Realtime Adaptive Streaming, LLC

Patent Trials and Appeals Board

Aug 17, 2020

IPR2019-00712 (P.T.A.B. Aug. 17, 2020)

Trials@uspto.gov Paper 30
571-272-7822 Entered: August 17, 2020

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

ADOBE, INC.,
Petitioner
v.

REALTIME ADAPTIVE STREAMING LLC,
Patent Owner.
___________

IPR2019-00712
Patent 8,929,442 B2
____________

Before, GEORGIANNA W. BRADEN, GREGG I. ANDERSON, and
KAMRAN JIVANI, Administrative Patent Judges.

ANDERSON, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining All Challenged Claims Unpatentable
35 U.S.C. § 318(a)

IPR2019-00712
Patent 8,929,442 B2

2
I. INTRODUCTION
In this inter partes review under 35 U.S.C. § 314, Adobe, Inc.
(“Petitioner”) filed a Petition (Paper 2, “Pet.”) pursuant to 35 U.S.C. §§ 311–
19 to institute an inter partes review of claims 1–15 (“challenged claims”) of
U.S. Patent 8,929,442 B2 (Ex. 1001, the “’442 patent”)1, filed on
December 19, 2013. Realtime Adaptive Streaming LLC (“Patent Owner”)
filed a Preliminary Response (Paper 6). We instituted an inter partes review
of all challenged claims (Paper 9, “Inst. Dec.”). After institution, Patent
Owner filed a Patent Owner Response (Paper 17, “PO Resp.”), Petitioner
filed a Reply (Paper 22, “Reply”), and Patent Owner filed a Sur-reply
(Paper 23, “Sur-reply”). An oral hearing was held on June 11, 2020, and a
transcript made of record (Paper 29, “Tr.”).
The Petition is supported by the Declaration of Dr. Clifford Reader
(Ex. 1003, “Reader Declaration”). The Reply is supported by the
Declaration of Dr. Clifford Reader in Support of the Petitioner’s Reply
(Ex. 1029, “Reader Reply Declaration”). The deposition of Dr. Reader was
taken by Patent Owner after the Reader Declaration (Ex. 2002). The
Response is supported by the Declaration of Kenneth A. Zeger, Ph.D. in
Support of Patent Owner’s Response (Ex. 2001, “Zeger Declaration”).
The deposition of Dr. Zeger was taken by Petitioner after the Zeger
Declaration (Ex. 1030).

1 Without conceding the date, Petitioner alleges that the ’442 patent’s
“earliest possible priority date” is the filing date of provisional application
No. 60/268,394 (“’394 application”) of February 13, 2001. Pet. 4 & n.1; see
also Ex. 1001, code (60), 1:15–16 (identifying the ’394 application as
related). The prior art references relied on were either filed or published
prior to this date. See Section I.E below. Patent Owner does not dispute that
the references relied on are prior art.
IPR2019-00712
Patent 8,929,442 B2

3
We have jurisdiction under 35 U.S.C. § 6. This Final Written
Decision is entered pursuant to 35 U.S.C. § 318(a) and 37 C.F.R. § 42.73.
For the reasons discussed below, Petitioner has shown by a preponderance
of the evidence that claims 1–15 of the ’442 patent are unpatentable.
II. BACKGROUND
A. Real Parties in Interest
The parties identify Adobe, Inc. and Realtime Adaptive Streaming,
LLC as the real parties in interest. Pet. 87; Paper 3, 1.
B. Related Matters
Petitioner advises us that the ’442 patent is the subject of the
following actions and administrative proceedings: Realtime Adaptive
Streaming v. Adobe Sys. Inc., Case No. 2-18-cv-09344 (C.D. Cal.); Realtime
Adaptive Streaming LLC v. Samsung Elecs. Co., Case No. 6-18-cv-00113
(E.D. Tex.); Realtime Adaptive Streaming LLC v. Adobe Sys. Inc., Case No.
1-18-cv-10355 (D. Mass.); Realtime Adaptive Streaming LLC v. Apple Inc.,
Case No. 1-17-cv-02869 (D. Colo.); Realtime Adaptive Streaming LLC v.
Sony Elecs. Inc., Case No. 1-17-cv-01693 (D. Del.); Realtime Adaptive
Streaming LLC v. Polycom Inc., Case No. 1-17-cv-02692 (D. Colo.);
Realtime Adaptive Streaming LLC v. Brightcove Inc., Case No. 1-17-cv-
01519 (D. Del.); Realtime Adaptive Streaming LLC v. Haivision Network
Video Inc., Case No. 1-17-cv-01520 (D. Del.); Realtime Adaptive Streaming
LLC v. Cisco Sys. Inc., Case No. 6-17-cv-00591 (E.D. Tex.); Realtime
Adaptive Streaming LLC v. Amazon.com Inc., Case No. 6-17-cv-00549 (E.D.
Tex.); and Sony Corp. v. Realtime Adaptive Streaming LLC, Case No.
IPR2018-01439 (PTAB).2 Pet. 87–88.

2 Dismissed. See ’1439 IPR, Paper 10.
IPR2019-00712
Patent 8,929,442 B2

4
Patent Owner advises us of the “following proceedings that may
affect, or be affected by,” a decision in this proceeding: Realtime Adaptive
Streaming LLC v. Haivision Network Video Inc., Case No. 1-17-cv-01520
(D. Del.); Realtime Adaptive Streaming LLC v. Adobe Systems Inc., Case
No. 1-18-cv-10355 (D. Mass); and Realtime Adaptive Streaming LLC v.
Arris Solutions, Inc., Case No. 19-cv-00585 (D. Colo.). Paper 3, 2.
C. Technology and the ’442 Patent
The ’442 patent describes “a system and method for compressing and
decompressing based on the actual or expected throughput (bandwidth) of a
system employing data compression and a technique of optimizing based
upon planned, expected, predicted, or actual usage.” Ex. 1001, 7:51–56.
1. Technology
The field of the invention is data compression and decompression. Id.
at 1:22–23. There are a variety of known data compression algorithms
where one or more parameters are changed, resulting in an associated
change in performance. Id. at 1:32–36. Lempel-Ziv is “a typical dictionary
based compression algorithm.” Id. at 1:36–38. The size of the dictionary
can affect performance resulting in a longer time to execute. Id. at 1:38–41.
“Algorithms that compress particularly well usually take longer to execute
whereas algorithms that execute quickly usually do not compress
particularly well.” Id. at 1:54–56. One of several problems identified is the
need to “provide dynamic modification of compression system parameters
so as to provide an optimal balance between execution speed of the
algorithm (compression rate) and the resulting compression ratio.” Id.
at 1:57–61.
IPR2019-00712
Patent 8,929,442 B2

5
2. ’442 Patent (Ex. 1001)
The ’442 patent selects an appropriate compression algorithm for a
particular application by balancing speed of compression against efficiency,
i.e., the compression ratio. Ex. 1001, 8:4–12. Higher compression ratios
result in “smaller compressed data sizes and greater storage capacity.”
Ex. 1003 ¶ 18. A problem for storage systems is that an optimal
compression ratio is often too slow and hinder data read/write operations.
Id. (citing Ex. 1001, 5:11–55).
The claimed method and apparatus of the ’442 patent selects
compression routines using a “controller [that] tracks and monitors the
throughput (data storage and retrieval) of a data compression system and
generates control signals to enable/disable different compression algorithms
when, e.g., a bottleneck occurs so as to increase the throughput and
eliminate the bottleneck.” Ex. 1001, 9:55–59. The ’442 patent equates
throughput to bandwidth. Id. at code (57) (“compressing and decompressing
data based on an actual or expected throughput (bandwidth) of a system”);
see also id. at 1:51–56, 9:12–14.
The ’442 patent explains that:
two categories of compression algorithms are defined—an
“asymmetrical” data compression algorithm and a
“symmetrical[”] data compression algorithm[s]. An
asymmetrical data compression algorithm is referred to herein as
one in which the execution time for the compression and
decompression routines differ significantly. In particular, with
an asymmetrical algorithm, either the compression routine is
slow and the decompression routine is fast or the compression
routine is fast and the decompression routine is slow. Examples
of asymmetrical compression algorithms include dictionary-
based compression schemes such as Lempel-Ziv.

IPR2019-00712
Patent 8,929,442 B2

6
Ex. 1001, 9:61–10:4.
The ’442 patent then describes “symmetry” and “asymmetry” in the
context of compression and decompression.
[S]ymmetry may be defined in terms of overall effective
bandwidth, compression ratio, or time or any combination
thereof. In particular, in instances of frequent data read/writes,
bandwidth is the optimal parameter for symmetry. In
asymmetric applications such as operating systems and
programs, the governing factor is net decompression bandwidth,
which is a function of both compression speed, which governs
data retrieval time, and decompression speed, wherein the total
governs the net effective data read bandwidth.

Id. at 10:16–24.
Figure 1 of the ’442 patent is reproduced below.

Figure 1 is “a high-level block diagram [that] illustrates a system for
providing bandwidth sensitive data compression/decompression according to
an embodiment of the present invention.” Ex. 1001, 10:31–34. Figure 1
illustrates “a host system 10 comprising a controller 11 (e.g., a file
management system), a compression/decompression (or data compression)
system 12, a plurality of compression algorithms 13, a storage medium 14,
and a plurality of data profiles 15.” Id. at 10:34–38. The “controller tracks
IPR2019-00712
Patent 8,929,442 B2

7
and monitors the throughput (e.g., data storage and retrieval)” of the system
and when the throughput of the system falls below a predetermined
threshold, the system generates control signals to enable/disable different
compression algorithms. Id. at 10:42–45. “In one embodiment, the system
throughput that is tracked by the controller 11 preferably comprises a
number of pending access requests to the memory system.” Id. at 10:43–45.
Still referring to Figure 1, “[t]he data compression system 12 is
operatively connected to the storage medium 14 using suitable protocols to
write and read compressed data to and from the storage medium 14.”
Ex. 1001, 10:46–48. “The data compression system 12 may maintain the
compressed data to be stored on the storage medium 14 and the
decompressed data that is retrieved from the storage medium 14 for
subsequent data processing, storage, or transmittal.” Id. at 10:64–11:1. Data
compression system 12 may receive compressed or uncompressed data via
I/O (input/output) port 16 from a remote location or transmit the data to
another network device for remote processing or storage. Id. at 11:1–8.
D. Illustrative Claim
Independent claims 1 and 8 of the challenged claims are independent
claims to a method and apparatus, respectively. Claims 2 through 7 depend
directly or indirectly from claim 1. Claims 9 through 15 depend directly or
indirectly from claim 8.
Claim 1 is reproduced below as illustrative.
1. [1.0]3 A method comprising:

3 We adopt Petitioner’s format, sometimes omitting the brackets, for
labeling the claim limitations. Petitioner identifies the claim number
followed by numbers for each limitation after the preamble, which is
numbered limitation 1.0. See Pet. 15. Patent Owner follows the same
IPR2019-00712
Patent 8,929,442 B2

8

[1.1] decompressing at least a portion of a compressed data
block that was selected from among a plurality of compressed
data blocks,

[1.2] wherein at least a portion of a data block having video or
audio data was compressed with one or more compression
algorithms selected from among a plurality of compression
algorithms based upon a throughput of a communication
channel and a parameter or an attribute of the at least the
portion of the data block to create one or more of the plurality
of compressed data blocks; and

[1.3] storing at least a portion of the decompressed data block.

Ex. 1001, 20:6–17.
E. Evidence
Tso, US 6,185,625 B1, filed December 20, 1996, issued February 6,
2001 (Ex. 1004).
Vishwanath, US 6,216,157 B1, filed June 17, 1998, issued April 10,
2001 (Ex. 1005).
Kidder, US 5,898,833, filed December 31, 1996, issued April 27,
1999 (Ex. 1006).
Petitioner further relies on the Reader Declaration (Ex. 1003) and the
Reader Sur-reply Declaration (Ex. 1029). See Section I above. Patent
Owner also relies on the Zeger Declaration (Ex. 2001). Id.
F. Asserted Challenges to Patentability
In the Institution Decision, we instituted trial on the following
challenges:

format, excluding the brackets. PO Resp. 13 (limitation 1.2). The inclusion
of Petitioner’s formatting and/or brackets does not impact our analysis.
IPR2019-00712
Patent 8,929,442 B2

9
Claims Challenged 35 U.S.C. §4 Reference(s)/Basis
1–15 102 Tso
1–15 103 Tso
1–15 103 Vishwanath, Kidder
Inst. Dec. 25–44.
III. ANALYSIS
A. Level of Ordinary Skill in the Art
Petitioner alleges a person of ordinary skill in the art at the time of the
invention would have
had a bachelor’s degree in electrical engineering, computer
engineering, computer science, or the equivalent and 2–3 years
of work experience with real-time systems implementing data
compression, storage retrieval, processing, and transmission, or
the equivalent.

