Equil IP Holdings LLC v. Akamai Technologies, Inc.

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE EQUIL IP HOLDINGS LLC Plaintiff, Civil Action No. 22-677-RGA Vv. AKAMAI TECHNOLOGIES, INC., Defendant.

AKAMAI TECHNOLOGIES, INC., Plaintiff, y Civil Action No. 22-1531-RGA AUTOMATED MEDIA PROCESSING SOLUTIONS, INC. D/B/A EQUILIBRIUM Defendant.

MEMORANDUM OPINION David E. Moore, Bindu A. Palapura, Andrew L. Brown, POTTER ANDERSON & CORROON LLP, Wilmington, DE; Jason R. Bartlett (argued), Jason A. Crotty, Marc J. Pernick, MASCHOFF BRENNAN GILMORE ISRAELSEN & MAURIEL LLP, San Francisco, CA; Steven Callahan, Christopher T. Bovenkamp, CHARHON CALLAHAN ROBSON & GARZA, PLLC; Dallas, TX Attorneys for Plaintiff Equil IP Holdings LLC and Defendant Automated Media Processing Solutions, Inc. d/b/a Equilibrium. Adam W. Poff, Robert M. Vrana, Alexis N. Stombaugh, YOUNG, CONAWAY, STARGATT & TAYLOR LLP, Wilmington, DE; James R. Batchelder, James L. Davis, Jr. (argued), Daniel W. Richards, ROPES & GRAY LLP, East Palo Alto, CA; Scott S. Taylor, ROPES & GRAY LLP, Boston, MA; Colin P. Dunn, ROPES & GRAY LLP, New York, NY Attorneys for Akamai Technologies, Inc. uly K, 2024

Bitch Gitte DGE: Before me is‘the issue of claim construction of multiple terms in U.S. Patent No. 8,543,667 (“the ’667 patent”) and U.S. Patent No. 9,158,745 (“the ’745 patent”). The parties submitted a Joint Claim Construction Brief and Appendix for each patent. (No. 22-1531, D.I. 43, 44; No. 22-677, D.I. 79, 80).! I heard oral argument on April 18, 2024. I. BACKGROUND On May 24, 2022, Equil IP Holdings LLC (“Equilibrium”) filed a complaint against Akamai Technologies, Inc. (““Akamai’’). Equilibrium alleged infringement of the ’745 patent as well as two other patents that have since been dismissed. (See D.I. 62). The ’745 patent is directed to “[o]ptimization of media content using generated intermediate media content.” (’745 patent, Title). Its priority date appears to be October 21, 1999. (‘745 patent at 1:8-17; see D.I. 79 at 1). The asserted claims are claims 1, 2, 5, 6, and 7. (D.I. 79 at 2). Claims 2, 5, 6, and 7 all depend from claim 1. On November 23, 2022, Akamai filed a complaint against Automated Media Processing Solutions, Inc, which does business as Equilibrium, alleging infringement of the ’667 patent. The patent “relates generally to computer-based methods and apparatuses, including computer program products, for policy-based content insertion.” (667 patent at 1:5-7). Its priority date appears to be its filing date, which is January 14, 2008. Akamai asserts three independent claims and numerous dependent claims. (D.I. 43 at 1).

' The parties filed supplemental materials at my request. (No. 22-1531, D.I. 47, 48; No. 22-677, D.I. 102, 110, 111). Unless otherwise indicated, references to Docket Items relating to the ‘667 patent refer to No. 22-1531 and relating to the ‘745 patent refer to No. 22-677. | refer to Equil as Equilibrium because the parties opposing Akamai in these two cases are, at least for present purposes, the same.

