Netflix, Inc. v. Divx, LLC

80 F.4th 1352

Opinion

Case: 22-1138 Document: 41 Page: 1 Filed: 09/11/2023

United States Court of Appeals for the Federal Circuit ______________________

NETFLIX, INC., Appellant

v.

DIVX, LLC, Appellee ______________________

2022-1138 ______________________

Appeal from the United States Patent and Trademark Office, Patent Trial and Appeal Board in No. IPR2020- 00646. ______________________

Decided: September 11, 2023 ______________________

THOMAS SAUNDERS, Wilmer Cutler Pickering Hale and Dorr LLP, Washington, DC, argued for appellant. Also rep- resented by DAVID P. YIN; MARK CHRISTOPHER FLEMING, JAMIE N. HADDAD, Boston, MA.

NATHAN NOBU LOWENSTEIN, Lowenstein & Weather- wax LLP, Santa Monica, CA, argued for appellee. Also rep- resented by PARHAM HENDIFAR, KENNETH J. WEATHERWAX. ______________________

Before HUGHES, STOLL, and STARK, Circuit Judges. Case: 22-1138 Document: 41 Page: 2 Filed: 09/11/2023

STOLL, Circuit Judge. Netflix, Inc. appeals the Patent Trial and Appeal Board’s final written decision in an inter partes review of U.S. Patent No. 8,472,792. Netflix challenges the Board’s finding that an asserted prior art reference fails to qualify as analogous art. We hold that the Board abused its dis- cretion in determining that Netflix failed to identify the field of endeavor for either the ’792 patent or the prior art and thus failed to establish analogous art under the field of endeavor test. The Board’s directive that Netflix more pre- cisely articulate the relevant field of endeavor to meet its burden was unduly strict. We therefore vacate the Board’s field of endeavor finding and remand for the Board to re- consider that factual question consistent with this opinion. BACKGROUND I The ’792 patent explains, in a section titled “Back- ground of the Invention,” that “[t]he present invention re- lates generally to encoding, transmission and decoding of multimedia files.” ’792 patent col. 1 ll. 20–21; see also id. at col. 1 ll. 38–40. The abstract of the patent similarly be- gins: “A multimedia file and methods of generating, dis- tributing and using the multimedia file are described.” Continuing, the abstract states: “One embodiment of a multimedia file in accordance with the present invention includes a series of encoded video frames, a first index that includes information indicative of the location within the file and characteristics of each encoded video frame and a separate second index that includes information indicative of the location within the file of a subset of the encoded video frames.” The detailed description of the invention describes a file structure compliant with Microsoft’s Re- source Interchange File Format (RIFF), which is used for storing multimedia information. Id. at col. 5 ll. 33–39. Rel- evant here is a special version of the RIFF format—the Au- dio Video Interleave (AVI) file—which contains additional Case: 22-1138 Document: 41 Page: 3 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 3

storage structures called chunks. Id. at col. 5 l. 51–col. 6 l. 17. As characterized by the ’792 patent, a standard AVI file contains an “idx1” chunk that contains information about every single frame. See id. at col. 15 ll. 15–18, col. 22 ll. 20–22. The invention implements a multimedia file based on the AVI file structure and requires an additional chunk—the “index” chunk—that contains information for a subset of frames. See id. at col. 15 ll. 15–18. DivX, the patent owner, emphasizes that the ’792 pa- tent describes and claims a feature called trick play func- tionality, which is the ability to “fast forward, rewind and scene skip[]” frames. Id. at col. 16 ll. 26–29; see Appellee’s Br. 5. The specification describes a process for implement- ing this functionality by locating a specific frame within a multimedia file using an “index” chunk. See ’792 patent Fig. 4.0.1, col. 4 ll. 29–32. After identifying the frame to be sought, the invention searches through the “index” chunk to find “tag” chunks that reference the frames closest to the desired frame (i.e., the desired frame is located between the frames referenced by the “tag” chunks) and returns the po- sition of the video frame and any audio referenced by the “tag” chunk within the multimedia file. Id. at col. 48 ll. 40– 46, col. 49 ll. 7–23. As claimed, the “index” chunk is “lo- cated prior to the series of encoded video frames and the first index,” id. at col. 51 ll. 40–45, which allows for trick play functionality “prior to the downloading of the ‘idx1’ chunk,” id. at col. 16 ll. 26–29. Claim 1 is illustrative of the claimed invention: 1. A decoder for decoding a multimedia file com- prising at least one video track and at least one au- dio track, the decoder comprising: a processor; and memory having a multimedia file including: a series of encoded video frames; Case: 22-1138 Document: 41 Page: 4 Filed: 09/11/2023