Pet. 7 (citing Ex. 1003 ¶¶ 97–99). Patent Owner adopts this definition for
purposes of this proceeding. PO Resp. 6–7 (citing Ex. 2001 ¶¶ 26–28). We
adopted this definition in the Institution Decision and maintain it for this
Decision. See Inst. Dec. 27.
B. Claim Construction
In an inter partes review requested in a petition filed on or after
November 13, 2018, we apply the same claim construction standard used in
district courts in a civil action under 35 U.S.C. § 282(b), namely that
articulated in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en
banc). See Changes to the Claim Construction Standard for Interpreting
Claims in Trial Proceedings Before the Patent Trial and Appeal Board, 83

4 The Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29, 125
Stat. 284, 287–88 (2011), amended 35 U.S.C. § 103. Because the effective
filing date of the ’442 patent is before the effective date of the relevant
amendment, the pre-AIA version of § 103 applies.
IPR2019-00712
Patent 8,929,442 B2

10
Fed. Reg. 51,340, 51,343 (Oct. 11, 2018) (amending 37 C.F.R. § 42.100(b)
effective November 13, 2018) (now codified at 37 C.F.R. § 42.100(b)
(2019)). In this case, the Petition was filed on February 27, 2019, and we,
therefore, apply the district court standard here. In so doing, we construe a
claim “in accordance with the ordinary and customary meaning of such
claim as understood by one of ordinary skill in the art and the prosecution
history pertaining to the patent.” 37 C.F.R. § 42.100(b). “[O]nly those
terms need be construed that are in controversy, and only to the extent
necessary to resolve the controversy.” See Nidec Motor Corp. v. Zhongshan
Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017); see also
Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir.
1999).
Both parties propose constructions for “asymmetric”/“compression
algorithms is asymmetric.” Reply 1; PO Resp. 7. The parties have separate
proposals for construing “throughput.” Reply 2; Sur-reply 3–4.
1. “asymmetric”/“compression algorithms is asymmetric”
We preliminarily construed “asymmetric”/“compression algorithms is
asymmetric” to mean “a compression algorithm . . . in which the execution
time for the compression and decompression routines differ significantly.”
Inst. Dec. 15–16. The construction is not in dispute because both parties
agree to the construction from the Institution Decision. Reply 1–2; PO
Resp. 7. For completeness of the record, we adopt the agreed upon
construction for purposes of this Decision.
2. “throughput”
During the course of the trial, a dispute arose between the parties
regarding construction of the term throughput. Petitioner proposes that
IPR2019-00712
Patent 8,929,442 B2

11
“throughput of a communication channel” should be construed to mean “a
rate of transmission of data.” Reply 8–9 (citing Ex. 1029 ¶¶ 19–27). Patent
Owner’s proposal for “throughput” is “the amount of data flowing through a
data channel.” Sur-reply 3–5 (citing Ex. 2001 ¶ 2001).
The claim language at issue is “one or more compression algorithms
selected from among a plurality of compression algorithms based upon a
throughput of a communication channel.” The claims require “throughput”
is “of a communication channel,” which does not favor either proposal. The
claim language of the ’442 patent does not suggest any specific
measurement of “throughput”, e.g., rate or amount, and does not support
either parties’ proposed construction.
As Petitioner notes, the Specification “recites various types of
throughputs such as: ‘actual throughput,’ ‘expected throughput,’ ‘desired
throughput,’ ‘system throughput,’ ‘overall throughput,’ ‘optimum
throughput,’ and ‘required/requested data throughput’ that allow
‘optimiz[ation] based upon planned, expected, predicted, or actual’
throughput.” Reply 5 (citing Ex. 1001, code (57), 1:22–30, 7:52–56, 9:11–
28, 10:43–45, 11:28–2, 12:25–27, 13:47–61). Patent Owner argues the
throughput types refer to “the amount of data flowing through a data
channel,” supporting its proposed construction. Sur-reply 4–5 (citing
Ex. 2001 ¶ 79).
Patent Owner argues the Specification uses “the term ‘throughput’ as
a synonym for ‘usage’ of a communication channel or the data flowing
through a channel.” Sur-reply 7. In support of its proposed construction,
Patent Owner argues that Specification describes “throughput” as including
the “number of pending transmission requests over the communication
IPR2019-00712
Patent 8,929,442 B2

12
channel,” the “number of pending access requests to a storage device,” and
“the number of pending disk accesses (access requests) to determine whether
a bottleneck is occurring.” Id. at 8 (citing Ex. 1001, 8:25–27, 8:18–20,
13:35–39) (emphasis added). Patent Owner does not explain why the
“number” of pending actions related to transmission require us to conclude
that “throughput” should be construed as limited to “the amount of data
flowing through a data channel.”
The prosecution history of the application for the ’442 patent is of
record here. See Ex. 1002. Neither party points to any evidence from the
prosecution history as supportive of their respective proposals.
The understanding of what “bandwidth” means is relevant to the
construction of “throughput” because “bandwidth” is related to
“throughput.” Indeed, as Petitioner alleges, the ’442 patent equates
“throughput” and “bandwidth.” Reply 7–8 (citing Ex. 1001, 7:57–8:3
(“system and method for compressing and decompressing based on the
actual or expected throughput (bandwidth) of a system employing data
compression”); see also id. at 8 (citing Zeger Deposition, 45:2–9 (“at least in
the abstract, where they have the ‘bandwidth’ as a parenthetical, that is being
used synonymously with ‘throughput.’”). With reference to the
Specification’s disclosure of “throughput (bandwidth),” Patent Owner
argues that “the term ‘bandwidth’ can inform what the appropriate
construction and understanding of the term ‘throughput’ would be.”
Tr. 50:2–5. We determine that, as used in the ’442 patent, “throughput”
equates to and means the same as “bandwidth.” See Ex. 1001, Abs., 7:57–
8:3, 9:11–14.
IPR2019-00712
Patent 8,929,442 B2

13
Patent Owner cites extrinsic evidence in the form of the Zeger
Declaration and argues we should adopt one of three understandings5 a
person of ordinary skill would have had of the term “bandwidth.” Ex. 2001
¶¶ 78–79. The understanding of “bandwidth” asserted by Dr. Zeger, and
relied on by Patent Owner for its proposed construction, is “the amount of
data flowing through a data channel.” PO Resp. 28–29 (citing Ex. 2001
¶ 78). However, the second listed understanding of “bandwidth” is capacity
“measured in bits per second,” i.e., the “rate of transmission of data” and is
the construction proposed by Petitioner. Neither the Zeger Declaration,
Patent Owner Response, nor the Sur-reply explain why the other two
understandings of “bandwidth” are inapplicable here. Rather, Patent Owner
summarily argues the third understanding of “bandwidth” (“the amount of
data flowing through a data channel”) is the “correct interpretation” and the
second understanding (capacity “measured in bits per second”) is not.
Tr. 48:17–50:16.
The United States Court of Appeals for the Federal Circuit requires
that declarant testimony “cannot be used as a wholesale substitute for
reasoned analysis and evidentiary support.” Arendi v. Apple, 832 F.3d 1355,
1361–1362 (Fed. Cir. 2016). Our Trial Practice Guide cautions, “in an

5 “(1) the difference between the largest and smallest frequencies in a band
of spectrum, i.e., the width of the frequency band (for example, ‘AM radio
stations use 10 KHz bandwidths.’); and (2) the capacity of a
communications channel, measured in bits per second, i.e., the maximum
about of data that could potentially be transmitted across the channel. For
example: ‘Some fiber optic internet connections have bandwidths up to 10
Gigabits per second.’ And (3) bandwidth may also be used to refer to the
amount of data flowing through a data channel, which is upper bounded by
the maximum capacity of the channel.” Ex. 2001 ¶ 78; PO Resp. 28–29.
IPR2019-00712
Patent 8,929,442 B2

14
obviousness analysis, conclusory assertions from a third party about general
knowledge in the art cannot, without supporting evidence of record, supply a
limitation that is not evidently and indisputably within the common
knowledge of those skilled in the art.” Trial Practice Guide,6 36 (citing K/S
Himpp v. Hear-Wear Techs., LLC, 751 F.3d 1362, 1365 (Fed. Cir. 2014)).
Patent Owner’s unsupported argument is not persuasive and we determine
that “bandwidth” includes “rate of transmission of data.”
Patent Owner argues other alleged “closely-related patents” support
Patent Owner’s proposed construction for the claims here. Sur-reply 9–10.
Patent Owner’s purported support for its construction is based on different
claims of allegedly related patents challenged in other inter partes review
proceedings.7 Id. For example, Patent Owner points to claims from U.S.
Patent No. 7,386,046 (“’046 patent”). Id. at 9. The relationship to the ’442
patent is not explained. The ’046 patent claim term is “tracking the
throughput.” Id. The claims here do not recite “tracking.”
Patent Owner argues our construction of these claims in the ’1222
IPR, which also challenges claims of the ’442 patent, is “incompatible” with
Petitioner’s proposed construction. Sur-reply 10. The construction in the
’1222 IPR is preliminary. The preliminary construction is further qualified
in specifying that “throughput” includes “at least the number of pending
transmission requests.” Id. (citing ’1222 IPR, Paper 7, 13–14) (emphasis

6 Patent Trial and Appeal Board Consolidated Trial Practice Guide (Nov.
2019) (“Trial Practice Guide”), available at
https://www.uspto.gov/sites/default/files/documents/tpgnov.pdf
7 Netflix, Inc. v. Realtime Adaptive Streaming, LLC, IPR2019-00209; Google
LLC v. Realtime Adaptive Streaming, LLC, IPR2019-01033; ARRIS
Solutions, Inc. v. Realtime Adaptive Streaming, LLC, IPR2019-01222
(“’1222 IPR”).
IPR2019-00712
Patent 8,929,442 B2

15
added). Neither proposed construction here is precluded by the preliminary
finding in the ’1222 IPR.
As discussed above (see n.5), one understanding of “bandwidth”
asserted by Patent Owner is
(2) the capacity of a communications channel, measured in bits
per second, i.e., the maximum about of data that could potentially
be transmitted across the channel. For example: “Some fiber
optic internet connections have bandwidths up to 10 Gigabits per
second.”

PO Resp. 28 (citing Ex. 2001 ¶ 78) (emphasis added). This understanding
comports with Petitioner’s proposed construction of “throughput.”
Petitioner cites extrinsic dictionary evidence defining “throughput” as
additional support for its construction. See Reply 4 (citing Ex. 1027,8 1179
(“data transmission” defined as: “[t]he total capability of equipment to
process or transmit data during a specified time period.”)); Ex. 1028,9 519
(“throughput” defined as: “[t]he data transfer rate of a network, measured as
the number of bits per second transmitted” or “[a] measure of the data
processing rate in a computer system.”). Thus, we construe “throughput” to
include, but not limited to, “a rate of transmission of data.” It is not
necessary to determine whether Patent Owner’s proposed construction is
also included in the construction of “throughput” because no issue here turns
on whether “throughput” also includes “the amount of data flowing through
a data channel.” See Section III.F.3.c below (Patent Owner arguing “rate of
transmission of data” is not “throughput”).

