasis added)).
`
`However, in the earlier institution of the Netflix petition, the Board
`
`preliminarily construed the claim term “access profile” to mean “information
`
`regarding the number or frequency of reads or writes.” IPR2018-01169, Paper No.
`
`20 at 12. It noted that a “final determination as to claim construction will be made
`
`at the close of the proceeding, after any hearing, based on all the evidence of
`
`record.” Id. at 10.
`
`On a review of that record, the Board found that Netflix had demonstrated a
`
`
`
`3 Imai, Japanese Patent App. Pub. No. H11331305, published Nov. 30, 1999.
`
`3
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`reasonable likelihood that Imai and Ishii teach or suggest the limitations of claims
`
`1-14 of the ’535 Patent. Id. at 21-22. In particular, it found that Ishii taught an
`
`“access profile,” as required by claims 1-14, under the construction “information
`
`regarding the number or frequency of reads or writes.” Id. at 20.
`
`Petitioner’s Supplemental Declaration of Dr. Scott Acton, DISH1029,
`
`provides further support for Petitioner’s alternative position here that Ground 3
`
`invalidates the challenged claims under the narrower construction preliminarily
`
`adopted in the Netflix IPR. For example, Patent Owner argued in the Netflix
`
`proceeding that the Ishii reference did not meet the “access profile” limitation.
`
`The Board, however, found that it did under at least one preliminary construction
`
`of that term. Dr. Acton’s supplemental declaration provides additional evidence
`
`that the combination of Dvir with Ishii would likewise meet the “access profile”
`
`limitation, even under a narrower claim construction.
`III. ARGUMENT
`
`The Board declined to construe “access profile,” and instituted Ground 3 of
`
`the Petitioner’s July 3, 2018 inter partes review petition, determining “that the
`
`Petition shows a reasonable likelihood that Petitioner would prevail with respect to
`
`the contention that . . . claims 3-6, 8, 11, and 12 would have been obvious over
`
`Dvir and Ishii.” IPR2018-01342, Paper No. 9 at 20-21. In the Netflix IPR,
`
`however, the Board preliminarily construed “access profile” to mean “information
`
`regarding the number or frequency of reads or writes.” IPR2018-01169, Paper No.
`
`4
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`20 at 11-12. Petitioner here could not have addressed the Board’s January 17,
`
`2019 preliminary construction of “access profile” before now.
`
`In the present trial, the Board has determined that no preliminary
`
`construction is necessary for the term “access profile.” IPR2018-01342, Paper
`
`No. 9 at 10. Petitioner does not dispute the Board’s determination. Petitioner
`
`asserts, however, that in the event the Board eventually determines to construe the
`
`“access profile” term under its BRI, the proper construction is “information that
`
`enables a controller to determine a compression routine that is associated with a
`
`data type of the data to be compressed,” as asserted in the petition. Id., Paper No.
`
`2 at 19-20. Petitioner’s proposed construction is consistent with the Board’s
`
`findings on institution that the ’535 patent teaches that “[a] bandwidth sensitive
`
`data compression routine may be selected based on access profiles that enable the
`
`controller to determine a compression routine associated with a data type of the
`
`data to be compressed.” Id., Paper No. 9 at 6 (citing the ’535 Patent at 8:4-8). The
`
`BRI of “access profile,” therefore, encompasses at least this teaching from the ’535
`
`patent, consistent with the construction of “access profile” proffered by Petitioner
`
`here. See Corning v. Fast Felt Corp., 873 F.3d 896, 900-01 (Fed. Cir. 2017)
`
`(reversing Board without remand and holding that Board erroneously construed the
`
`“roofing or building cover material” term too narrowly under the BRI to exclude
`
`teachings from the patent’s specification).
`
`While the scope of Petitioner’s proposed construction of “access profile” is
`
`5
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`supported by the ’535 patent teachings, Petitioner nevertheless seeks to offer
`
`supplemental evidence in the form of a supplemental declaration from Dr. Scott
`
`Acton. DISH1029.4 Petitioner offers Dr. Acton’s supplemental declaration in the
`
`alternative only, as the declaration makes clear. See DISH1029, ¶ 223. That is,
`
`Petitioner continues to assert that its previously proposed construction of the
`
`“access profile” term offered with its petition is the proper BRI construction of
`
`“access profile.” Nevertheless, if the Board ultimately construes “access profile”
`
`more narrowly, as the Board preliminarily did in the Netflix IPR, then Petitioner
`
`asserts as an alternative argument under this narrower construction that Ground 3
`
`renders the claims of the ’535 patent unpatentable. See DISH1029.