a first index that includes information in- dicative of the location within the file and characteristics of each encoded video frame; and a separate second index that includes infor- mation indicative of the location within the file of a subset of the encoded video frames, the separate second index located prior to the series of encoded video frames and the first index, the first and second indexes en- abling trick play functionality. Id. at col. 51 ll. 31–45. II In its petition for inter partes review, Netflix asserted that claims 1, 5, 8, 9, 13–15, 18, and 21–23 of the ’792 pa- tent would have been obvious in view of Zetts 1 as modified by Kaku. 2 The issue on appeal is whether the secondary reference, Kaku, is analogous art to the ’792 patent. In a section titled “Field of the Invention,” Kaku’s specification states that the invention “relates to motion image apparat- uses and, more particularly, to a motion image reproducing apparatus which is applicable to a digital camera for repro- ducing motion image data recorded on a recording me- dium.” Kaku col. 1 ll. 6–10. Kaku’s abstract discloses that, in a reproduce mode, “image data accommodated in a de- sired AVI file of a memory card is read out frame by frame in an intermittent fashion.” Kaku’s summary of the inven- tion refers to handling and manipulating data files contain- ing frames of image data and sound. Id. at col. 1 ll. 35–39, col. 2 ll. 28–31. In the detailed description of the preferred embodiments, Kaku discloses using an index chunk in the AVI file header to show image data and/or play sound data.

1 U.S. Patent No. 7,212,726. 2 U.S. Patent No. 6,671,408. Case: 22-1138 Document: 41 Page: 5 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 5

Id. at col. 5 ll. 22–30; see also id. Fig. 2. It also discloses compressing and decompressing image data to display im- ages to reproduce motion images. Id. at col. 2 ll. 19–24. Although Kaku describes the “primary object” of its inven- tion as focusing on reproducing a motion image as it relates to memory size of the apparatus (i.e., a problem prevalent with digital cameras), id. at col. 1 ll. 30–34, Kaku also clar- ifies that “the invention is applicable to every electronic ap- pliance to reproduce motion images.” Id. at col. 11 ll. 58– 61. In its petition, Netflix argued that Zetts disclosed a sys- tem and method of inserting an abridged index (“GOP off- set table”) at the beginning of a multimedia file to facilitate trick play, J.A. 6058–61, and that Kaku disclosed “using the AVI file format to store video/audio data, with an ‘index chunk,’ written to the end of the AVI file having the begin- ning addresses of each frame,” J.A. 6026. Netflix con- tended that a person of ordinary skill in the art would have been motivated to apply Zetts’s teachings of inserting a GOP offset table at the beginning of a multimedia file to facilitate trick play functionality in an AVI file as taught by Kaku because “the GOP offset table is compact” and al- lows users to jump to desired locations without being forced to view the entire video. J.A. 6032. DivX argued in its pre- liminary response that Kaku was cumulative art already considered by the examiner. J.A. 6122–24. The Board in- stituted the inter partes review. In its patent owner response, DivX raised the argu- ment that Kaku is non-analogous art. J.A. 6291–315. DivX relied on testimony from its expert, Dr. Chandrajit Bajaj, that: [A person of skill in the art] would understand that the field of endeavor of the ’792 Patent relates to facilitating trick play functionality in multimedia content that is streamed or downloaded over the in- ternet. There is, however, no indication that Kaku Case: 22-1138 Document: 41 Page: 6 Filed: 09/11/2023