8 IEEE, THE AUTHORITATIVE DICTIONARY OF IEEE STANDARDS TERMS,
(7TH Ed. 2000).
9 Microsoft Corporation, MICROSOFT COMPUTER DICTIONARY (5th Ed.
2002).
IPR2019-00712
Patent 8,929,442 B2

16
C. Legal Standard for Anticipation
Anticipation “requires that every element and limitation of the claim
was previously described in a single prior art reference, either expressly or
inherently, so as to place a person of ordinary skill in possession of the
invention.” Sanofi-Synthelabo v. Apotex, Inc., 550 F.3d 1075, 1082 (Fed.
Cir. 2008) (citing Schering Corp. v. Geneva Pharms., Inc., 339 F.3d 1373,
1379 (Fed. Cir. 2003); Continental Can Co. v. Monsanto Co., 948
F.2d 1264, 1267–69 (Fed. Cir. 1991)).
“[U]nless a reference discloses within the four corners of the
document not only all of the limitations claimed but also all of the
limitations arranged or combined in the same way as recited in the claim, it
cannot be said to prove prior invention of the thing claimed and, thus, cannot
anticipate under 35 U.S.C. § 102.” Net MoneyIN, Inc. v. VeriSign, Inc., 545
F.3d 1359, 1371 (Fed. Cir. 2008). Whether a claim is invalid as anticipated
is a two-step inquiry. See Power Mosfet Tech., LLC v. Siemens AG, 378
F.3d 1396, 1406 (Fed. Cir. 2004). The first step requires construction of the
claim. Id. The second step in the analysis requires a comparison of the
properly construed claim to the prior art. Id.
D. Legal Standard for Obviousness
A patent claim is invalid as obvious if the differences between the
claimed subject matter and the prior art are “such that the subject matter as a
whole would have been obvious at the time the invention was made to a
person having ordinary skill in the art to which said subject matter pertains.”
See 35 U.S.C. § 103(a).
“The ultimate determination of obviousness is a question of law, but
that determination is based on underlying factual findings.” In re Nuvasive,
IPR2019-00712
Patent 8,929,442 B2

17
Inc., 842 F.3d 1376, 1381 (Fed. Cir. 2016). The underlying factual findings
include: (1) “the scope and content of the prior art”; (2) “differences
between the prior art and the claims at issue”; (3) “the level of ordinary skill
in the pertinent art”; and (4) “the presence of secondary considerations of
nonobviousness such” “as commercial success, long felt but unsolved needs,
failure of others,” and unexpected results. Id. (citing, inter alia, Graham v.
John Deere Co., 383 U.S. 1, 17–18 (1966)).
“To satisfy its burden of proving obviousness, a petitioner cannot
employ mere conclusory statements. The petitioner must instead articulate
specific reasoning, based on evidence of record, to support the legal
conclusion of obviousness.” In re Magnum Oil Tools Int’l, Ltd., 829 F.3d
1364, 1380 (Fed. Cir. 2016). Furthermore, in assessing the prior art, the
Board must consider whether a person of ordinary skill would have been
motivated to combine the prior art to achieve the claimed invention.
Nuvasive, 842 F.3d at 1381. As the Federal Circuit has noted, in quoting
from the Supreme Court’s decision in KSR Int’l Co. v. Teleflex Inc., 550
U.S. 398, 418–419 (2007), “‘because inventions in most, if not all, instances
rely upon building blocks long since uncovered, and claimed discoveries
almost of necessity will be combinations of what, in some sense, is already
known,’ ‘it can be important to identify a reason that would have prompted a
person of ordinary skill in the relevant field to combine the elements in the
way the claimed new invention does.’” Personal Web Techs., LLC v. Apple,
Inc., 848 F.3d 987, 991–992 (Fed. Cir. 2017).
IPR2019-00712
Patent 8,929,442 B2

18
E. Alleged Anticipation/Obviousness of Claims 1–15 Over Tso
Petitioner alleges claims 1–15 would have been “anticipated and/or
rendered obvious by Tso.” Pet. 4, 15–53 (emphasis omitted). The Petition
is supported by the Reader Declaration. Ex. 1003 ¶¶ 126–238.
1. Tso (Ex. 1004)
Tso describes “[a] system for enhancing data access over a
communications link.” Ex. 1004, code (57). The system retrieves an object
over a computer network having a remote scaling server, a network client, a
remote proxy, and an encode service provider. Id.
An encode service provider selectively encodes or scales, i.e.,
compresses, an object based on a predetermined characteristic thereof, such
as a datatype. Ex. 1004, 2:60–62.10 Greater compression rates are achieved
“without negatively affecting software that ultimately processes the
compressed data.” Id. at 2:63–66. Higher compression rates may be
achieved by “reducing the quality of a picture and/or by using compression
techniques tuned for specific datatypes.” Id. at 2:66–3:2. “For example,
representing certain types of images, such as photographic images, using
JPEG or fractal compression algorithms may result in a 10x improvement in
compression rate over the popular GIF format.” Id. at 3:2–6. The system
can “actively control the trade-off between download time and content
quality by selecting whether or not to scale content and/or the extent of such
scaling.” Id. at 11:44–48.

10 Both experts agree that “scaled” means “compressed.” Ex. 2001 ¶ 41;
Ex. 1003 ¶ 114; see also Ex. 1004, 6:33–36 (The service provider has the
“capability to compress and/or scale” various selection criteria.).
IPR2019-00712
Patent 8,929,442 B2

19
Figure 5 of Tso is reproduced below.

Figure 5 of Tso is “a schematic diagram illustrating an embodiment of the
present invention directed to an enabled network client.” Ex. 1004, 3:29–31.
As illustrated in Figure 5, browser 12 renders data streams having non-
standard MIME (Multipurpose Internet Mail Extensions) types as they are
received by network client 3. Id. at 9:16–21, 13:41–44. Network client 3 is
in communication with remote scaling server 1 and receives requests for
refinement “to obtain higher quality representations of said images.” Id.
at 13:44–50. Network client 3 may contain specialized decoding software to
support more sophisticated scaling features. Id. at 13:51–54.
Still referencing Figure 5, “encode manager 7 is coupled to HTTP
remote proxy 6.” Ex. 1004, 6:27–28. The encode manager “controls one or
more encode service providers 8 which may be used to selectively scale
content based on a predetermined characteristic thereof.” Id. at 6:30–32.
The service provider has the “capability to compress and/or scale” various
selection criteria such as image, video, or HTML (HyperText Markup
Language). Id. at 6:33–36. One example of a “predetermined selection
IPR2019-00712
Patent 8,929,442 B2

20
criteria” includes “speed and/or quality of client/server communications
link 4.” Id. at 6:36–44.
Link speed may be determined in a variety of ways, including “by
sampling the rate of one or more prior transfers of data between itself and
network client 3.” Ex. 1004, 10:21–24. “If the derived link speed indicates
a relatively slow communications link, HTTP remote proxy 6 may determine
that all content to be transmitted to network client 3 should be scaled by
default.” Id. at 10:33–36. A relatively fast communications link may not
justify using scaling. Id. at 10:36–39. In either case, “HTTP remote proxy 6
records this determination in client preference table 11,” which is updated to
store the link speed. Id. at 10:24–27, 10:40–44.
2. Claim 1
The preamble of the ’442 patent, designated limitation 1.0, recites “[a]
method comprising.” Petitioner argues Tso discloses a “method for
providing a client with an object to be rendered to a user.” Pet. 15 (citing
Ex. 1004, code (57), Figs. 8–10, 21:63–22:16 (claim 21); Ex. 1003 ¶ 127).
Neither party suggests, nor do we find, that the preamble is limiting. Patent
Owner does not dispute that the preamble is disclosed or taught by Tso. We
find Petitioner has shown persuasively that Tso discloses or teaches a
method as per the preamble of claim 1.
a. Limitation 1.1
Limitation 1.1 recites “decompressing at least a portion of a
compressed data block that was selected from among a plurality of
compressed data blocks.” Petitioner’s annotation of Tso’s Figure 5 from
page 16 of the Petition is reproduced below.
IPR2019-00712
Patent 8,929,442 B2

21
Petitioner’s Annotation of Tso Fig. 5

Figure 5 is a schematic diagram illustrating an embodiment of the invention
of Tso directed to an enabled network client. Ex. 1004, 3:29–31. As shown
in the annotation to Figure 5, Tso’s system “includes a ‘network client 3
[that] accesses Internet 2 through a remote scaling server 1’ implemented as
a network server.” Ex. 1003 ¶ 128; see also Pet. 15–16 (citing Ex. 1004,
6:3–12). Petitioner argues, in part, that “Tso’s server uses a compression
algorithm to create compressed data blocks.” Pet. 17 (citing Ex. 1004, Fig. 6
(annotated)), 22 (citing Ex. 1003 ¶ 153). Furthermore, Petitioner cites to
Tso’s teaching that the compressed data is received as a “block of data” and
that, in addition, blocks of data would have been obvious to a person of
ordinary skill. Pet. 16–17 (citing Ex. 1004, 11:6–16; Ex. 1003 ¶¶ 128–132).
Patent Owner does not dispute Petitioner’s argument and supporting
evidence that Tso discloses or teaches limitation 1.1. Given the entirety of
the record, we find Petitioner has shown persuasively that Tso discloses or
teaches a server and client encoding and decoding compressed data
transmitted in data blocks, as recited in limitation 1.1.
IPR2019-00712
Patent 8,929,442 B2

22
b. Limitation 1.2-Undisputed Portions
Limitation 1.2 recites:
wherein at least a portion of a data block having video or audio
data was compressed with one or more compression algorithms
selected from among a plurality of compression algorithms based
upon a throughput of a communication channel and a parameter
or an attribute of the at least the portion of the data block to create
one or more of the plurality of compressed data blocks.

Petitioner alleges Tso anticipates and/or renders obvious claim 1 of the ’442
patent, including limitation 1.2 reproduced above. Pet. 22. Patent Owner
disputes only whether Tso discloses that portion of limitation 1.2 reciting
“one or more compression algorithms selected from among a plurality of
compression algorithms based upon a throughput of a communication
channel.” PO Resp. 13. For the reasons discussed below, we find that Tso,
as understood by a person of ordinary skill in the art, discloses limitation
1.2, including the disputed portion. See Pet. 22–29.
Figure 7 of Tso is reproduced below.

Tso’s Figure 7 is a schematic diagram of one embodiment of “an encode
service provider configured to scale image data.” Ex. 1004, 3:36–38.
IPR2019-00712
Patent 8,929,442 B2

23
Petitioner argues Tso’s remote scaling server shown in Figure 7 “chang[es]
content received from Internet 2 prior to returning it to a requesting network
client 3” by compressing the data prior to transmission. Pet. 22 (citing
Ex. 1004, 10:45–11:17, 6:13–45); see also id. at 23 (Petitioner’s annotation
of Tso Figure 7).
Relying on the Reader Declaration, Petitioner argues the “server uses
a compression algorithm to create compressed data blocks.” Pet. 22 (citing
Ex. 1003 ¶ 153). Petitioner’s showing regarding the “compressed data
blocks” recited in limitation 1.1 is incorporated as support for this
limitation 1.2. Id.; see also id. at 15–21 (showing for limitation 1.1).
Tso teaches the encode service provider selectively scales content
based on “predetermined characteristics” or selection criteria.
Ex. 1004, 6:30–36. Selection criteria may include “different type of data
content, such as image, video, or HTML (HyperText Markup Language).”
Id. at 6:32–36. In addition to data type, selection criteria for selectively
scaling include presentation capabilities of network client 3 and “speed
and/or quality of client/server communications link 4.” Id. at 6:36–43.
Petitioner cites to the preceding disclosures of Tso to argue a person of
ordinary skill would have understood Tso’s objects are data blocks and
compressed objects are compressed data blocks. Id. at 23–24 (citing Ex.
1003 ¶¶ 153–155; Ex. 1004, 6:27–45, 4:17–50, 11:6–17).
Petitioner argues that Tso’s different encode service providers “may
include the capability to compress and/or scale a different type of data type
(i.e., a plurality of compression algorithms).” Pet. 24 (citing Ex. 1003
¶ 156). Petitioner alleges data types associated with encode service
providers include “image, video, or HTML (HyperText Markup Language).”
IPR2019-00712
Patent 8,929,442 B2

24
Id. at 24 (citing Ex. 1004, 6:27–45, 7:3–17, quoting Ex. 1004, 6:33–36).
Petitioner then argues a person of ordinary skill “would have understood that
Tso uses a video encoder for compressing video data, an HTML encoder for
compressing HTML data, etc.” Id. (citing Ex. 1003 ¶ 157).
With the above as a foundation, Petitioner argues that the portion of
limitation 1.2 reciting selection of a compression algorithm “based upon . . .
a parameter or an attribute of the at least the portion of the data block” is
disclosed or taught by Tso’s server which “compresses requested objects
‘based on a predetermined characteristic,’ such as a ‘type of data content.’”
Pet. 28 (citing Ex. 1004, 6:27–36). Petitioner also relies on the encode
manager of Tso which “‘dynamically determines the content type for the
data stream (e.g., image/jpeg, image/gif, video/mpeg)’ and matches the
content type to a particular encode service provider.” Id. (citing Ex. 1004,
10:60–11:4); see also id. (citing Ex. 1004, 10:60–11:4; Ex. 1003 ¶ 164).
Patent Owner does not dispute Petitioner’s argument and supporting
evidence regarding the above elements of limitation 1.2. Based on the
entirety of the record, we agree with Petitioner’s showing summarized above
and find Petitioner has shown persuasively the undisputed portions of
limitation 1.2 are disclosed in Tso.
c. Limitation 1.2-Disputed Portion
The central issue is whether Tso teaches simply determining whether
or not to compress or, instead, discloses or teaches selecting a compression
algorithm, i.e., “which compression algorithm should be applied.” PO Resp.
14. Stated in the context of limitation 1.2, what is in dispute is “one or more
compression algorithms selected from among a plurality of compression
algorithms based upon a throughput of a communication channel.” See PO
IPR2019-00712
Patent 8,929,442 B2