`
`Dr. Acton’s supplemental declaration addresses Ground 3’s combination of
`
`Dvir in view of Ishii under the construction of “access profile” as “information
`
`regarding the number or frequency of reads or writes,” from the Netflix IPR
`
`proceeding. The combination renders the ’535 patent claims unpatentable and
`
`teaches claim element 1.1 because Ishii’s file compression portion 105 and file
`
`compression method selection portion 104 select an “access profile” from a
`
`plurality of access profiles corresponding to high-, medium-, and low-access
`
`
`
`4 Dr. Acton previously submitted a declaration in support of Petitioner’s petition in
`the instant matter. See DISH1003. There, Dr. Acton provided evidence of the
`unpatentability of the ’535 patent claims under Petitioner’s proposed construction
`of the term “access profile.”
`
`6
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`frequency. DISH1029, ¶¶ 226-233. Ishii teaches that its file compression portion
`
`105 actually compresses data of the searched file to be compressed by selecting an
`
`appropriate data compression method for compression of the file. Ishii selects an
`
`appropriate data compression method according to the attributes of the file to be
`
`compressed, including access frequency and file type. Id., ¶ 228. For both of
`
`Ishii’s elements 104 and 105, the “access frequency” corresponds to a data
`
`attribute of a file, namely the number of times the file has been accessed. Id.,
`
`¶ 231.
`
`Moreover, claim element 1.2 is taught by the same combination because in
`
`Ishii, selecting a compression method with suitable compression ratio and
`
`compression speed or decompression speed depends on the file access frequency
`
`and data attribute. Using this information, the file compression method with a
`
`shorter decompression time is selected for files with higher access frequency and
`
`the file compression method with a higher compression ratio is selected for files
`
`with lower access frequency. Id., ¶¶ 234-240.
`IV. CONCLUSION
`
`In view of the foregoing, Petitioner respectfully requests that supplemental
`
`information in DISH1029 be admitted into the record.
`
`
`
`
`
`
`
`7
`
`
`
`
`
`
`
`
`
`
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Respectfully submitted,
`
`
`
`
`/Adam R. Shartzer/
`Adam R. Shartzer, Reg. No. 57,264
`
`Fish & Richardson P.C.
`3200 RBC Plaza
`60 South Sixth Street
`Minneapolis, MN 55402
`T: 202-783-5070
`F: 877-769-7945
`
`Attorney for Petitioners
`
`
`
`
`
`
`
`
`
`
`
`
`
`Dated: March 21, 2019
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`8
`
`
`
`Attorney Docket: 45035-0002IP4
`Case IPR2018-01342
`
`CERTIFICATE OF SERVICE
`
`Pursuant to 37 CFR §§ 42.6(e), the undersigned certifies that on March 21,
`
`2019, a complete and entire copy of this Motion to File Supplemental Information
`
`Pursuant to 37 C.F.R. § 42.123 and its supporting exhibit were provided via email,
`
`to the Patent Owner by serving the email correspondence addresses of record as
`
`follows:
`
`
`
`
`
`
`
`
`
`Neil A. Rubin
`Kent Shum
`Russ August & Kabat
`12424 Wilshire Blvd., 12th Floor
`Los Angeles, CA 90025
`
`Email: nrubin@raklaw.com
`Email: kshum@raklaw.com
`Email: rak_realtimedata@raklaw.com
`
`
`
`
`
`
`
`/Edward G. Faeth /
`Edward G. Faeth
`Fish & Richardson P.C.
`60 South Sixth Street, Suite 3200
`Minneapolis, MN 55402
`(202) 626-6420
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`1
`
`
`
`PROPOSED
`
`PROPOSED
`EXHIBIT 1029
`
`EXHIBIT 1029
`
`
`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`James J. Fallon et al.
`In re Patent of:
`8,934,535 Attorney Docket No.: 45035-0002IP4
`U.S. Patent No.:
`January 13, 2015
`
`Issue Date:
`Appl. Serial No.: 4/033,245
`
`Filing Date:
`September 20, 2013
`
`Title:
`SYSTEMS AND METHODS FOR VIDEO AND AUDIO
`DATA STORAGE AND DISTRIBUTION
`
`
`Mail Stop Patent Board
`Patent Trial and Appeal Board
`U.S. Patent and Trademark Office
`P.O. Box 1450
`Alexandria, VA 22313-1450
`
`
`
`SUPPLEMENTAL DECLARATION OF
`PROFESSOR SCOTT T. ACTON, Ph.D.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Dated: ______March 21, 2018______
`
`
`
`
`
`
`
`
`By: _________________________
`
`
`Scott T. Acton
`
`Exhibit 1029
`
`
`
`X.
