Patent opinions from the Eastern District of Texas and the U.S. Court of Appeals for the Federal Circuit

Weekly Digest
Fed. Cir.

Bright Data Ltd. v. Code200, UAB, No. 23-2144 (Fed. Cir. Aug. 1, 2025)

Affirmed
Court
U.S. Court of Appeals for the Federal Circuit
Case No.
No. 23-2144
Decided
August 1, 2025
Judge
Stark, J.
Document
Nonprecedential Opinion
Length
13 pages

NOTE: This disposition is nonprecedential. United States Court of Appeals

for the Federal Circuit

BRIGHT DATA LTD.,

Appellant

v.

CODE200, UAB, TESO LT, UAB, METACLUSTER LT, UAB, OXYSALES, UAB, THE DATA COMPANY

TECHNOLOGIES INC., MAJOR DATA UAB,

CORETECH LT, UAB,

Appellees

2023-2144, 2023-2145, 2023-2146, 2023-2147, 2023-2414,

2023-2415, 2023-2442, 2023-2443

Appeals from the United States Patent and Trademark Office, Patent Trial and Appeal Board in Nos. IPR2021-01492, IPR2021-01493, IPR2022-00103, IPR2022-00135, IPR2022-00138, IPR2022-00353, IPR2022-00861, IPR2022-00862, IPR2022-00915, IPR2022-00916.

Decided: August 1, 2025

ROBERT M. HARKINS, JR., Cherian LLP, Berkeley, CA, argued for appellant. Also represented by KORULA T. CHERIAN; THOMAS M. DUNHAM, RONALD WIELKOPOLSKI,

2Washington, DC.

DANIEL LEVENTHAL, Norton Rose Fulbright US LLP, Houston, TX, argued for all appellees. Appellees Code200, UAB, Teso LT, UAB, Metacluster LT, UAB, Oxysales, UAB, coretech lt, UAB also represented by STEPHANIE DEBROW, MARK T. GARRETT, Austin, TX; JONATHAN S. FRANKLIN, Washington, DC.

MICHAEL N. RADER, Wolf Greenfield & Sacks, PC, New York, NY, for appellee The Data Company Technologies Inc. Also represented by ADAM R. WICHMAN, Boston, MA. JASON R. BARTLETT, Maschoff Brennan, San Francisco, CA, for appellee Major Data UAB. Also represented by WENSHENG MA.

Before HUGHES, CUNNINGHAM, and STARK, Circuit Judges.

Patent Owner Bright Data Ltd. (“Bright Data”) appeals the decision of the Patent Trial and Appeal Board (“Board”) in ten inter partes reviews (“IPRs”), finding the challenged claims of four of its patents unpatentable.1 Bright Data argues that the Board erred in its constructions of two claim terms, its reading of prior art references, and its findings regarding secondary considerations of non-obviousness. We disagree and affirm the Board.

3I

The patents at issue – U.S. Patent Nos. 11,044,342; 10,257,319; 10,484,510; and 11,044,344 – are part of the same family. They share a common specification.

Each of the four patents is directed to a “system designed for increasing network communication speed for users,” by “releasing congestion from the Web by fetching [user requested] information from multiple sources, and relieving traffic from Web servers by offloading the data transfers from them to nearby peers.” J.A. 1137 (’342 pat. abstract).2 The claimed system includes “multiple communication devices,” each of which may, at various times, “serve as a client, peer, or agent, depending upon requirements of the network.” J.A. 1160 (’342 pat. col. 4:44-50); see also J.A. 1163 (’342 pat. col. 9:20-25 (“separate [software] modules that run in parallel” are activated depending on “specific role that the communication device 200 is partaking in . . . at a given time”). The patents purport to reduce “infrastructure costs” faced by internet service providers by eliminating the need for proxy servers “at every point around the world where the Internet is being consumed.” J.A. 1159 (’342 pat. cols. 1:50-53, 2:26-29).