25
Resp. 14–22; Reply 13–22; Sur-reply 13–18. That portion of limitation 1.2
reciting the selection of a compression algorithm based on “a parameter or
an attribute of the at least the portion of the data block” is not disputed. See
Section III.E.2.a above.
Preliminarily, Petitioner argues that as long as the selection is based
on both “bandwidth” and a “parameter” that is sufficient and how the initial
determination is made is not required. See Reply 13–14 (citing Ex. 1029
¶¶ 33–34; Ex. 1030, 19:25–21:2, 10:10–30). In other words, Petitioner
argues a broad interpretation of “based upon a throughput” requires only “a
determination of whether to compress.” Tr. 8:8–11. We agree with Patent
Owner that Petitioner’s broad interpretation reads the phrase “selected from
among a plurality of compression algorithms” out of limitation 1.2. See Sur-
reply 15. Based on the entirety of the record, Petitioner does not argue
persuasively how its broad construction proposal addresses the “selected
from” language. The claim clearly and unambiguously requires the selection
of a compression algorithm. See Phillips, 415 F.3d at 1314 (claim
construction unnecessary when “the ordinary meaning of claim language as
understood by a person of skill in the art [is] readily apparent.”).
Accordingly, we do not adopt Petitioner’s proposed construction.
We turn to whether Tso discloses or teaches compressing video or
audio data “with one or more compression algorithms selected from among a
plurality of compression algorithms based upon a throughput of a
communications channel.” Petitioner argues Tso’s scaling server “selects a
compression algorithm from among a plurality by selecting a particular
encode service provider.” Pet. 22, 25 (citing Ex. 1003 ¶ 158). Petitioner
argues that “Tso’s encoders are tailored to various criteria, including
IPR2019-00712
Patent 8,929,442 B2

26
‘available network bandwidth,’ which is a characteristic of the client/server
communications link.” Id. at 26 (citing Ex. 1004, 3:6–14). Petitioner then
alleges “Tso ‘selectively encod[es] [a requested] object based on one or
more predetermined selection criteria,’ [citing Ex. 1004, claim 13] where the
‘predetermined selection criteria comprise a characteristic of the
communications link’ over which the requested object is transmitted. [citing
Ex. 1004, claim 16].” Id.
Petitioner points to the encode manager of Tso as managing the
compression of data to be transmitted from the server to network client.
Pet. 26 (citing Ex. 1004, 6:27–45). Petitioner argues the encode manager
also controls “one or more encode service providers 8 . . . to selectively scale
content based on a predetermined characteristic,” such as the “speed and/or
quality of client/server communications link 4.” Id. (citing Ex. 1004, 6:27–
45). Petitioner concludes that the recited selection of a compression
algorithm is disclosed or rendered obvious by Tso’s encode manager’s
selection of an encode service provider. Id. (citing Ex. 1003
¶ 160). Furthermore, according to Petitioner, the selection of a compression
algorithm is based on bandwidth, which is disclosed in Tso as “link speed.”
Id. (citing Ex. 1004, 1:29–67; Ex. 1003 ¶ 160).
Petitioner supports its argument and evidence for the disputed
limitation by citing to additional disclosures from Tso that disclose or teach
automatic updates of the link speed/quality information over the
communications link “to adapt to changing conditions, such as increased
congestion on client/server communications link 4.” Pet. 26–27 (citing
Ex. 1004, 10:11–44; see also id. at 27 (citing Ex. 1004, 12:1–19 (separate
embodiment describing “whether and how content is to be scaled”)).
IPR2019-00712
Patent 8,929,442 B2

27
Petitioner also cites Tso’s disclosure or teaching of load balancing
procedures which “result in changes to the throughput of the client/server
communications link.” Id. at 27 (citing Ex. 1004, 13:15–22; Ex. 1003
¶ 161). Petitioner concludes by relying on the Reader Declaration that a
person of ordinary skill “would have understood that Tso uses the
throughput/bandwidth of the client/server communications link—over which
the compressed data is received—to select an appropriate compression
algorithm to perform network load-balancing.” Id. (citing Ex. 1003 ¶¶ 161,
163); see also Pet. 22 (“server uses a compression algorithm to create
compressed data blocks.”).
We agree with Petitioner and find that “link speed” is the claimed
“throughput.” Reply 16 (citing Ex. 1004, 10:11–44 (“link speed” is the
“speed of client/server communications link 4”), 6:13–7:17, 8:43–9:30;
Ex. 1003 ¶¶ 152–165); Pet. 26 (citing Ex. 1004, 6:27–45, 1:29–67; Ex. 1003
¶ 160). This finding is supported by Tso’s teaching that selectively scaling
is based on a link speed of a communications link. See Reply 16 (citing
Pet. 22–29; Ex. 1004, 6:13–17, 8:43–9:30; Ex. 1003 ¶¶ 152–165). This
finding is consistent with our construction of “throughput” as including “a
rate of transmission of data.” See Section III.B.2 above.
We also find Tso’s encode manager controls “one or more encode
service providers 8 . . . to selectively scale content based on a predetermined
characteristic,” such as the “speed and/or quality of client/server
communications link 4.” Pet. 23 (citing Ex. 1004, 6:27–45; 4:17–50, 11:6–
17), 26 (citing Ex. 1004, 6:27–45); Reply 17–18 (citing Ex. 1004, 6:27–45;
Pet. 22–29; Ex. 1004, 6:32–36; Ex. 1029 ¶ 39). As set forth in the Petition
as summarized above, we find Tso’s encode service provider may scale or
IPR2019-00712
Patent 8,929,442 B2

28
compress content, including “image, video, or HTML (HyperText Markup
Language).” Ex. 1004, 6:32–36. We credit the Reader Declaration that
Tso’s selection of an encoder is tailored to characteristics of the object being
compressed. Ex. 1003 ¶ 157; see also Ex. 1004, 3:6–14 (“datatype-specific
compression techniques”). In other words, the encode service provider
selects different compression algorithms depending on the type of data
content, “such as image, video, or HTML,” associated with other encode
service providers 8.” Ex. 1004, 6:27–45. We agree with Petitioner and
Dr. Reader that a person of ordinary skill “would have understood that Tso
uses a video encoder for compressing video data, an HTML encoder for
compressing HTML data, etc.” Pet. 24 (citing Ex. 1003 ¶ 157). We credit
the Reader Declaration and find that the recited selection of a compression
algorithm is disclosed or rendered obvious by Tso’s encode manager’s
selection of an encode service provider. Id. (citing Ex. 1003 ¶ 160).
Separate from the Reader Declaration, we find Tso discloses or
teaches selection of a compression algorithm. At page 19 of its Reply,
Petitioner cites additional disclosure of Tso that “[the] non-enabled network
client 3 may optionally be provided the ability to actively control the trade-
off between down load time and content quality by selecting whether or not
to scale content and/or the extent of such scaling.” Ex. 1004, 11:44–48
(emphasis added). The preceding was cited to specifically respond to
“Realtime’s second proposition—that Tso only teaches choosing whether to
compress, not which compression algorithm should be applied.” See
Reply 16. The disclosure at column 11 beginning at line 44, like that of
column 6, refers to the embodiment illustrated in Figure 2 of Tso. See
Ex. 1004, 6:3–4 (“Referring now to FIG. 2, in a first embodiment of the
IPR2019-00712
Patent 8,929,442 B2

29
present invention . . .”), 11:17–65 (still referring to Figure 2, for example,
network client 3, encode manager 7, and encode service provider 8).
At the hearing, Patent Owner argued that the citation to Tso’s
column 11 at line 44 was an improper reply. Tr. 86:25–87:9. We disagree.
First, Patent Owner did not raise the issue in its Sur-reply or seek to file a
motion to strike, either of which would have been appropriate means for
making a record supporting the argument. Second, the only asserted basis
for striking the evidence was that “[t]he portion of the Tso reference here
being relied upon by Petitioner wasn't included in the petition itself.” Id.
at 86:26–87:2. We find that the evidence was a proper response to the
argument in the Response, as discussed above. See Reply 16; see also Trial
Practice Guide, 74 (“Generally, a reply or sur-reply may only respond to
arguments raised in the preceding brief.”); Apple Inc. v. Andrea Electronics
Corp., 949 F.3d 697, 706–707 (Fed. Cir. 2020) (“As we have regularly held,
‘the petitioner in an inter partes review proceeding may introduce new
evidence after the petition stage if the evidence is a legitimate reply to
evidence introduced by the patent owner.’” (citations omitted).
Patent Owner disputes the Reader Declaration testimony that Tso’s
“selective scaling” discloses or teaches the disputed portion of limitation 1.2.
PO Resp. 14 (citing Ex. 1004, 6:27–45; Ex. 1003 ¶ 160). Patent Owner
argues that column 6 of Tso, which is relied on by Dr. Reader, “teaches only
the ‘yes-no dimension of the decision to compress or scale’ by Tso’s HTTP
proxy server.” Id. at 14–15 (citing Ex. 2002,11 61:20–24, 61:17–64:16;
Ex. 2001 ¶¶ 52–53).

11 Realtime Adaptive Streaming LLC, v. Adobe, Inc., 2:18-cv-09344 (C.D.
Cal.), Deposition of Clifford Reader, Ph.D. (“Reader Deposition”).
IPR2019-00712
Patent 8,929,442 B2

30
We find Patent Owner’s argument unpersuasive. The primary support
for this argument is the Zeger Declaration, which relies principally on the
Reader Deposition. See Ex. 2001 ¶¶ 52–53. Tso is cited separately but only
in conclusory fashion and then only to give context to the Reader
Deposition. See, e.g., Ex. 2001 ¶ 52 (citing Ex. 2002, 61:17–64:16) (“the
cited disclosure teaches ‘selectively scal[ing] content’ in the sense that it
uses ‘speed and/or quality of client/server communications link 4’ to select
whether or not to compress content, and not ‘how much’ to compress it.”).
The Reader Deposition does not support Patent Owner or the Zeger
Declaration opinions. Patent Owner’s only specific cite to the Reader
Deposition is Dr. Reader’s understanding as to whether “‘selectively
scale’. . . means a yes-no decision or whether it’s a how-much decision.”
Ex. 2002, 61:17–23. Patent Owner cites several pages of the Reader
Deposition and then picks one sentence out of Dr. Reader’s answer that
Tso’s disclosure is “not quite clear at this point.” PO Resp. 15 (citing
Ex. 2002, 61:7–24; Ex. 2001 ¶ 53); see also Reply 21 (testimony regarding
“selectively scaling objects”). We do not find the testimony raises any issue
about the credibility of the Reader Declaration testimony at paragraphs 160
or 161.
For its second argument that Tso’s “link speed” does not disclose or
teach selection of a compression algorithm, Patent Owner disputes
Petitioner’s evidence from the Reader Declaration that a person of ordinary
skill would have understood that Tso “uses the throughput/bandwidth of the
client/server communications link,” to “select an appropriate compression
algorithm to perform network load balancing.” PO Resp. 16–17 (citing
Ex. 1003 ¶ 161 (citing Ex. 1004, 10:11–44, 12:1–19, 13:15–22)). Patent
IPR2019-00712
Patent 8,929,442 B2