`
`UNPATENTABILITY OF THE ’535 PATENT (SUPPLEMENTAL)
`
`221. I, Scott T. Acton, of Charlottesville, VA, being over the age of 18 and
`
`competent to make the statements herein, hereby declare the following:
`
`222. I previously submitted a declaration in support of the Inter Partes
`
`Review petition Sling TV L.L.C., Sling Media L.L.C., DISH Network L.L.C., and
`
`DISH Technologies L.L.C. (collectively “DISH”) submitted before the Patent Trial
`
`and Appeal Board regarding U.S. Patent No. 8,934,535 (“the ’535 patent”). See
`
`generally DISH1003 (Acton Decl.)
`
`223. I incorporate my original declaration fully herein. I continue to hold
`
`the opinions I expressed in my original declaration, and I do not intend for this
`
`declaration to alter them. The opinions expressed in this supplemental declaration
`
`are relevant to this proceeding only in the situation where the Board adopts a
`
`construction for the term “access profile” that is “information regarding the
`
`number or frequency of reads or writes.”
`
`224. Similarly, for this supplemental declaration, DISH’s counsel asked me
`
`to consider the claim term “access profile” if construed as “information regarding
`
`the number or frequency of reads or writes.” Particularly, I was asked whether,
`
`under this construction, the prior art grounds I described in my original declaration
`
`Exhibit 1029 – pg. 1
`
`
`
`would still render the ’535 patent claims unpatentable. In my opinion, they would
`
`still render the ’535 patent claims unpatentable.
`
`221. I first reiterate my understanding that DISH proposes, for the purposes
`
`of this proceeding, that the BRI of the term “access profiles” is “information that
`
`enables a controller to determine a compression routine that is associated with a
`
`data type of the data to be compressed.” DISH1003-¶44. As I stated in my
`
`original declaration, the prior art grounds on which I opined satisfy the “access
`
`profiles” limitation under this construction. See DISH1003-¶¶96-117, 141-146,
`
`153, 173-174.
`
`222. I am further informed by DISH’s counsel that in the institution
`
`decision for this inter partes review, the Board held that claim construction of the
`
`term “access profile” was not necessary at that time to determine that institution
`
`was proper. I am further informed by DISH’s counsel that the Board determined
`
`that the ’535 patent teaches that “[a] bandwidth sensitive data compression routine
`
`may be selected based on access profiles that enable the controller to determine a
`
`compression routine associated with a data type of the data to be compressed.”
`
`The Board’s findings regarding the teachings of the ’535 patent are consistent with
`
`the claim construction for “access profiles” under the broadest reasonable
`
`interpretation that I applied in my original declaration: “information that enables a
`
`Exhibit 1029 – pg. 2
`
`
`
`controller to determine a compression routine that is associated with a data type of
`
`the data to be compressed.”
`
`223. Under the alternative interpretation of “access profiles” I have been
`
`asked to consider—“information regarding the number or frequency of reads or
`
`writes”—it is my opinion that claims of the ’535 patent are also rendered obvious.
`
`Particularly, and for the reasons I will describe below, it is my opinion that U.S.
`
`Patent No. 6,557,001 to Dvir et al. (“Dvir”) read in view of U.S. Patent No.
`
`5,675,789 to Ishii et al. (“Ishii”), which I identified as Ground 3 in my original
`
`declaration, renders obvious at least claims 3-6, 8, 11, and 12 of the ’535 patent.
`
`224. I understand that the “access profile” term occurs in limitations 1.1
`
`and 1.2 of the ’535 patent’s claim 1, from which claims 3-6, 8, 11, and 12 depend.
`
`Below I will describe how Dvir in view of Ishii (Ground 3) satisfies these
`
`limitations if “access profile” is construed as “information regarding the number or
`
`frequency of reads or writes.” I refer to and incorporate my original declaration for
`
`satisfaction of the remaining claim limitations.
`
`225. Claim Element 1.1: selecting an access profile from among a plurality
`
`of access profiles based upon the determined parameter or attribute.
`
`226. Ishii’s file compression portion 105 and file compression method
`
`selection portion 104 select an “access profile” from a plurality if access profile is
`
`construed as “information regarding the number or frequency of reads or writes.”