has anything to do with enabling trick play func- tionality, streamed content, or dual indexes. Nota- bly, as discussed above, Kaku utilizes M-JPEG files in limited memory cameras and does not even gen- erally relate to streaming or the types of sophisti- cated multimedia files addressed by the ’792 Patent. Thus, Kaku and the ’792 patent have dis- tinct fields of endeavor. J.A. 4042 ¶ 45. DivX also argued that Kaku is not reason- ably pertinent to the problem the inventor of the ’792 pa- tent sought to address. See J.A. 6302–05 (relying on its expert’s testimony that the problem addressed by the ’792 patent was “facilitating trick play functionality in stream- ing services”). Netflix responded in its reply brief that Kaku was, in fact, analogous art: It is undisputed that Kaku teaches the AVI file for- mat . . . Under Federal Circuit precedent, Kaku must be considered for its AVI teachings, which cannot be ignored even if Kaku were primarily di- rected to camera embodiments. Moreover, Kaku is not so limited. Kaku teaches that, “although the embodiments were explained using a digital cam- era, it is needless to say that the invention is appli- cable to every electronic appliance to reproduce motion images.” Kaku includes embodiments di- rected to particular implementations of the AVI file format, e.g., the “data file.” . . . Here, the ’792 patent refers to AVI as prior art. Therefore, Kaku is in the same field of endeavor. “The Supreme Court’s decision in KSR . . . directs us to construe the scope of analogous art broadly.” Here, the ’792 patent broadly defines its scope as including “encoding . . . and decoding of multime- dia files.” Kaku teaches encoding (e.g., Case: 22-1138 Document: 41 Page: 7 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 7

“compressed” image data) and decoding (e.g., “de- compression”) of image data in AVI files. Under Federal Circuit precedent, Kaku is reasonably per- tinent. J.A. 6403–05 (citations omitted). DivX argued in its sur-reply that Netflix failed to meet its burden to demonstrate that Kaku was analogous art to the ’792 patent because Netflix did not explicitly identify the field of endeavor or the reasonably pertinent problems for either the ’792 patent or Kaku. J.A. 6471–76. The Board agreed with DivX. It rejected Netflix’s obvi- ousness argument because it found Netflix had not met its burden of showing that Kaku is analogous art to the ’792 patent under either the field-of-endeavor test or the reasonable-pertinence test. J.A. 33–34. With respect to the field of endeavor, the Board held that because Netflix “failed to identify the field of endeavor of either the ’792 patent or Kaku, [Netflix] cannot demon- strate that Kaku and the claimed invention are in the same field, and therefore fail[ed] to meet its burden of establish- ing analogous art under the field of endeavor test.” J.A. 21; see also J.A. 20 (finding that Netflix’s reply “failed to mean- ingfully respond to [DivX]’s arguments by identifying the field of endeavor of the ’792 patent and explaining why Kaku belonged in that field”). In addition, the Board found the use of AVI files alone too narrow to define the field of endeavor for either the ’792 patent or Kaku. J.A. 21. The Board noted that Netflix appeared to present multiple dif- ferent positions on what it viewed as the field of endeavor throughout the case—i.e., “file formats, AVI” or “encoding and decoding multimedia files.” J.A. 23–24. In the Board’s view, this supported its determination that Netflix failed to clearly identify a field of endeavor and, thus, failed to meet its burden. J.A. 23–24. Case: 22-1138 Document: 41 Page: 8 Filed: 09/11/2023

With respect to whether Kaku was reasonably perti- nent to the problem addressed by the ’792 patent, the Board found that Netflix failed to identify the problem ad- dressed by the ’792 patent or Kaku. J.A. 26. Netflix had argued that Kaku was reasonably pertinent because the ’792 patent’s scope broadly includes encoding and decoding multimedia files. But the Board faulted Netflix for merely pointing to encoding and decoding—something “well- known in the art”—instead of addressing “the problem that the ’792 patent seeks to solve.” J.A. 26–27; see also J.A. 26 (explaining that “the problems to which both [the inven- tion-in-dispute and the prior art] relate must be identified and compared” (quoting Donner Tech., LLC v. Pro Stage Gear, LLC, 979 F.3d 1353, 1359 (Fed. Cir. 2020))). The Board instead agreed with DivX and its expert “that the problem that the ’792 patent seeks to solve is facilitating trick play functionality in streaming media.” J.A. 27. Be- cause, in the Board’s view, Kaku was not reasonably perti- nent to that problem, the Board concluded that Kaku was not analogous art under the reasonably pertinent prong. Netflix appeals. We have jurisdiction under 28 U.S.C. § 1295(a). DISCUSSION We use “[t]wo separate tests [to] define the scope of analogous art: ‘(1) whether the art is from the same field of endeavor, regardless of the problem addressed and, (2) if the reference is not within the field of the inventor’s en- deavor, whether the reference still is reasonably pertinent to the particular problem with which the inventor is in- volved.’” Airbus S.A.S. v. Firepass Corp., 941 F.3d 1374, 1379 (Fed. Cir. 2019) (quoting In re Bigio, 381 F.3d 1320, 1325 (Fed. Cir. 2004)). The Board found that Kaku does not qualify as analogous art under either test. For the rea- sons that follow, we hold that the Board abused its discre- tion in requiring Netflix to explicitly identify the field of endeavor using specific language, but that the Board’s Case: 22-1138 Document: 41 Page: 9 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 9