31
Owner argues the Tso disclosures are directed only to using link speed to
make a determination whether or not to compress. Id. at 17 (citing Ex. 2001
¶ 58).
Patent Owner’s argument is not persuasive. Paragraph 58 of the
Zeger Declaration is merely a denial of the opinions expressed in
paragraph 161 of the Reader Declaration. Paragraph 161 of the Reader
Declaration is in addition to what is disclosed in paragraph 160, which relies
on column 6 of Tso. See Ex. 1003 ¶ 161 (“Additionally, Tso maintains link
speed/quality information . . .”). As discussed above, we find Petitioner has
shown persuasively that Tso’s column 6 teaches selection of a compression
algorithm based on, inter alia, “link speed.” Id. ¶ 160 (citing Ex. 1004,
6:27–45).
We find additional support in column 10 of Tso, that “Tso uses the
throughput/bandwidth of the client/server communications link—over which
the compressed data is received—to select an appropriate compression
algorithm to perform network load-balancing.” Ex. 1003 ¶ 161. Patent
Owner argues column 10 of Tso is limited a discussion of how to calculate
“link speed.” PO Resp. 17 (citing Ex. 1004, 10:11–44; Ex. 2001 ¶ 59).
Patent Owner explains that column 10 describes that one method that “link
speed” may be determined is by “measuring the time i[t] takes a ‘ping’ sent
from HTTP remote proxy 6 to network client 3 to return to HTTP remote
proxy 6.” Id. (citing Ex. 1004, 10:21–30). Patent Owner then asserts that
link speed is used to configure client preference table 11, which in turn only
determines whether to compress content or not. Id. (citing Ex. 1004, 10:21–
44). Patent Owner asserts that “link speed” is not disclosed “in any other
way, including, for example, affecting which compression algorithm is
IPR2019-00712
Patent 8,929,442 B2

32
applied or how much compression is obtained.” Id. at 17–18 (citing
Ex. 2001 ¶ 59; Ex. 2002, 68:14–23).
Patent Owner contends that the client preference table of Tso stores IP
addresses of the network client, including a value indicative of whether or
not to scale the content, i.e., compress. PO Resp. 18 (citing Ex. 1004,
10:11–20 (whether or not to scale the content to be returned to network
client is based on “link speed”)). Accordingly, Patent Owner asserts
whether or not to compress is based on “link speed” but selection of a
compression algorithm is not. Id. at 19 (citing Ex. 2001 ¶¶ 59–61). Patent
Owner supports its argument by noting that Tso’s encode manager stores a
scaled version, i.e., a compressed version, even if link speed is high enough
that compression is not needed. Id. at 19–20 (citing Ex. 1004, 11:6–19).
The arguments relating to column 10 and the client preference table of
Tso are unpersuasive because they do not respond to the contentions made in
the Petition and the Reader Declaration testimony. Column 10 is cited as
additional support that Tso discloses or teaches selecting a compression
algorithm based, in part, on link speed. Column 10 discloses that if link
speed indicates a slower communications link, the “HTTP remote proxy 6
may determine that all content to be transmitted to network client 3 should
be scaled by default.” Ex. 1004, 10:33–36 (emphasis added). This
disclosure teaches scaling takes place and the client preference table stores
“derived link speed” (Ex. 1004, 10:31–33) and adapts to “changing
conditions.” See Reply 14–15 (citing Ex. 1004, 10:40–44) (emphases
added). Patent Owner does not explain how scaling by default or adapting
to changing conditions is only a determination of whether to scale or
compress. As taught in column 10 of Tso just described, we find that “Tso
IPR2019-00712
Patent 8,929,442 B2

33
teaches not only choosing whether or not to scale, but also choosing ‘the
extent of such scaling,’ based upon network conditions such as the ‘speed
and/or quality of client/server communications link.’” Id. at 18–19 (citing
Ex. 1029 ¶ 78).
Patent Owner argues Petitioner’s and Dr. Reader’s reliance on
columns 12 and 13 of Tso is “flawed for similar reasons” to those asserted
above for column 10. PO Resp. 20 (citing Ex. 1004, 12:1–19, 13:15–22;
Ex. 1003 ¶ 161; Ex. 2001 ¶¶ 63–64). The column 12 disclosure is directed
to a graphical user interface of Tso shown in Figure 3, which shows a
“SCALING INTERFACE.” Id. However, column 12 describes Figure 3,
which is a different embodiment from what Petitioner relies on in columns 6
and 10 of Tso. See Ex. 1004, 12:1–3 (“In this embodiment, as illustrated in
FIG. 3 . . .”); see also Tr. 58:11–10 (Patent Owner confirming Tso
column 12 is a different embodiment from column 6). We do not rely on the
embodiment disclosed in column 12 to support this Decision. The argument,
therefore, is not persuasive.
According to Patent Owner column 13 of Tso describes Figure 3’s
“SCALING INTERFACE” and a JavaScript implementation for updating
the user preference table based on changing conditions related to, for
example, “increased congestion on client/server communications link 4.”
PO Resp. 22 (citing Ex. 1003, 13:1–22, 10:40–41 (regarding congestion);
Ex. 2002, 71:3–22; Ex. 2001 ¶ 67). This argument is not persuasive for the
same reasons discussed in connection with column 12 above.
Petitioner characterizes Patent Owner’s argument as “throughput”
alone is determinative of which compression algorithm should be applied.
Reply 12 (citing PO Resp. 16). We accept Patent Owner’s representation
IPR2019-00712
Patent 8,929,442 B2

34
that its argument is not based only on “throughput.” See Sur-reply 13–14.
There is no dispute, and we find, that a person of ordinary skill would have
read limitation 1.2 as requiring the algorithm selection to be based upon both
“bandwidth” and a “parameter or attribute.” See Section III.E.2.a above.
Based on the entirety of the record, we find Petitioner has shown
persuasively that Tso discloses or teaches the subject matter of limitation
1.2.
d. Limitation 1.3
Limitation 1.3 recites “storing at least a portion of the decompressed
data block.” Among other showings, Petitioner cites Tso’s network as
including “a client-side cache memory 22” in which “‘the
decompressed/translated data stream’ is stored.” Pet. 29 (citing Ex. 1004,
13:51–65, 15:53–65 (“The decompressed/translated data stream may then be
simultaneously stored in client-side cache memory 22.”); Ex. 1004, Fig. 5
(annotated) above (cache memory 22)).
Patent Owner does not dispute Petitioner’s argument and supporting
evidence regarding the above elements of limitation 1.3. Based on the
entirety of the record, we find Petitioner has shown persuasively that Tso
discloses storing compressed data blocks, as recited in limitation 1.3.
Based on the entire record before us, we find Petitioner has shown
persuasively that Tso discloses the method of independent claim 1.
3. Dependent Claim 7
Claim 7 depends from claim 1 and recites “wherein the at least the
portion of the compressed data block was compressed, prior to
decompression, based upon a user command.” Like claim 1, claim 7 does
IPR2019-00712
Patent 8,929,442 B2

35
not include the term the “compression algorithm is asymmetric,” which is
recited in dependent claim 2. See Section III.E.4 below.
Petitioner argues Tso’s compression and decompression process is
initiated when a user “requests” an object. Pet. 42 (citing Ex. 1004, 9:3–60).
Tso teaches that the user’s request is transmitted as “an HTTP request for the
. . . object to remote scaling server 1 over client/server communications
link 4.” Ex. 1004, 9:44–60. Petitioner argues that “[o]nce the request is
received at remote scaling server 1, encode manager 7 of the server
selectively compresses the data based on a predetermined characteristic.”
Pet. 42–43 (citing Ex. 1004, 6:13–45, 9:60–11:5; Ex. 1003 ¶ 204).
Patent Owner argues claim 7 is patentable for the reasons argued in
connection with claim 1 but does not otherwise dispute Petitioner’s
argument and supporting evidence that Tso discloses or teaches claim 7. PO
Resp. 44. Based on the entirety of the record before us, we find Petitioner
has shown persuasively that Tso discloses or teaches the compressing data
prior to decompressing that data, as recited in claim 7.
4. Dependent Claim 2
Claim 2 depends from claim 1 and recites “wherein at least one of the
plurality of compression algorithms is asymmetric.” The asymmetric term
was construed above in Section III.B.1.
Petitioner argues “at least one of Tso’s plurality of compression
algorithms is asymmetric.” Pet. 29 (citing Ex. 1003 ¶ 169). For example,
Petitioner argues “Tso discloses compression algorithms for particular data
types, including “video/mpeg” [Ex. 1004, 10:60–65] and ‘fractal
compression.’” Id. (citing Ex. 1003 ¶ 169; Ex. 1004, 3:2–13, 11:39–43,
14:33–41, 16:20–46, 16:58–17:2, 17:55–18:20).
IPR2019-00712
Patent 8,929,442 B2

36
Patent Owner argues claim 2 is patentable for the reasons argued in
connection with claim 1 but does not otherwise dispute Petitioner’s
argument and supporting evidence that Tso discloses or teaches claim 2. PO
Resp. 44. Based on the entire record before us, we find Petitioner has shown
persuasively that Tso discloses or teaches asymmetric compression
algorithms, as recited in claim 2.
5. Independent Claim 8
We have reviewed Petitioner’s arguments regarding claim 8. See
Pet. 28–41; Ex. 1003 ¶¶ 208–220. Claim 8 is an independent apparatus
claim of similar scope to claim 1. Petitioner relies on much of its arguments
for claim 1 in asserting claim 8 is unpatentable. See Pet. 28–41.
Exemplary of Petitioner’s analysis for claim 8 is its analysis of
limitation 8.1, which recites “a data decompression system configured to
decompress a compressed data block.” Limitation 1.1 is similar in reciting
“decompressing at least a portion of a compressed data block.” Petitioner
incorporates its showing for limitation 1.1 to show limitation 8.1. Pet. 45.
Similarly, Petitioner’s arguments for limitations 8.2 and 8.3 incorporate the
arguments for limitations 1.3 and 1.2, respectively. Id. at 46–48.
Limitation 8.4 recites the same language as used in dependent claim 2,
analyzed above. Petitioner relies on the arguments made for claim 2 for
limitation 8.4. Pet. 48 (citing Ex. 1003 ¶¶ 219–220).
Patent Owner disputes Petitioner’s argument and supporting evidence
that Tso discloses or teaches independent claim 8 for the same reasons it
argued with respect to limitation 1.2, which argument of Patent Owner we
find unpersuasive. PO Resp. 45. Based on the entire record before us, we
IPR2019-00712
Patent 8,929,442 B2

37
find Petitioner has shown persuasively that Tso discloses or teaches
independent claim 8.
6. Dependent Claims 3–6 and 9–15
As discussed above in Section III.E.1–5, we have determined that
Petitioner has sufficiently shown all the limitations of independent claims 1,
2, 7, and 8 are disclosed or taught by Tso. We have reviewed Petitioner’s
argument and supporting evidence challenging patentability of claims 3–6
and 9–15. Pet. 71–78, 83–87; Ex. 1003 ¶¶ 174–202, 221–238.
Patent Owner disputes Petitioner’s argument and supporting evidence
that Tso discloses or teaches claims 3–6, which depend from claim 1, and 9–
15, which depend from claim 8, for the same reasons it argues with respect
to claims 1 and 8, which argument of Patent Owner we find unpersuasive.
PO Resp. 44–45. Based on the entirety of the record, we find Petitioner has
shown persuasively that Tso discloses or teaches claims 3–6 and 9–15.
7. Single Reference Obviousness of Claims 1–15 Over Tso
Patent Owner argues that the Reader Declaration testimony alone is
insufficient to establish obviousness over Tso. PO Resp. 26–27. As
discussed above, Tso alone anticipates the challenged claims. Petitioner
does not specifically argue inherency, but it argues that, if the disputed
portion of limitation 1.2 is not anticipated, selecting “a compression
algorithm based upon throughput,” would have been obvious to a person of
ordinary skill. See Pet. 26 (citing Ex. 1004, 6:27–45; Ex. 1003 ¶ 160).
Patent Owner’s argument, based in part on the Reader Deposition,
focuses on the lack of testimony that a person of ordinary skill “would
modify Tso.” See PO Resp. 26 (citing Ex. 2002, 75:17–22; Ex. 2001 ¶ 73).
Patent Owner argues “Petitioner’s single-reference obviousness theory is
IPR2019-00712
Patent 8,929,442 B2