Figure 3, reproduced below, shows an embodiment in which several communication devices loaded with software switch functions, with each device serving at times as a client, peer, or agent. J.A. 1160-61 (’342 pat. cols. 4:54-5:48). When a communication device is designated as a client device, it requests information from the internet through a web browser. J.A. 1161, 1163 (’342 pat. cols. 5:21-25, 9:27-436. An agent recognizes that the identical information has been accessed recently from other communication devices, which each have portions of the requested data. J.A. 1163 (’342 pat. col. 9:44-50) (describing job of “agent[] [as] obtain[ing] a list of peers within the communication network 100 that contain requested information”). Those communication devices then are designated to work as peers and supply their respective portions of information to the agent. J.A. 1161 (’342 pat. col. 5:37-43).

Figure on page 4 of the opinion

J.A. 1146.

Claim 1 of the ’342 patent is representative and recites: A method for use with a web server that responds to Hypertext Transfer Protocol (HTTP) requests and stores a first content that is identified by a first Uniform Resource Locator (URL), the method by a first client device comprising:

[a] executing, by the client device, a web browser application or an email application;
[b] establishing a Transmission Control Protocol (TCP) connection with a second server;
5[c] receiving, the first content from the web server over an Internet; and
[d] sending the received first content, to the second server over the established TCP connection, in response to the receiving of the first URL.

J.A. 1168 (emphasis added to show disputed limitations).

Several entities, including Code200 (“Petitioner”), petitioned for, and the Board instituted, IPRs. E.g., J.A. 1-150. As part of its proceedings, the Board construed two disputed claim terms: “client device” and “second server.” The parties’ fundamental dispute with respect to both terms was whether they should be construed based on their function, as Petitioner contended was their plain and ordinary meaning, or if they should instead be more narrowly construed to require each have a different structure, as Bright Data preferred. J.A. 10-40. The Board rejected Bright Data’s proposal and construed “client device” to mean “a ‘communication device that is operating in the role of a client’” and “second server” to mean “a ‘server that is not the client device.’” J.A. 38-39.3

6Here, after resolving the parties’ claim construction disputes, the Board analyzed Petitioner’s prior art. Pertinent to this appeal is an article by Michael K. Reiter entitled “Crowds: Anonymity for Web Transactions,” ACM TRANSACTIONS ON INFORMATION AND SYSTEM SECURITY, Vol. 1, No. 1, November 1998, at 66–92 (“Crowds”). J.A. 2391-417. Crowds proposes a system for web transactions allowing a user to join a crowd of users, “that collectively issues requests on behalf of its members,” to protect users’ anonymity. J.A. 2391. The Board found that Crowds both anticipated and rendered obvious the claims at issue on appeal.4 In reaching its conclusion as to obviousness, the Board considered Bright Data’s evidence of secondary considerations of non-obviousness. It found that Bright Data’s evidence lacked probative value because Bright Data failed to demonstrate the requisite nexus between its evidence and its claims.

Bright Data timely appealed. We have jurisdiction under 28 U.S.C. § 1295(a)(4)(A).

II

“The Board’s claim constructions . . . are determinations of law reviewed de novo where based on intrinsic evidence, with any Board findings about facts extrinsic to the patent record reviewed for substantial-evidence support.” St. Jude Med., LLC v. Snyders Heart Valve LLC, 977 F.3d 1232, 1238 (Fed. Cir. 2020). “Anticipation is a question of fact subject to substantial evidence review.” IOENGINE, LLC v. Ingenico Inc., 100 F.4th 1395, 1402 (Fed. Cir. 2024) (internal quotation marks omitted). “Substantial evidence7 is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Id. (internal quotation marks omitted). “[T]he possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency’s finding from being supported by substantial evidence.” Consolo v. Fed. Mar. Comm’n, 383 U.S. 607, 620 (1966). We “defer to the Board’s findings concerning the credibility of expert witnesses.” In-cept LLC v. Palette Life Scis., 77 F.4th 1366, 1377 (Fed. Cir. 2023).