`
`Exhibit 1029 – pg. 3
`
`
`
`As I will describe, Ishii contains a high-access, a medium-access, and a low-access
`
`access profile. Based on a file’s access frequency, or the number of reads to the
`
`file, Ishii selects one of these access profiles.
`
`227. As I explained in my original declaration, “Ishii operates on data
`
`blocks consisting of files using, among other things, ‘a file compression portion
`
`which compresses a file on said file unit’ and ‘a compression method selector’ that
`
`selects compression methods based on ‘the data type and the access frequency of
`
`the file for compression.’” DISH1003-¶83 (Acton Decl.) (emphasis added)
`
`(quoting DISH1005 (Ishii) at Abstract; 5:51-6:17).
`
`228. Ishii also describes that its “file compression portion 105 actually
`
`compresses data of the searched file to be compressed” by “select[ing] an
`
`appropriate data compression method for compression of the file.” DISH1005-
`
`5:60-65. Ishii selects an appropriate data compression method “[a]ccording to the
`
`attributes of the file to be compressed including access frequency and file type.”
`
`Id. (emphasis added); see also, e.g., id. at 1:54-55.
`
`229. Ishii also describes how its file compression method selection portion
`
`104 selects an “access profile” if access profile is construed as “information
`
`regarding the number or frequency of reads or writes.” As I explained in my
`
`original declaration, Ishii “discloses a file compression method selection portion
`
`that ‘selects the [suitable] method . . . depending on the file access frequency and
`
`Exhibit 1029 – pg. 4
`
`
`
`data attribute’” and “Ishii’s compression method is performed by a system that
`
`uses a compression processor to determine a ‘data attribute’ of the received data to
`
`be compressed and determine ‘the access frequency . . . from the last access date
`
`and the number of accesses.’” DISH1003-¶199 (Acton Decl.) (quoting DISH1005
`
`(Ishii), 7:15-21 (emphasis added)).
`
`230. I also remarked in my original declaration on the different
`
`compression methods that the file compression method selection portion 104
`
`selects based on access frequency. Specifically, I noted that Ishii describes how its
`
`“file compression method selection portion 104 here selects the method with
`
`suitable compression ratio and compression/decompression speed depending on
`
`the file access frequency and data attribute.” DISH1003-¶165 (quoting
`
`DISH1005, 7:16-21). This selection is between different compression methods
`
`such that “[t]he file compression method with a shorter decompression time is
`
`selected for files with higher access frequency” and “the file compression method
`
`with a higher compression ratio is selected for files with lower access frequency.”
`
`DISH1003-¶165 (quoting DISH1005, 7:16-34 (emphasis added)); see also id.
`
`(quoting DISH1005, 1:32-45, 2:16-30).
`
`231. The “access frequency” considered by Ishii’s file compression portion
`
`105 and file compression method selection portion 104 corresponds to a data
`
`attribute of a file, namely the number of times the file has been accessed. Ishii
`
`Exhibit 1029 – pg. 5
`
`
`
`teaches that the “access frequency of a file is determined based on the last access
`
`date and the number of accesses for that file.” DISH1005-5:54-56 (emphasis
`
`added). I clarified in my original declaration that “number of accesses,” as cited in
`
`Ishii, is “synonymous” with “number of reads.” DISH1003-¶199. That is, when
`
`Ishii measures access frequency based on “number of access,” a POSITA would
`
`understand that to mean the number of times the system has read the file. Thus,
`
`the disclosure I have recited above corresponds to the selection of an “access
`
`profile” under the interpretation “information regarding the number or frequency
`
`of reads or writes.”
`
`232. Ishii explains that upon determining a file’s access frequency, it is
`
`classified among options of “high, medium, and low” access. Id. at 5:55-56
`
`(emphasis added). In other words, one access profile is selected from the plurality
`
`of high, medium, or low. Therefore, it is my opinion that Ishii teaches the
`
`limitation “selecting an access profile from among a plurality of access profiles
`
`based upon the determined parameter or attribute” even if “access profile” is
`
`construed as “information regarding the number or frequency of reads or writes.”
`
`233. As I described in my original declaration, a POSITA would have had
`
`reason to use these teachings of Ishii in conjunction with Dvir’s device. See
`
`DISH1003-¶¶161-172. One of the motivating reasons for combining Ishii’s
`
`teachings with Dvir’s device, I remarked, is to take advantage of Ishii’s insight of
`
`Exhibit 1029 – pg. 6
`
`
`
`selecting based upon access frequency. Specifically, “a POSITA would have
`
`understood the advantages that Ishii provides by reducing computing load . . . and
`
`would have been motivated to modify Dvir to incorporate Ishii’s access frequency-
`
`based compression selection to receive these benefits.” DISH1003-¶168 (emphasis
`
`added).