reasonable-pertinence determination is supported by sub- stantial evidence. I We begin by addressing the Board’s dispositive conclu- sion that Netflix failed to identify the field of endeavor for Kaku and the patent-in-suit. We review the Board’s proce- dural or administrative decisions for an abuse of discretion. See 5 U.S.C. § 706(2)(A), (E); Game & Tech. Co. v. Wargam- ing Grp. Ltd., 942 F.3d 1343, 1348–49 (Fed. Cir. 2019). An abuse of discretion occurs if a decision “(1) is clearly unrea- sonable, arbitrary, or fanciful; (2) is based on an erroneous conclusion of law; (3) rests on clearly erroneous fact find- ing; or (4) involves a record that contains no evidence on which the Board could rationally base its decision.” Intel- ligent Bio-Sys., Inc. v. Illumina Cambridge Ltd., 821 F.3d 1359, 1367 (Fed. Cir. 2016) (quoting Bilstad v. Wakalopu- los, 386 F.3d 1116, 1121 (Fed. Cir. 2004)). Whether a prior art reference qualifies as analogous prior art is a question of fact that we review for substantial evidence. Bigio, 381 F.3d at 1324. We do not charge a per- son of ordinary skill in the art to know all arts; rather, we presume he knows the teachings of “all the prior art in the field of his endeavor” at the time the invention was made. Airbus S.A.S., 941 F.3d at 1380 (quoting In re Wood, 599 F.2d 1032, 1036 (CCPA 1979)). Thus, for an obvious- ness determination, a reference may only qualify as prior art that a person of ordinary skill in the art would look to if it is “analogous to the claimed invention.” Bigio, 381 F.3d at 1325; see also Manual of Patent Examining Procedure (MPEP) § 2141.01(a). We analyze whether prior art is analogous with “the foresight of a person of ordinary skill, not with the hind- sight of the inventor’s successful achievement.” See Sci. Plastic Prod., Inc. v. Biotage AB, 766 F.3d 1355, 1359 (Fed. Cir. 2014); see also KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 421 (2007) (“A factfinder should be aware, of Case: 22-1138 Document: 41 Page: 10 Filed: 09/11/2023

course, of the distortion caused by hindsight bias . . . .”). This inquiry “is meant to defend against hindsight.” See In re Kahn, 441 F.3d 977, 987 (Fed. Cir. 2006); In re Oetiker, 977 F.2d 1443, 1447 (Fed. Cir. 1992) (“The combination of elements from non-analogous sources, in a manner that re- constructs the applicant’s invention only with the benefit of hindsight, is insufficient to present a prima facie case of obviousness.”). We determine the field of endeavor “by reference to ex- planations of the invention’s subject matter in the patent application, including the embodiments, function, and structure of the claimed invention.” Bigio, 381 F.3d at 1325; see In re Deminski, 796 F.2d 436, 442 (Fed. Cir. 1986) (affirming that the references were within the field of endeavor of the invention where the references shared the same “function and structure”); MPEP § 2141.01(a) (considering “similarities and differences in structure and function” when assessing analogous art). The field of en- deavor is “not limited to the specific point of novelty, the narrowest possible conception of the field, or the particular focus within a given field.” Unwired Planet, LLC v. Google Inc., 841 F.3d 995, 1001 (Fed. Cir. 2016); see also Jeffrey T. Burgess, The Analogous Art Test, 7 BUFF. INTELL. PROP. L.J. 63, 72 (2009) (explaining that field of endeavor may be broadly defined because it relies on “the complete disclo- sure of all embodiments in the patent’s specification”); Lance Leonard Barry, Cézanne and Renoir: Analogous Art in Patent Law, 13 TEX. INTELL. PROP. L.J. 243, 244 (2005) (stating analogous art test merely gauges whether prior art is “too remote” to be treated as analogous art); id. at 247 (“[C]ourts generally take an expansive view of what consti- tutes analogous art.”). Importantly, “the scope of any field of endeavor will vary with the factual description of each invention.” Bigio, 381 F.3d at 1326. Unlike the reasona- ble-pertinence test, the field-of-endeavor test does not look to the problem that the patent purports to address. In re Clay, 966 F.2d 656, 658–59 (Fed. Cir. 1992). It is enough Case: 22-1138 Document: 41 Page: 11 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 11