“The ultimate question of obviousness is a legal question that we review de novo with underlying factual findings that we review for substantial evidence.” Roku, Inc. v. Universal Elecs., Inc., 63 F.4th 1319, 1324 (Fed. Cir. 2023). “What the prior art discloses and whether a person of ordinary skill would have been motivated to combine prior art references are both fact questions that we review for substantial evidence.” Intel Corp. v. PACT XPP Schweiz AG, 61 F.4th 1373, 1378 (Fed. Cir. 2023). Findings relating to secondary considerations (i.e., objective indicia) of non-obviousness are reviewed for substantial evidence. See Finisar Corp. v. DirecTV Grp., Inc., 523 F.3d 1323, 1338-39 (Fed. Cir. 2008) (citing Graham v. John Deere Co., 383 U.S. 1, 17-18 (1966)).

III

Bright Data challenges the Board’s claim constructions, its understanding of Crowds, and its findings regarding secondary considerations of non-obviousness. We address each issue in turn.

A

We agree with the Board’s constructions of “client device” and “second server.” As the Board explained, a person of ordinary skill in the art would understand that these terms, as used in Bright Data’s patents, refer to the role a device is playing, and not to its particular structure (i.e.,8 hardware). Thus, the Board properly construed “client device” as a “communication device that is operating in the role of a client.” J.A. 38. It properly construed “second server” as a “device that is operating in the role of a server” and is “not the client device.” J.A. 38-39.

The Board rejected Bright Data’s contention that the disputed terms should be construed as hardware-based, instead of role-based. The specification explains that a communication device has software that allows it to toggle between roles as a client, peer, or agent. See, e.g., J.A. 1160 (’342 pat. col. 4:41-50) (“software stored within each communication device” allows each device to serve in different roles “depending upon the requirements of the network”). The specification also expressly notes: “The acceleration application 220 also contains three separate modules that run in parallel, namely, a client module 224, a peer module 226, and an agent module 228, each of which comes into play according to the specific role that the communication device 200 is partaking in the communication network 100 at a given time.” J.A. 1163 (’342 pat. col. 9:20-25) (emphasis added). These statements provide strong support for a role-based construction as opposed to a hardware-based construction.

The Board was not persuaded by Bright Data’s argument that role-based constructions would be unworkable because they would somehow imply that the “first client device” must operate as both a client and server. See J.A. 31-32. As the Board correctly explained, Bright Data’s argument mistakenly presumes that the Board’s constructions require that a device “act exclusively in only one role . . . at all times,” which they do not. J.A. 31. We agree with the Board on these points.

Bright Data’s far narrower proposed constructions – which would limit “client device” to a “consumer computer” and “second server” to a commercial server – are not correct. The only reference to “client device” that refers to a9 “consumer computer” appears in a background example, which even Bright Data agrees discusses a prior art system and not the claimed system. See J.A. 1159 (’342 pat. col. 2:43-46). All subsequent references to “client device” throughout the specification use the term based on its role.

Bright Data argues that its construction is nonetheless correct because the patentee acted as its own lexicographer. We disagree.

As the Board concluded, the record here does not meet the high standard for lexicography. Lexicography involves “clearly set[ting] forth a definition of the disputed claim term” and “clearly express[ing] an intent to redefine the term.” Thorner v. Sony Comput. Ent. Am., LLC, 669 F.3d 1362, 1365 (Fed. Cir. 2012) (internal quotation marks omitted). The only support Bright Data cites for its purported lexicography defines not the “client devices” of its claims but, instead, the “client devices” of the prior art.