`
`234. Claim Element 1.2: compressing the at least the portion of the data
`
`block with one or more compressors using asymmetric data compression and
`
`information from the selected access profile to create one or more compressed data
`
`blocks, the information being indicative of the one or more compressors to apply to
`
`the at least the portion of the data block.
`
`235. Ishii teaches compressing at least a portion of the data block using
`
`information from the selected “access profile” if access profile is construed as
`
`“information regarding the number or frequency of reads or writes.” As I
`
`explained in my original declaration, “Ishii’s compression method is performed by
`
`a system that uses a compression processor to determine a ‘data attribute’ of the
`
`received data to be compressed and determine ‘the access frequency . . . from the
`
`last access date and the number of accesses.’” DISH1003-¶199 (Acton Decl.)
`
`(quoting DISH1005 (Ishii), 7:15-34 (emphasis added)). I also explained that
`
`“Ishii’s file compression method selection portion ‘selects the file compression
`
`method suitable for the data type and the access frequency and instructs the
`
`Exhibit 1029 – pg. 7
`
`
`
`compression means to execute processing by the selected compression method.’”
`
`DISH1003-¶200 (quoting DISH1005, 2:26-29); see also id. ¶165 (quoting
`
`DISH1005, 7:17-34, 6:7-17).
`
`236. Additional disclosure in Ishii reinforces its teaching of compressing
`
`based upon access frequency. For example, Ishii explains that “[i]t selects the data
`
`compression method suitable for the file considering . . . access frequency” and
`
`“[t]he file compression method selection portion 104 selects the method used by
`
`the data compression portion 105 to compress data of the file to be compressed.”
`
`DISH1005, 5:66-6:3; see also id. at 2:40-47).
`
`237. Ishii determines a compression method based on the access profile it
`
`has selected. As I explained in the previous section, Ishii uses the number of times
`
`a file has been read to classify it among either “high, medium and low” access-
`
`frequency selections. DISH1005, 5:55-56 (emphasis added). This classification is
`
`significant because in Ishii “select[ing] the method with suitable compression ratio
`
`and compression/decompression speed depend[s] on the file access frequency and
`
`data attribute.” Id. at 7:17-20. Using this information, “[t]he file compression
`
`method with a shorter decompression time is selected for files with higher access
`
`frequency and the file compression method with a higher compression ratio is
`
`selected for files with lower access frequency.” Id. at 7:28-31.
`
`Exhibit 1029 – pg. 8
`
`
`
`238. Ishii describes how its determination of a compression method would
`
`apply to specific compression methods that would be known to a POSITA. For
`
`example, it explains that “Lempel–Ziv coding has a lower compression ratio than
`
`Huffman and arithmetic coding but has a higher compression/decompression
`
`speed” and thus, “[t]he file compression method selection portion 104 selects
`
`Lempel–Ziv coding for files with high access frequency and Huffman or arithmetic
`
`coding for files with low access frequency.” DISH1005, 7:46-52. In other words,
`
`Ishii discloses selecting a compressor from several options based on access
`
`frequency.
`
`239. Considering this disclosure, it is my opinion that a POSITA would
`
`understand Ishii as teaching data compression according to an “access profile” if
`
`interpreted to mean “information regarding the number or frequency of reads or
`
`writes.”
`
`240. I further note that a POSITA would know how to take this insight
`
`from Ishii and apply it to other known compression methods with varying
`
`compression ratio and speed. This would include, for example, Dvir’s use of
`
`MPEG compression. See DISH1004 (Dvir), 5:13-19. To this point, MPEG is an
`
`asymmetric compression algorithm, as I explained in my original declaration. See
`
`DISH1003-¶¶111-114. The Lempel-Ziv compression algorithm disclosed in Ishii
`
`is also asymmetric, which the ’535 patent itself acknowledges. See DISH1001,
`
`Exhibit 1029 – pg. 9
`
`
`
`10:2-4. Substitution of one of these compression algorithms for another
`
`compression algorithm is well within the skill of a POSITA, who “would have
`
`been motivated to modify Dvir to incorporate Ishii’s access frequency-based
`
`compression selection to receive these benefits.” DISH1003-¶168 (Acton Decl.).
`
`Exhibit 1029 – pg. 10
`
`