that the prior art reference falls within the relevant field of endeavor of the patent-in-suit. We have affirmed findings of analogous art where the references shared a general field of endeavor. See, e.g., Un- wired Planet, 841 F.3d at 1001–02 (affirming Board’s find- ing of analogous art where the patent-in-dispute and prior art were both “in the field of interface design, with [the prior art] focusing on graphical user interfaces and the [pa- tent-in-dispute] focusing on interfaces for location-based services,” because “[t]hese two areas of focus overlap within the broader field of interface design because the teachings in graphical user interface design . . . have relevance in in- terfaces for location-based applications”); In re Mettke, 570 F.3d 1356, 1359 (Fed. Cir. 2009) (affirming Board’s identification of the field of endeavor as “pay-for-use pub- lication communication terminations” where the specifica- tion described various communication media, not just “providing access to the Internet” as directed by the claim- in-dispute). Here, Netflix argues that the Board erroneously re- quired it to specifically state both Kaku’s and the ’792 pa- tent’s field of endeavor, using the exact words “field of endeavor,” to meet its burden in response to DivX’s argu- ment concerning analogous art. Appellant’s Br. 37–38; see Oral Arg. at 0:47–0:57, https://oralarguments.cafc. uscourts.gov/default.aspx?fl=22-1138_03072023.mp3. We agree. The Board abused its discretion in determining that Netflix failed to articulate a field of endeavor. Netflix iden- tified two alternative theories for what it viewed as Kaku’s and the patent-in-suit’s overlapping fields of endeavor— (1) AVI files; or (2) encoding and decoding multimedia files. Although Netflix’s reply brief before the Board did not for- mulaically articulate a field of endeavor using those exact words, our precedent does not require the use of magic words. The Board erred by imposing a higher burden than Case: 22-1138 Document: 41 Page: 12 Filed: 09/11/2023

that required by our precedent. The Board acknowledged that Netflix argued “that ‘Kaku must be considered for its AVI teachings,’ and that ‘Kaku includes embodiments di- rected to particular implementations of the AVI file for- mat.’” J.A. 20 (quoting J.A. 6404 (Netflix reply)). We fail to see how this was not enough for the Board to understand Netflix’s position that Kaku’s field of endeavor is AVI files. Addressing the ’792 patent, Netflix argued that the inven- tion “refers to AVI as prior art” and cited sections of the ’792 patent that discussed the AVI file and how the “chunks” of the invention’s multimedia file “are defined as part of the AVI file format.” J.A. 6404–05 (citing ’792 pa- tent col. 5 ll. 33–38, col. 5 l. 51–col. 6 l. 17, col. 22 ll. 20– 22). Taken together and in context, Netflix sufficiently ar- gued that the field of endeavor for both the ’792 patent and Kaku is AVI file formats. Furthermore, Netflix alternatively argued in its reply brief before the Board that the ’792 patent’s broad scope includes “encoding . . . and decoding of multimedia files,” J.A. 6405 (citing ’792 patent col. 1 ll. 20–21); see also Oral Arg. at 8:50–9:09, and that Kaku teaches “encoding . . . and decoding” for image data in AVI files through compression and decompression, J.A. 6405 (citing Kaku col. 2 ll. 19–24). Although the Board characterizes this argument as exclu- sive to the issue of reasonable pertinence, J.A. 23, we disa- gree with the Board’s unduly rigid view of the analogous art framework. As we have recognized, the evidence and analysis relating to the field of endeavor and reasonably pertinent prongs may overlap. See Unwired Planet, 841 F.3d at 1001–02 (finding that the patent-at-issue and the asserted prior art were in the same field of endeavor and “reasonably pertinent to the problem of displaying ad- dress information,” based on the same disclosures concern- ing graphical user interface design based on geography and the supporting expert testimony). Contrary to the Board’s unduly rigid requirement that a petitioner explicitly iden- tify a field of endeavor, there are instances—like in Case: 22-1138 Document: 41 Page: 13 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 13