In the background portion of the specification, the patent states: “Fig. 2 is a schematic diagram providing an example of a peer-to-peer file transfer network 50. In the network 50, files are stored on computers of consumers, referred to herein as client devices 60.” J.A. 1159 (’342 pat. col. 2:43-46). While it is true that terms like “referred to herein as” can be used to signal lexicography, see, e.g., Kyocera Senco Indus. Tools Inc. v. Int’l Trade Comm’n, 22 F.4th 1369, 1378 (Fed. Cir. 2022); Abbott Lab’ys v. Andrx Pharms., Inc., 473 F.3d 1196, 1210 (Fed. Cir. 2007), they do not always do so. It is necessary to examine precisely how the term is used and its place in the full context of the specification and in relation to the claims. Here, the language on which Bright Data relies provides a definition of, if anything, “client devices 60,” not “client devices” generally and throughout the patent, including in the claims. The claims are not drawn to “client devices 60” but, instead, to “client devices,” and this claim term is not defined in the patent. The passing reference in the background10 section of the patent to “client devices 60” would not be read by a skilled artisan as carrying through to the claims but, rather, as limited to describing the prior art. See J.A. 33 (Board noting Bright Data’s purported lexicographic declaration was used only in reference to describing prior art peer-to-peer file sharing system, not in reference to claimed invention). There is simply no “clear” definition or expression of intent to use “client devices” in the claims in the manner Bright Data argues.

Bright Data also directs us to the prosecution history, but it is no more helpful to it. Bright Data relies principally on statements made by the applicant to an examiner in prosecution of a great-grandfather patent to the ’342. There, the applicant described the prior art as follows:

Client devices, such as client 105 in the [prior art] Garcia reference, are end-units that request information from servers, use client-related software such as Web browser software, communicate over the Internet using ISP connection, and are typically consumer owned and operated . . . . In contrast, server devices are known in the art to be dedicated devices to store information objects, to be provided to clients upon request.

J.A. 3410.

In this statement, the applicant was distinguishing its invention from a prior art reference known as Garcia. In context, a person of ordinary skill in the art would understand the applicant to have been distinguishing its client devices and servers from Garcia’s, not defining devices and servers more generally. Specifically, the applicant explained that while a client device in Garcia was “typically consumer owned and operated,” this was not the case with the claimed invention’s client devices. See id. (Applicant arguing “[t]he Garcia reference is silent, and [e]ffectively teaches away, from using clients as content source for other clients”).

11Thus, again, we agree with the Board’s constructions and reject Bright Data’s arguments to the contrary.

B

Applying its constructions of the disputed terms, the Board held that Crowds either anticipates the appealed claims or, in combination with other references or alone, renders them obvious. Substantial evidence supports the Board’s findings and, therefore, we agree that Petitioner proved the challenged claims are obvious.

Crowds teaches a system for web transactions allowing a user to join a crowd of users, to anonymize a search request. In Crowds, a request from a user is passed on to at least one random member of the crowd before being submitted to an end server. In this way, the request is anonymized. The anonymous users (shown as circled numbers in Figure 2 from Crowds, reproduced below) are known as “jondo’s” (short for “John Doe’s”). J.A. 2398-99.

Figure on page 11 of the opinion

The Board found that the first jondo will send a request to a second jondo to anonymize it; in the example above, jondo 5 sends a request to jondo 4. J.A. 44 (relying on annotated figure from Petitioner). Then the second jondo, jondo 4, will act as a second server passing along the12 request to a client device, jondo 6, which passes it to a web server (shown as boxes in Figure 2). See J.A. 42. Jondo 6, the Board found, “is operating in the role of a client executing a web browser application because jondo 6 serves as a client of web server 5 when web requests originating at jondo 5 are sent by jondo 6 to web server 5 in the mapped path.” J.A. 46. Thus, because jondo 6 acts as a client, Crowds teaches the claimed client device of the appealed claims. In reaching this conclusion, the Board relied on Crowds’ disclosure as well as the declaration of Petitioner’s expert, as it is free to do. J.A. 45 (citing Exhibits 1003 and 1004).

The Board also relied on Crowds’ disclosure and Petitioner’s expert to find that Crowds teaches a jondo acting as a second server. As the Board found, a person having ordinary skill in the art would understand that, in Crowds, the client device (jondo 6) establishes a TCP connection with the second server (jondo 4). See J.A. 46, 50. Because jondo 4 sends the “received jondo 5’s web request on to jondo 6 and send[s] web server 5’s response back to jondo 5,” it “operates in the role of a server.” J.A. 48.