38
both unexplained and unsupported by anything other than conclusory
statements.” Sur-reply 19 (citing Arendi, 832 F.3d at 1361–1363).
Petitioner responds that modification is unnecessary and Tso “at least
implicitly” discloses or teaches the arguably missing limitation to a person
of ordinary skill. See Reply 22–23 (citing Pet., 17–22, 34–36; Ex. 1003
¶¶128–151, 185–187).
Arendi is inapposite because it decided a missing limitation could not
be found based on “common sense” alone. Arendi, 832 F.3d at 1361. In
addition, Arendi’s expert evidence was not specifically directed to the
missing limitation. Id. at 1365. Neither is the case here. See Ex. 1003 ¶ 160
(Tso’s “encode manager 7 selects an encode service provider 8 and
corresponding compression algorithm based on the bandwidth/throughout of
the client/server communications link.”). The Reader Declaration is not
conclusory but rather is based on specific disclosures of Tso. Id. (citing
Ex. 1004, 6:27–45, 1:29–67).
Petitioner’s argument is supported by Federal Circuit case law.
Reply 22 (citing Monsanto Tech. LLC v. E.I. DuPont de Nemours & Co.,
878 F.3d 1336, 1346 (Fed. Cir. 2018)). Monsanto is an appeal of an inter
partes review and is similar to the facts here. The Federal Circuit held that
petitioner’s expert declaration on anticipation was not prior art and instead
was extrinsic evidence used to support what is “necessarily present” in a
prior art’s teaching. Monsanto, 878 F.3d at 1345. Furthermore, the extrinsic
evidence of the expert was not rebutted. Id. at 1346. Here, Patent Owner’s
only expert rebuttal by Dr. Zeger is directed to the Reader Deposition
testimony. See Ex. 2001 ¶ 73. The Zeger Declaration does not rebut the
Reader Declaration testimony that Tso, alone and without modification,
IPR2019-00712
Patent 8,929,442 B2

39
would teach or suggest to a person of ordinary skill selecting a compression
algorithm based upon a throughput. Neither paragraph 73 nor any other
testimony from the Zeger Declaration present evidence of a lack of
motivation for us to consider. Instead, the Zeger Declaration addresses only
that the Reader Declaration, as the Reader Deposition acknowledges, does
not assert any modification of Tso. See Ex. 2001 ¶ 73.
Petitioner has not argued that a motivation for modifying Tso is
necessary. Reply 23 (citing Ex. 1029 ¶ 47). Indeed, Petitioner does not rely
on combining elements from Tso to find obviousness. Rather, the exact
same disclosure of Tso that is asserted for anticipation is relied on for
obviousness. Pet. 26 (citing Ex. 1003 ¶ 160); see also Reply 23 (citing
Ex. 1029 ¶¶ 46–48).
“A disclosure that ‘anticipates under § 102 also renders the claim
invalid under § 103, for anticipation is the epitome of obviousness.’”
Realtime Data, LLC v. Iancu, 912 F.3d 1368, 1373 (Fed. Cir. 2019).
Motivation for combining two references is not necessary where one
anticipates, even if anticipation is not alleged. Id. We find the holdings of
Monsanto and Realtime support our conclusion that motivation is
unnecessary to find obviousness here based on a single reference. Monsanto
focused on the expert testimony and, motivation was required to “combine
elements” of the prior art. Realtime supports a conclusion that there is,
under circumstances such as relevant here, a thin line between obviousness
and anticipation.12 Petitioner here relies on the same disclosure for

12 The thin line is illustrated by Cont’l, 948 F.2d at 1268–1269 (Anticipation
determination may be supported by “technological facts known to those in
the field of the invention.”). Knowledge of those in the field, “is not,
IPR2019-00712
Patent 8,929,442 B2

40
anticipation and obviousness and the claims and alleges, if not anticipated,
the claims would have been obvious.
Based on the entire record before us, we find Petitioner has shown
persuasively that Tso would have taught or suggested to a person of ordinary
skill that selecting a compression algorithm based upon a throughput of a
communications channel would have been obvious. Thus, Petitioner has
shown by a preponderance of the evidence that claims 1–15 would have
been obvious over Tso.
8. Summary (Ground 1)
We determine Petitioner has shown by a preponderance of the
evidence that Tso anticipates or renders obvious claims 1–15.
F. Alleged Obviousness of Claims 1–15 Over Vishwanath and Kidder
Petitioner alleges claims 1–15 would have been obvious over
Vishwanath and Kidder. Pet. 4, 53–87. Petitioner also relies on the Reader
Declaration to support its position. See Ex. 1003 ¶¶ 239–350.
1. Vishwanath (Ex. 1005)
Vishwanath describes modifying an output based on characteristics of
application data and of the medium of transmission in a client-server system.
Ex. 1005, code (57), 1:9–13. “[T]he client decodes the adapted output to
produce a modified version of the interactive application that is adapted for
the client.” Id. at 2:38–41. The adapted output may be a selected
compression algorithm dependent upon one characteristic, including speed
or bandwidth of the transmission medium. Id. at 5:64–6:5, 6:50–61.

however, a substitute for determination of patentability in terms of § 103.”
Cont’l, 948 F.2d at 1269.
IPR2019-00712
Patent 8,929,442 B2

41
Compression algorithms are selected based on input data type and
decompressed by a decoder. Id. at 7:1–4.
In one embodiment, Vishwanath describes a server side and a client
side connected through a transmission medium. Ex. 1005, 4:47–49. Figure
2 of Vishwanath is reproduced below.

Figure 2 illustrates the server side and the client side of Vishwanath’s
invention connected through a transmission medium. Ex. 1005, 3:36–38.
“[T]he server side includes an appliance specific transducer 150 and an
adaptive-transmission transducer 152.” Id. at 4:51–53. Appliance-specific
transducer 150 and adaptive-transmission transducer 152 modify
respectively an application based on the capability of the client and
transmission medium 154. Id. at 4:53–57.
Types of applications modified by the adaptive transmission include
documents, internet (including “streaming multimedia”), email, and forms.
Ex. 1005, 5:4–11. Similarly, the transmission medium delivering the
IPR2019-00712
Patent 8,929,442 B2

42
application may be modified based on “the bandwidth, acceptable error
rates, and the latency of the transmission medium.” Id. at 2:62–64, 3:14–17,
4:27–33, 5:67–6:5.
Adaptive-transmission transducer 152, as described above, uses the
characteristics of the application and transmission medium 154 to modify
appliance-specific output 179 to generate adapted output 183. Ex. 1005,
4:53–57. “After the appliance-specific output 179 is modified to generate
the adapted output 183, the server sends the adapted output 183 through the
transmission medium to the client.” Id. at 6:5–8. Generally, selection of the
compression algorithm is done in an adaptive manner. See, e.g., id. at 7:21–
24.
The transmission transducer includes multimode compressor 250 and
adaptive packetizer 252. Ex. 1005, 6:8–12. The packetizer selects the
network protocol to generate the adapted output. Id. at 6:12–14. Multimode
compressor 250 also may automatically select “the compression algorithm in
view of the application, the transmission medium 154 and the client.” Id.
at 6:33–35. Parameters used to select the compression algorithm may
include input data type and bandwidth versus quality. Id. at 6:50–61.
Compression algorithms disclosed include H.263 and MPEG 1 and 2. Id.
at 6:62–67.
2. Kidder (Ex. 1006)
Kidder describes a method and apparatus where “a network client
requests multiple downloads of a video clip stored on the network server.”
Ex. 1006, code (57). “In response to each request, the network server
compresses the video clip and transmits the compressed data in a scalable
bitstream.” Id.
IPR2019-00712
Patent 8,929,442 B2

43
The apparatus described improves the quality of graphic or audio
information over a network. Ex. 1006, code (57). In one embodiment, a
network client requests multiple downloads of a video clip stored on the
network server, “compresses the video clip[,] and transmits the compressed
data in a scalable[13] bitstream.” Id. The server uses “H.263 or MPEG
[compression standards] for compressing the data to be transmitted.” Id.
at 6:48–50.
Kidder explains that it is desirable to improve “the quality of a
transmitted video clip by increasing the effective bandwidth available for the
transmission and playback of the video clip” and to use cached data to
“improve the quality of subsequent viewing instances.” Ex. 1006, 2:37–42.
“[A] node coupled to the network receives and stores a first set of data
which represents a data object and receives a second set of data which
represents the data object.” Id. at 2:47–49. Kidder receives compressed
data, decompresses the data, and stores the decompressed data in a cache.
Id. at 7:2–8:28.
3. Claim 1
Petitioner argues Vishwanath describes the preamble of claim 1, a
“method comprising,” by teaching a “[m]ethod . . . to deliver an application
to a client.” Pet. 58 (citing Ex. 1005, Abstract, Fig. 3, claims 10–13). If
limiting, Patent Owner does not dispute Petitioner’s argument and
supporting evidence that Vishwanath teaches the preamble. Based on the

13 The Reader Declaration asserts that “scaled” means “compressed.”
Ex. 1003 ¶ 114.
IPR2019-00712
Patent 8,929,442 B2

44
entire record, we find Petitioner has shown persuasively that Vishwanath
teaches the preamble of claim 1.14
a. Limitation 1.1
Limitation 1.1 recites “decompressing at least a portion of a
compressed data block that was selected from among a plurality of
compressed data blocks.” Petitioner relies on Vishwanath’s teaching of a
client that communicates with a server over a transmission medium.
Pet. 60–61 (citing Ex. 1005, 4:46–51, annotated Fig. 2). Petitioner further
argues “Vishwanath’s client ‘generat[es a] modified version of an
application from [the] adapted [i.e., compressed] output’ [Ex. 1005, 2:54–
59] by decoding it with decoder 156.” Id. at 61 (citing Ex. 1005, 2:22–42,
claims 12–13). Petitioner relies on the understanding of a person of ordinary
skill relating to transmission of packets as including selected data blocks of
variable size. Id. at 62 (citing Ex. 1003 ¶¶ 48–88, 257–261; Ex. 1001, 7:3–7
(files are comprised of data blocks); Ex. 1005, 7:61–8:23 (“Incremental
reconstruction of the output at the client.”)).
Petitioner cites to both Vishwanath and Kidder as teaching “receiving
compressed data as a stream, and decompressing the stream of data.”
Pet. 63 (citing Ex. 1005, 5:1–12, 7:18–42, 7:61–8:22; Ex. 1006, 6:25–47,
7:2–8:28). Relying on the Reader Declaration, Petitioner contends a person
of ordinary skill would understand a data stream to include a plurality of
compressed blocks. Id. at 63–64 (citing Ex. 1003 ¶¶ 265–268 (citing

14 Petitioner also cites to Kidder as teaching various limitations. See, e.g.,
Pet. 59 (Kidder also describes a method). Unless otherwise stated, for
purposes of this Decision we analyze only Vishwanath.
IPR2019-00712
Patent 8,929,442 B2

45
Ex. 1005, 5:1–12, 7:18–42, 7:61–8:22; Ex. 1006, 6:25–47, 7:2–8:28;
Ex. 1014,15 Abstract, 5:27–31; Ex. 1024,16 3:33–54)).
Patent Owner does not dispute Petitioner’s argument and supporting
evidence that Vishwanath teaches limitation 1.1. Based on the record before
us, we find Petitioner has shown persuasively that Vishwanath teaches a
server and client encoding and decoding compressed data transmitted in data
blocks, as recited in limitation 1.1.
b. Limitation 1.2-Undisputed Portions
Limitation 1.2 recites:
wherein at least a portion of a data block having video or audio
data was compressed with one or more compression algorithms
selected from among a plurality of compression algorithms based
upon a throughput of a communication channel and a parameter
or an attribute of the at least the portion of the data block to create
one or more of the plurality of compressed data blocks.

Pet. 64.
Petitioner cites to Vishwanath’s teaching of “many different types of
media,” including audio and video. Pet. 64 (citing Ex. 1005, 2:10–13, 5:4–
11, 6:50–57). Petitioner argues Vishwanath shows one or more compression
algorithms because, among other showings, “Vishwanath’s transducer
selects, as illustrated in FIG. 7, an ‘appropriate compression algorithm’
[Ex. 1005, 6:7–15] to apply to the requested data from among ‘[a] number of
compression algorithms[] applicable,’ including ‘LZ . . . MPEG1, MPEG2,
[and] H.263.” Id. at 65 (citing Ex. 1005, 6:32–37).