Netflix’s reply in this case—where general language is suf- ficient to allow the Board to consider alternative argu- ments on the merits. In other words, even where a petitioner does not explicitly define a field of endeavor, its briefing may nonetheless present an argument on that is- sue when taken as a whole. Our conclusion that the Board erred is bolstered by the Board’s own analysis. After holding that Netflix failed to identify a field of endeavor and thus failed to meet its bur- den, the Board purported to analyze the field-of-endeavor question without clearly articulating what it viewed as the field of endeavor. The Board acknowledged that DivX “re- lies on Dr. Bajaj’s testimony on the field of endeavor of the ’792 patent and of Kaku” and found “Dr. Bajaj’s testimony supported by the cited disclosures.” J.A. 25. As noted above, Dr. Bajaj testified that “the field of endeavor of the ’792 Patent relates to facilitating trick play functionality in multimedia content that is streamed or downloaded over the internet.” J.A. 4042 ¶ 45. In the very next sentence, however, the Board suggested a different, broader field of endeavor, noting that “[t]he ’792 patent’s Background of the Invention makes clear that the patent relates to encod- ing, transmission, and decoding of multimedia files.” J.A. 25. Contrary to the strict requirement it imposed on Netflix, the Board at no point in its analysis used specific language to articulate or explicitly identify the field of en- deavor—e.g., “the field of endeavor is trick play functional- ity.” Moreover, like Netflix, the Board identified two possibilities for the ’792 patent’s field of endeavor. See J.A. 25 (“The ’792 patent claims are each directed to using multiple indexes and enabling trick play functionality in multimedia files.”). Given the Board’s own articulation of two potential fields of endeavor without the use of magic Case: 22-1138 Document: 41 Page: 14 Filed: 09/11/2023

words, 3 we have difficulty understanding how it could view Netflix’s reply brief as insufficient for failure to affirma- tively identify a single field of endeavor. See J.A. 24. We thus remand to allow the Board to decide this fac- tual question of whether Kaku is directed to the same field of endeavor as the patent-in-suit based on the arguments fairly presented by the parties, including Netflix’s argu- ments that Kaku and the ’792 patent are both directed to AVI files and/or that they are both directed to the encoding and decoding of multimedia files. In so holding, we note the unusual circumstances of this case. We understand it is not our role to reweigh evi- dence; nor is it the Board’s role to make arguments on be- half of the parties. But in this unique case, where the Board found Netflix’s reply brief so deficient as to not pre- sent any argument regarding the field of endeavor, we are reluctant to affirm the Board’s factual finding, which rests on a failure to identify a field of endeavor rather than a clear analysis of why Kaku is not, in fact, directed to the same field of endeavor. In these circumstances, we find that the Board abused its discretion and remand for the

3 We also note that one of the potential fields of en- deavor identified by the Board—“encoding, transmission, and decoding of multimedia files”—mimics the argument presented by Netflix when it stated that “the ’792 patent broadly defines its scope as including ‘encoding . . . and de- coding of multimedia files.’” J.A. 6405; ’792 patent col. 1 ll. 20–21. On remand, the Board should consider whether both the ’792 patent and Kaku relate to this possible field of endeavor. Case: 22-1138 Document: 41 Page: 15 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 15

Board to fully consider the question of whether the ’792 pa- tent and Kaku share the same field of endeavor. 4 II We next consider whether the Board’s application of the reasonably pertinent test is supported by substantial evidence. See Bigio, 381 F.3d at 1325. “[W]e only presume knowledge from those arts reasonably pertinent to the par- ticular problem with which the inventor was involved.” Airbus S.A.S., 941 F.3d at 1381 (quoting Wood, 599 F.2d at 1036). Prior art that is outside an inventor’s field of en- deavor is reasonably pertinent “only if its subject matter logically would have commended itself to an inventor’s at- tention in considering his problem.” Id. at 1382 (quota- tions omitted). Put another way, a prior art reference is reasonably pertinent only if “a person of ordinary skill would reasonably have consulted those references and ap- plied their teachings in seeking a solution to the problem that the inventor was attempting to solve.” In re GPAC Inc., 57 F.3d 1573, 1578 (Fed. Cir. 1995). “In order to de- termine whether a reference is ‘reasonably pertinent,’ . . . a reasonable factfinder should consider record evidence cited by the parties to demonstrate the knowledge and per- spective of a person of ordinary skill in the art at the time of the invention.” Airbus S.A.S., 941 F.3d at 1383. Netflix challenges the Board’s determination that Kaku is not reasonably pertinent to the problem addressed by the ’792 patent.