In addressing Bright Data’s arguments about the prior art, the Board recognized that “most of the[m] . . . are based on claim constructions that we have not adopted.” J.A. 49. Bright Data does the same on appeal and it fares no better. See, e.g., Open. Br. at 54 (arguing Crowds disclosure “indicates that servers are different from user computers”); id. at 58 (arguing another piece of prior art fails to teach appealed claims because its servers are “not different types of network components”). The Board’s analysis in its final written decisions was consistent with the correctly construed claim language and was supported by substantial evidence.

C

Substantial evidence also supports the Board’s finding that Bright Data failed to introduce evidence of secondary13 considerations of non-obviousness sufficient to overcome Petitioner’s showing of obviousness. The Board concluded that Bright Data’s evidence did not satisfy the nexus requirement. Its evidence of commercial success, for instance, reflects the marketplace success of “residential proxy service[s], residential consumer computers, and residential IP addresses,” which are not limitations of any of the claims. J.A. 62. Because the features of the products it relies on to show commercial success are not claimed, Bright Data “failed to establish a nexus between the challenged claims and the products” relied on. J.A. 61-63. The product neither embodies nor is coextensive with the challenged claims. J.A. 62. The Board similarly rejected Bright Data’s assertions of long-felt need, copying, and industry praise, again for lack of nexus. J.A. 64-67.

The Board’s findings are supported by the evidence it cited. See J.A. 58-68. The Board committed no error.

IV

We have considered Bright Data’s remaining arguments and find them unpersuasive. Accordingly, for the foregoing reasons, we affirm the judgment of the Board.

AFFIRMED

Footnotes

  1. 1 IPR2022-00861 and IPR2022-00862 were joined to IPR2021-01492 and IPR2021-01493, respectively, and then terminated. J.A. 39297, J.A. 46356.
  2. 2 We cite to the ’342 patent, J.A. 1137-69, and the Board’s final written decision in IPR2022-00103, which considered the ’342 patent, J.A. 1-75. The other patents and final written decisions are not materially different with respect to the issues presented in this appeal.
  3. 3 In doing so, the Board reached the same construction that a district court had reached when handling three suits involving Bright Data’s patents. See Bright Data v. Oxylabs f/k/a Teso LT, UAB et al., No. 2:19-cv-00395, ECF No. 191 (E.D. Tex. Dec. 7, 2020) (court disagreeing with Bright Data’s hardware-based distinctions, instead finding specification teaches use of structurally identical “communication devices” that can serve in different roles at different times); Bright Data v. Code200, UAB, et al., No. 2:19-cv-00396, ECF No. 97 (E.D. Tex. Feb. 8, 2021) (adhering to same constructions adopted in Oxylabs); Bright Data v. NetNut Ltd., No. 2:21-cv-225, ECF No. 146 (E.D. Tex. May 10, 2022) (same); see also J.A. 2944-48, 2971-73, 3009-11, 6588-99.
  4. 4 On appeal, Bright Data limits its arguments to the independent claims of the challenged patents, all of which the Board determined were unpatentable based on Crowds. See Open. Br. at 3, 53-66; J.A. 73; J.A. 234; J.A. 407–08; J.A. 567; J.A. 710–11; J.A. 798–99; J.A. 1042; J.A. 1134.

Text extracted from the official PDF and reflowed for reading. It is not a certified transcript — the PDF linked above is the authoritative document.

Analysis is coming soon.

A plain-English account of the issues, facts and holding, written by Patent Case Watch. The opinion itself — its full text, its PDF and everything else on this page — stays free to read for everyone, always.

Cite this opinion

Bright Data Ltd. v. Code200, UAB, No. 23-2144 (Fed. Cir. Aug. 1, 2025).

Record ID
CAFC-23-2144-20250801
Permalink
https://patentcasewatch.com/opinions/CAFC-23-2144-20250801

No reporter citation is shown because the official source does not supply one; none is inferred here. Check the opinion and a citator before relying on this form.