15 Kalmanek et al., EP 0739138 A2; published April 10, 1996.
16 Porter et al., U.S. 6,519,286 B1; issued February 11, 2003.
IPR2019-00712
Patent 8,929,442 B2

46
Petitioner further argues “Vishwanath’s system selects a compression
algorithm based on parameters such as ‘[i]nput data type’ and available network
bandwidth.” Pet. 66 (citing Ex. 1005, 6:50–62, Fig. 9B). According to Petitioner,
compression of the data block “based upon a throughput of a communication
channel” is shown because Vishwanath “seeks to improve data quality and user
experience over varying transmission media” and automatically modifies the
requested data based on “the resources available.” Id. (citing Ex. 1005, 1:9–13,
2:60–3:33). In addition, Petitioner contends Vishwanath “selects a compression
algorithm for a particular application based on parameters, such as the type of data
within the application in addition to available network bandwidth.” Id. at 67
(citing Ex. 1005, 6:50–62).
Patent Owner does not dispute any of the preceding argument and
supporting evidence regarding limitation 1.2. We agree with Petitioner’s
showing summarized above and find Petitioner has shown persuasively the
undisputed portions of limitation 1.2.
c. Limitation 1.2-Disputed Portion
Patent Owner disputes that Vishwanath teaches selecting a
compression algorithm based upon “throughput.” See PO Resp. 28–34;
Reply 24–27; Sur-reply 20. Patent Owner does not contest that Vishwanath
teaches the remainder of limitation 1.2. See Section III.F.3.a above.
We frame the disputed issue as follows. Patent Owner argues that
Petitioner’s construction of “throughput” as “a rate of transmission of data”
is not correct. See Section III.B.2 above. Specifically, Patent Owner argues
that Vishwanath teaches “channel capacity,” which is not the same at
“throughput.” See PO Resp. 28–30. Furthermore, Patent Owner argues that
Vishwanath’s disclosure of “bandwidth” is not the recited “throughput.” Id.
IPR2019-00712
Patent 8,929,442 B2

47
at 30–34; see, e.g., 31–32 (citing Ex. 2001 ¶¶ 82–83 (“the word ‘bandwidth’
in Vishwanath refers to the channel capacity (i.e. maximum possible
throughput)”). For reasons discussed below, these arguments are not
persuasive.
Petitioner argues “compression selection is based on ‘the bandwidth
[i.e., throughput] . . . of the transmission medium.’” Pet. 67 (citing
Ex. 1005, 2:62–64, 6:50–67). In addition, Petitioner contends Vishwanath
“selects a compression algorithm for a particular application based on
parameters, such as the type of data within the application in addition to
available network bandwidth.” Id. at 67 (citing Ex. 1005, 6:50–62).
Petitioner argues Vishwanath shows one or more compression algorithms
because, among other showings, “Vishwanath’s transducer selects, as
illustrated in FIG. 7, an ‘appropriate compression algorithm’ [Ex. 1005, 6:7–
15] to apply to the requested data from among ‘[a] number of compression
algorithms[] applicable,’ including ‘LZ . . . MPEG1, MPEG2, [and] H.263.”
Id. at 65 (citing Ex. 1005, 6:32–37).
Patent Owner first argues a person of ordinary skill “would not equate
the term ‘throughput’ in the context of the ’442 patent claims with
Vishwanath’s disclosure of channel capacity.” PO Resp. 28–30 (citing
Ex. 2001 ¶ 77. Relying on the Zeger Declaration, Patent Owner identifies
three separate understandings17 a person of ordinary skill would have for the
term “bandwidth.” Id. at 28–30 (citing Ex. 2001 ¶¶ 78–79); see also id.
at 29–30 (citing Ex. 1005, 7:65–8:3, 8:33–36, 9:56; Ex. 2001 ¶ 79)
(arguing ’442 patent describes other methods for controlling “throughput”).
Patent Owner alleges Dr. Reader’s testimony “incorrectly assumes that the

17 These “understandings” were discussed generally in Section II.B.2 above.
IPR2019-00712
Patent 8,929,442 B2

48
claimed ‘throughput’ is the same thing as channel capacity (which may be
an alternative definition of “bandwidth”), which does not and cannot be
changed by a compression algorithm or routine.” Id. at 30 (citing Ex. 2001
¶ 80). Based on the preceding, Patent Owner contends “throughput” “is not
limited to and not the same thing as the capacity (i.e., maximum throughput)
of a given channel.” Id. at 29.
Neither the ’442 patent nor Dr. Reader use the term “channel
capacity” in conjunction with “throughput” or otherwise. See Reply 25.
Dr. Zeger does use the term but, without any support, attributes it to
Dr. Reader. See Ex. 2001 ¶¶ 77, 80. Indeed, Petitioner asserts, and we
agree, it is Patent Owner that is incorrect. Id. at 24. Petitioner re-alleges its
proposed construction that “throughput” means “a rate of transmission of
data.” Id. at 25; see also Section III.B.2 above (adopting Petitioner’s
construction). That the Zeger Declaration is based on a mistaken
understanding of the Reader Declaration causes us to give it little weight on
this point. See id. (Patent Owner “‘therefore applies an incorrect
understanding’ to its analysis of the Petition’s and Dr. Reader’s arguments.”)
(citing PO Resp. 28; Ex. 2001 ¶ 77). Other than alleging generally
“bandwidth” is not “throughput,” Patent Owner has no response to the
assertion that it has “an incorrect understanding.” See Sur-reply 20.
Patent Owner argues next that “Vishwanath’s discussion of
‘bandwidth’ is not the ‘throughput of a communications channel’ in the ‘442
patent claims.” PO Resp. 30–34. Patent Owner argues Vishwanath’s
teachings of “bandwidth” should be interpreted to mean “channel capacity,”
the “maximum possible throughput.” Id. at 31 (citing Ex. 1005, 1:27–31,
7:51–53 (both describing bandwidth differences based on transmission
IPR2019-00712
Patent 8,929,442 B2

49
medium); Ex. 2001 ¶ 82 (Vishwanath’s use of “bandwidth” means channel
capacity and does not use the word “throughput”).
Patent Owner argues Vishwanath does not teach “throughput” as it
proposes we construe the term. PO Resp. 31–33 (citing Ex. 2001 ¶¶ 83–85).
Patent Owner attempts to rebut the Reader Declaration by arguing
Dr. Reader’s citations to Vishwanath do not teach selection of a compression
algorithm based on “any consideration of the amount of data flowing
through the communication channel,” asserting Patent Owner’s construction
of “throughput.” Id. at 32–33 (citing Ex. 2001 ¶¶ 83–85). Patent Owner
further attempts to distinguish Vishwanath’s “bandwidth” teaching from
“how the term is used in the ’442 patent,” contending the Reader Declaration
“incorrectly equates . . . ‘bandwidth’” to “throughput.” Id. at 33 (citing
Ex. 2001 ¶¶ 83–86).
Patent Owner’s arguments based on its construction of “throughput”
are also not persuasive because we determine that “throughput” includes
“bandwidth” measured at a rate of transmission, i.e., bits per second. See
Section III.B.2 above. Patent Owner relies all but exclusively on the Zeger
Declaration and its rebuttal of the Reader Declaration. See, e.g., PO
Resp. 33 (citing Ex. 2001 ¶¶ 83–86). As discussed above, the Zeger
Declaration contains an error which causes us to give it little weight relating
to what Vishwanath teaches. Beyond the error, the Zeger Declaration
simply repeats the allegations in the Patent Owner Response and does not
provide sufficient basis for us to find it supports Patent Owner’s position.
Conversely, we find the Reader Declaration credible.
We find substantial evidence from the ’442 patent that “bandwidth,”
which Vishwanath discloses, is the equivalent of “throughput.” See,
IPR2019-00712
Patent 8,929,442 B2

50
Ex. 1001, 1:25, 7:54, 9:13 (“throughput (bandwidth)”); see also Pet. 66–67
(citing Ex. 1005, 2:62–64, 6:50–67) (“the parameters used to choose the
compression algorithm include . . . the compromise between bandwidth
versus quality”); Ex. 1003 ¶¶ 271, 273; Ex. 1029 ¶¶ 55–57 (person of
ordinary skill “would have equated Vishwanath’s discussion of ‘bandwidth’
with the claimed ‘throughput of a communications channel.’”). Again, other
than alleging generally “bandwidth” is not “throughput,” Patent Owner has
no response. See Sur-reply 20.
Based on the entire record, we find Petitioner has shown persuasively
that Vishwanath teaches the subject matter of limitation 1.2.
d. Limitation 1.3
Limitation 1.3 recites “storing at least a portion of the decompressed
data block.” Petitioner relies on Kidder’s teaching that decompressed data is
stored in a cache. Pet. 68 (citing Ex. 1006, 7:2–8:28). Patent Owner does
not dispute that Kidder teaches limitation 1.3 but does dispute that a person
of ordinary skill would have combined Kidder with Vishwanath. PO Resp.
34–44.
Petitioner cites Kidder to teach limitation 1.3. Pet. 68–70. Petitioner
argues that “Vishwanath does not explicitly describe storage of the
decompressed data,” as recited in limitation 1.3. Id. at 68. According to
Petitioner, Kidder meets the storage limitation by teaching receipt of
“compressed data, decompresses the data, and stores the decompressed data
in a cache.” Id. (citing Ex. 1006, 7:2–8:28). Patent Owner argues Petitioner
has not shown how and why Vishwanath and Kidder would have been
combined. PO Resp. 34–44.
IPR2019-00712
Patent 8,929,442 B2

51
Petitioner argues the reason a person of ordinary skill would have
been motivated to combine Kidder with Vishwanath is found in
Vishwanath’s description of “incremental reconstruction of the output at the
client” for “smart-loading which improves the user experience by providing
them with some data quickly.” Pet. 53–54 (citing Ex. 1005, 8:10–15).
Petitioner notes that details of the output reconstruction process are lacking
in Vishwanath but are found in Kidder’s teaching of caching data for
“‘iterative improvement’ of data quality.” Id. at 54 (citing Ex. 1006, 6:25–
8:28), 55–56 (citing Ex. 1003 ¶¶ 239–246 (Kidder’s caching technique)).
Petitioner argues a person of ordinary skill
would have understood how to implement Kidder’s storage
structure and iterative transmission techniques in Vishwanath’s
system with a reasonable expectation of success, stemming from
the significant overlap across the references in their teachings
and suggested approaches for adaptive compression and
transmission of data over communications channels of varying
quality.

Pet. 56 (citing Ex. 1003 ¶¶ 239–247) (emphasis added).
That Kidder stores different versions of the compressed data is a
feature related to the “adaptive compression” taught in both references.
Ex. 1003 ¶¶ 241–243 (citing Ex. 1005, 2:8–65; Ex. 1006, 6:25–8:28).
Storage of decompressed data, as shown in Kidder would be an asset to data
quality, i.e., “iterative improvement” using a cache. Ex. 1003 ¶ 243. Using
cache memory feature is a “detail” of improving data quality not found in
Vishwanath. Ex. 1003 ¶ 245.
Petitioner argues a person of ordinary skill
would have understood how to implement Kidder’s storage
structure and iterative transmission techniques in Vishwanath’s
system with a reasonable expectation of success, stemming from
IPR2019-00712
Patent 8,929,442 B2

52
the significant overlap across the references in their teachings
and suggested approaches for adaptive compression and
transmission of data over communications channels of varying
quality.

Pet. 56 (citing Ex. 1003 ¶¶ 239–247) (emphasis added). Petitioner uses an
annotation combining Figure 2 of Vishwanath with Figures 1 and 2 of
Kidder, which annotation is reproduced below. Id. at 57.

Petitioner’s Annotation of Vishwanath Fig. 2
and Kidder Figs. 1 and 2

Id. The above annotation shows how a person of ordinary skill would “have
implemented Kidder’s storage and adaptive compression system, as shown
in Kidder’s FIGS. 1 and 2, in Vishwanath’s system by modifying
Vishwanth’s client to cache requested data that was received and
IPR2019-00712
Patent 8,929,442 B2

53
decompressed (like Kidder’s client 102).” Id. at 56–57. Petitioner alleges
“[i]mplementing these modifications would require only routine knowledge
of electronics hardware design and conventional programming proficiency,
which were well within the skill of a [a person of ordinary skill] prior to the
Critical Date.” Id. (citing Ex. 1003 ¶ 250).
The Reader Declaration includes testimony supporting the arguments
described above. See Ex. 1003 ¶¶ 241–246. Kidder “supplies details” of an
iterative process, i.e., receiving compressed data, decompressing the data,
and storing the decompressed data in cache memory. Id. ¶ 245 (citing
Ex. 1006, 6:25–8:28). Dr. Reader testifies that
[b]ecause Kidder describes a method for decompressing and
caching different versions of requested data to iteratively
improve data quality [(Ex. 1006, 6:25–8:28)], one of ordinary
skill would have been motivated to modify Vishwanath’s
adaptive transmission system to also store decompressed data
[by employing Kidder’s caching technique].