4 Netflix also argues the merits of the Board’s field- of-endeavor analysis. Appellant’s Br. 38–48. Because the Board’s analysis of the merits is intertwined with its deter- mination that Netflix waived this argument by failing to identify a field of endeavor, we remand for the Board to consider this issue anew. Case: 22-1138 Document: 41 Page: 16 Filed: 09/11/2023

We hold that substantial evidence supports the Board’s finding. For the ’792 patent, the Board considered the spec- ification, claims, and prosecution history in determining that the ’792 patent related to the problem of facilitating trick play in streaming multimedia. To support its finding, the Board looked to the specification’s discussion of “dis- playing a multimedia presentation contained within a mul- timedia file non-sequentially,” J.A. 27 (quoting ’792 patent col. 48 ll. 22–25); “skipping in an irregular fashion between different portions,” id. (quoting ’792 patent col. 48 ll. 25– 27); and using “file formats for multimedia information [such as RIFF and AVI] to enable standardized generation, distribution and display of multimedia information,” J.A. 27–28 (quoting ’792 patent col. 1 ll. 25–28, col. 5 ll. 33– 56). The Board noted that every claim recited “the first and second indexes enabling trick play functionality.” J.A. 28 (quoting ’792 patent col. 51 l. 31–col. 54 l. 21). The Board also credited testimony from DivX’s expert Dr. Bajaj, who opined that the ’792 patent dealt with facilitating trick play in streamed or downloaded multimedia. J.A. 28–29; J.A. 4045 ¶¶ 49–50. Regarding Kaku, the Board found that it addressed a different problem than the ’792 patent: the problem of im- age compression “to accommodate as lengthy a recording as possible in a camera’s internal memory.” J.A. 29. To reach this conclusion, the Board relied primarily on Kaku’s specification. Specifically, the Board cited extensively to the Background of the Invention and Summary of the In- vention sections of Kaku, which refer to the problem of lim- ited “memory capacity” in digital cameras and the desire to “reproduc[e] motion images for a long time period regard- less of a memory capacity.” J.A. 29–30 (citing Kaku col. 1 ll. 8–12, col. 1 ll. 14–33, col. 1 l. 65–col. 2 l. 4, col. 2 ll. 18– 27). The Board noted that the problems addressed by the ’792 patent—“trick play,” indexing to implement trick play, and streaming multimedia—do not appear in Kaku. J.A. 30. The Board again credited Dr. Bajaj’s testimony— Case: 22-1138 Document: 41 Page: 17 Filed: 09/11/2023

NETFLIX, INC. v. DIVX, LLC 17

that the ’792 patent and Kaku are directed to different problems—finding his testimony “consistent with the teaching[s] of the ’792 patent and Kaku.” See J.A. 32 (cit- ing J.A. 4044–48 ¶¶ 47–54). The Board concluded that a skilled artisan considering the ’792 patent’s problem of trick play would not have looked to Kaku, which addresses a different problem. See J.A. 30, 32. In light of the evi- dence supporting the Board’s finding, as well as the Board’s thorough consideration of the record, we cannot say that the Board’s finding is unreasonable. See Airbus S.A.S., 941 F.3d at 1383. Accordingly, our remand does not in- clude reconsideration of the reasonably pertinent test for determining analogous art. CONCLUSION For the foregoing reasons, we vacate the Board’s obvi- ousness determination as to claims 1, 5, 8, 9, 13–15, 18, and 21–23, and remand for further proceedings consistent with this opinion. We affirm the Board’s decision in all other respects. AFFIRMED-IN-PART, VACATED-IN-PART, AND REMANDED COSTS No costs.