Id. ¶ 246.
Petitioner argues Vishwanath and Kidder are related art. According to
Petitioner, “Vishwanath describes an adaptive compression and transmission
system that selects an appropriate compression algorithm based on a
parameter of requested multimedia (e.g., audio and/or video) content and
available bandwidth of a transmission medium over which the multimedia
content is delivered.” Pet. 53 (citing Ex. 1005, 2:8–65; Ex. 1003 ¶¶ 239–
251). Petitioner argues Kidder is like Vishwanath in describing “a system
that adaptively compresses data based on the ‘available bandwidth of [a]
network’ over which the data is transmitted.” Id. (citing Ex. 1006, 6:25–
8:28). Petitioner argues both references also describe compression and
transmission systems that “incrementally improve data by transmitting
IPR2019-00712
Patent 8,929,442 B2

54
different versions of requested data to the same client.” Id. (citing Ex. 1005,
code (57), 7:61–8:23; Ex. 1006, 6:25–8:28).
Patent Owner argues Petitioner has not shown “why” Vishwanath
would have been modified by Kidder. See PO Resp. 34 (citing Ex. 2001
¶¶ 88–108). Patent Owner argues that Petitioner’s alleged motivation for the
combination, that “‘[s]torage shown in Kidder would be an asset to speed,
i.e., using a cache’ and ‘[u]sing [a] cache memory feature is a ‘detail’ of
improving data quality not found in Vishwanath,’” and functions differently
from Vishwanath. Id. at 35–36 (citing Ex. 1003 ¶¶ 245–246). Patent Owner
contends Kidder’s cache memory “involve[s] the storage and delivery of a
‘scalable bitstream.’” Id. (citing Ex. 1006, 6:1–5; Ex. 2002, 83:5–98:13;
Ex. 2001 ¶¶ 90–92).
Patent Owner asserts the scalable bitstream is compressed “only once
. . . to generate a single video bitstream . . . that in total represents the
highest possible quality video available.” PO Resp. 36–37 (citing Ex. 2002,
135:20–136:5; Ex. 2001 ¶ 92); see id. at 38 (that the bitstream compressed
only once saves bandwidth) (citing Ex. 2001 ¶¶ 94–95; Ex. 1006, 7:45–49,
10:29–35). According to Patent Owner, decompression of the bitstream
requires storing subsequent bitstreams to obtain higher quality video. Id.
(citing Ex. 2001 ¶ 95; Ex. 1006, 10:29–35 (different versions of the video
are sent to iteratively improve each version to achieve quality of original
uncompressed signal)).
Patent Owner states the scalable bitstream technology of Kidder is not
the predominant model for Internet video streaming. PO Resp. 38–41.
Among the reasons alleged for not adopting scalable bitstream technology
include the need for multiple encodings, technical complexity, cost, the
IPR2019-00712
Patent 8,929,442 B2

55
introduction of H.264–SVC (“Scalable Video Codec”), and limitations of
one bitstream over two independent bitstreams. Id. at 39–40 (citing
Ex. 2001 ¶¶ 96–98; Ex.2002, 92:22–93:13, 96:22–97:5). Patent Owner’s
conclusion is that a person of ordinary skill would find scalable bitstream
technology was undesirable, “and, importantly, did not require storage of
partial bitstreams in the local memory of users’ devices before the user could
view the highest possible quality video given bandwidth constraints.” Id.
at 40–41 (citing Ex. 2001 ¶ 99; Ex. 2002, 100:6–13).
Patent Owner contrasts Vishwanath, which does not use a single
scalable bitstream, to Kidder. PO Resp. 41. Patent Owner argues
Vishwanath “does not require persistent local storage.” Id. (citing Ex. 1005,
3:3–6 (“server can adapt the applications accordingly” to “different media,
such as wired and wireless, digital and analog”); Ex. 2001 ¶ 100). Patent
Owner alleges Vishwanath tailors the encoded video data using its
“adaptive-transmissision encoder 152.” Id. at 41–42 (citing Ex. 1005, 2:33–
37; Ex. 2001 ¶ 101). According to Patent Owner, Vishwanath does not
combine iterations of transmitted video files. Id. at 42 (citing Ex. 2001
¶ 102). Furthermore, Patent Owner contends a person of ordinary skill
would not combine Kidder’s “undesirable, costly, and cumbersome”
technology with the non-scalable streaming approach of Vishwanath. Id.
at 42–43 (citing Ex. 2001 ¶ 103).18
Patent Owner’s arguments are not persuasive, because, as described
above, we find on this record that Vishwanth and Kidder are sufficiently

18 We need not analyze Patent Owner’s contention that Kidder does not
“expressly disclose selecting a compression algorithm based on throughput
or bandwidth.” PO Resp. 44 (citing Ex. 2001 ¶ 108). As already analyzed,
Petitioner relies on Vishwanath for limitation 1.2.
IPR2019-00712
Patent 8,929,442 B2

56
related that one of ordinary skill would have consulted Kidder when looking
to improve the speed and data quality of Vishwanath. For example, both
references teach use of “H.263 or MPEG” compression algorithms.
Ex. 1003 ¶ 248 (citing 1005, 6:62–70; Ex. 1006, 6:48–50). As shown in part
by the annotation of Figure 2 of Vishwanath and Figures 1 and 2 of Kidder
above, a person of ordinary skill would have had a reasonable expectation of
success in combining Kidder with Vishwanath. We find the annotation
persuasive, as we did in the Institution Decision. See Inst. Dec. 38.
We are not persuaded that the differences between Vishwanath and
Kidder overcome the similarities between the two that would suggest the
combination. We credit the Reader Reply Declaration regarding Kidder that
“the data is encoded at least once for each iteration, or layer of scalability,”
not “only” once. Reply 32 (citing Ex. 1029 ¶ 64). We also agree that
whether the Kidder technology has not been widely adopted is irrelevant.
See id. Kidder is prior art and its teachings would have been before a person
of ordinary skill. See Beckman Instruments, Inc. v. LKB Produkter AB, 892
F.2d 1547, 1551 (Fed. Cir. 1989) (reference is “prior art for all it teaches”).
Based on the entire record, we find Petitioner has shown persuasively
that a person of ordinary skill would have motivation to combine
Vishwanath and Kidder and have a reasonable expectation of success in
doing so. We find Petitioner has shown persuasively that Kidder teaches
storing compressed data blocks, as recited in limitation 1.3.
We find that Petitioner has shown that the subject matter of claim 1
would have been obvious over Vishwanath and Kidder.
IPR2019-00712
Patent 8,929,442 B2

57
4. Claim 7
Claim 7 recites “wherein the at least the portion of the compressed
data block was compressed, prior to decompression, based upon a user
command.” Petitioner argues Vishwanath allows a user to enter a user
preference after which “the system adaptively compresses the data and
‘deliver[s] it to’ the user’s client device through a transmission medium.”
Pet. 79 (citing Ex. 1005, 2:22–42, 4:12–38). Relying in part on the Reader
Declaration, Petitioner asserts the “user’s preference initiating the adaptive
compression process renders obvious use of a user command, and that the
compressed data block was compressed, prior to decompression at the client,
based upon this user command.” Id. (citing Ex. 1003 ¶ 318).
Patent Owner argues claim 7 is patentable for the reasons argued in
connection with claim 1 but does not otherwise dispute Petitioner’s
argument and supporting evidence that the combination of Vishwanath and
Kidder teaches claim 7. PO Resp. 44. Based on the record before us, we
find Petitioner has shown persuasively that the combination of Vishwanath
and Kidder teaches a user command for compressing data prior to
decompressing that data, as recited in claim 7.
5. Claim 2
Claim 2 recites “wherein at least one of the plurality of compression
algorithms is asymmetric.” Petitioner acknowledges that “neither
Vishwanath nor Kidder uses the word ‘asymmetric,’ a [person of ordinary
skill] reviewing Vishwanath or Kidder would have understood or found
obvious the asymmetric nature of these compression algorithms.” Pet. 70
(citing Ex. 1005, 6:62–67, Fig. 7; Ex. 1006, 6:48–60 (“H.263 or MPEG”);
Ex. 1003 ¶¶ 252–287). Relying in part on the Reader Declaration, Petitioner
IPR2019-00712
Patent 8,929,442 B2

58
asserts MPEG and H.263 were known asymmetric techniques. Id. at 70–71
(citing Ex. 1003 ¶¶ 68–87, 287).
Patent Owner argues claim 7 is patentable for the reasons argued in
connection with claim 1 but does not otherwise dispute Petitioner’s
argument and supporting evidence that both Vishwanath and Kidder teaches
claim 2. PO Resp. 44. Based on the entire record, we find Petitioner has
shown persuasively that the combination of Vishwanath and Kidder teaches
claim 2.
6. Claim 8
Petitioner relies on much of its argument for claim 1 in asserting
claim 8 is unpatentable. Pet. 80–83. Exemplary is limitation 8.1, which
recites “a data decompression system configured to decompress a
compressed data block.” Limitation 1.1, which recites, in principal part,
“decompressing at least a portion of a compressed data block.” Petitioner
incorporates its arguments for limitation 1.1 to show limitation 8.1. Id. at 80
(“See Ground 2 [1.1] incorporated here.”). Similarly, Petitioner’s argument
for limitations 8.2 and 8.3 incorporate the argument for limitations 1.3 and
1.2, respectively. Id. at 81–83.
Limitation 8.4 recites the same language as used in dependent claim 2,
analyzed above. Petitioner relies on the argument made for claim 2 for
limitation 8.4. Pet. 83 (citing Ex. 1003 ¶¶ 284–292, 336–337).
Patent Owner disputes Petitioner’s argument and supporting evidence
that Vishwanath and Kidder teach independent claim 8 for the same reasons
it argued with respect to limitation 1.2, which we find unpersuasive. PO
Resp. 45. Based on the entire record, we find Petitioner has shown
IPR2019-00712
Patent 8,929,442 B2

59
persuasively that the combination of Vishwanath and Kidder teaches claim
8.
7. Dependent Claims 3–6 and 9–15
We have reviewed Petitioner’s arguments regarding dependent
claim 3–6 and 9–15. See Pet. 71–78 (claims 3–6), 83–87 (claims 9–15);
Ex. 1003 ¶¶ 293–316 (claims 3–6), 338–350 (claims 9–15).
Patent Owner disputes Petitioner’s argument and supporting evidence
that the combination of Vishwanath and Kidder teaches claims 3–6, which
depend from claim 1, and 9–15, which depend from claim 8, for the same
reasons it argued with respect to claims 1 and 8, which we find
unpersuasive. PO Resp. 44–45. Based on the entire record, we find
Petitioner has shown persuasively that the combination of Vishwanath and
Kidder teaches dependent claim 3–6 and 9–15.
8. Summary (Ground 2)
Petitioner has shown by a preponderance of the evidence that
claims 1–15 would have been obvious over the combination of Vishwanath
and Kidder.
IV. CONCLUSION19
Petitioner has shown by a preponderance of the evidence that
claims 1–15 of the ’442 patent would have been unpatentable as anticipated

19 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
IPR2019-00712
Patent 8,929,442 B2

60
and obvious over Tso. Among other findings, we specifically find Tso
discloses or teaches selecting a compression algorithm from a plurality of
compression algorithms.
Petitioner has shown by a preponderance of the evidence that
claims 1–15 of the ’442 patent would have been unpatentable as obvious
over the combination of Vishwanath and Kidder. Among other findings, we
specifically find that Vishwanath’s “bandwidth” is the recited “throughput.”
In summary:

V. ORDER
In consideration of the foregoing, it is hereby:
ORDERED that claims 1‒15 of the ’442 patent are determined to be
unpatentable; and
FURTHER ORDERED that, because this is a Final Written Decision,
parties to the proceeding seeking judicial review of the decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
Claims

35
U.S.C.
§
Reference(s)/Basis Claims
Shown
Unpatentable
Claims
Not Shown
Unpatentable
1–15 102 Tso 1–15
1–15 103 Tso 1–15
1–15 103 Vishwanath, Kidder 1–15
Overall
Outcome
1–15
IPR2019-00712
Patent 8,929,442 B2

61
FOR PETITIONER:

Jeremy J. Monaldo
Jonathan J. Lamberson
W. Karl Renner
FISH & RICHARDSON P.C.
jjm@fr.com
lamberson@fr.com
axf-ptab@fr.com

FOR PATENT OWNER:
Philip Wang
Neil Rubin
C. Jay Chung
Kent Shum
Reza Mirzaie
James S. Tsuei
RUSS AUGUST & KABAT
pwang@raklaw.com
nrubin@raklaw.com
jchung@raklaw.com
kshum@raklaw.com
rmirzaie@raklaw.com
jtsuei@raklaw.com