Il. LEGAL STANDARD “It is a bedrock principle of patent law that the claims of a patent define the invention to which the patentee is entitled the right to exclude.” Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed. Cir. 2005) (en banc) (cleaned up). ““[T]here is no magic formula or catechism for conducting claim construction.’ Instead, the court is free to attach the appropriate weight to appropriate sources ‘in light of the statutes and policies that inform patent law.’” SoftView LLC v. Apple Inc., 2013 WL 4758195, at *1 (D. Del. Sept. 4, 2013) (alteration in original) (quoting Phillips, 415 F.3d at 1324). When construing patent claims, a court considers the literal language of the claim, the patent specification, and the prosecution history. Markman v. Westview Instruments, Inc., 52 F.3d 967, 977-80 (Fed. Cir. 1995) (en banc), aff'd, 517 U.S. 370 (1996). Of these sources, “the specification is always highly relevant to the claim construction analysis. Usually, it is dispositive; it is the single best guide to the meaning of a disputed term.” Phillips, 415 F.3d at 1315 (cleaned up). “While claim terms are understood in light of the specification, a claim construction must not import limitations from the specification into the claims.” Deere & Co. v. Bush Hog, LLC, 703 F.3d 1349, 1354 (Fed. Cir. 2012) (citing Phillips, 415 F.3d at 1323). “(T]he words of a claim ‘are generally given their ordinary and customary meaning.’ . . . [It is] the meaning that the term would have to a person of ordinary skill in the art in question at the time of the invention, i.e., as of the effective filing date of the patent application.” Phillips, 415 F.3d at 1312-13 (citations omitted). “[T]he ‘ordinary meaning’ of a claim term is its meaning to [an] ordinary artisan after reading the entire patent.” Jd. at 1321. “In some cases, the ordinary meaning of claim language as understood by a person of skill in the art may be readily apparent even to lay judges, and claim construction in such cases involves little more than the application of the widely accepted meaning of commonly understood words.” Jd. at 1314.

When a court relies solely on the intrinsic evidence—the patent claims, the specification, and the prosecution history—the court’s construction is a determination of law. See Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318, 331 (2015). The court may also make factual findings based on consideration of extrinsic evidence, which “consists of all evidence external to the patent and prosecution history, including expert and inventor testimony, dictionaries, and learned treatises.” Phillips, 415 F.3d at 1317-19 (quoting Markman, 52 F.3d at 980). Extrinsic evidence may assist the court in understanding the underlying technology, the meaning of terms to one skilled in the art, and how the invention works. /d. Extrinsic evidence, however, is less reliable and less useful in claim construction than the patent and its prosecution history. Jd. CONSTRUCTION OF DISPUTED TERMS The parties agree that claim | of the ’667 patent and claim 1 of the ’745 patent are representative for the purpose of claim construction. Those claims state: 1. A method for content insertion, the method comprising: receiving, at a content server, a content request from a requesting computing device; modifying, at the content server, the content request by adding data to the content request, the data comprising one or more headers, attributes, or both, based on the content request; transmitting, by the content server, the modified content request to a second server based on one or more routing policies; and receiving, at the content server, a reply message responsive to the modified content request from the second server, the reply comprising a content insertion instruction comprising data indicative of content to transmit to the requesting computing device. patent at 16:46-60) (disputed terms bolded and italicized). 1. A method in a host computer for developing transformation processing operations to optimize media content playback to a plurality of playback

devices connected with the host computer in a network, the method comprising: receiving a first request from a first playback device for media content; wherein the first request contains information, the information indicating a first original media content, first content generation operations, and first transformation operations; determining whether a previously-generated first intermediate media content is available for reuse, the previously-generated first intermediate media content having been created using the first original media content and the first set of content generation operations; and responsive to determining that a previously-generated first intermediate media content is available, creating a first optimized media content for the first playback device by performing the first set of transformation operations on the previously-generated first intermediate media content; and responsive to determining that a previously-generated first intermediate media content is not available, creating a first optimized media content for the first playback device by creating a first intermediate content using the first original media content and the first set of content generation operations, and performing the first set of transformation operations on the first intermediate media content; and sending the first optimized media content to the first playback device. patent at 23:9-39) (disputed terms bolded and italicized). 1. “A method in a host computer for developing transformation processing operations to optimize media content playback to a plurality of playback devices connected with the host computer in a network, the method comprising” (’745 patent, claim 1). At oral argument the parties agreed that “method in a host computer” is limiting. (Tr. 15:14-25).? The parties agreed the remainder of the preamble is not limiting, so there is no construction needed. (/d.).

3 Citations to the transcript of the argument, not yet filed, are in the format “(Tr. []).”

The parties dispute whether Finjan applies to the claim at issue and requires only one computer --“‘a host computer” in the preamble of claim 1-- to perform all the steps in the method. Finjan LLC v. SonicWall, Inc., 84 F Ath 963, 974 (Fed. Cir. 2023). In Finjan, the Federal Circuit found that claims 1 and 22 of the patent at issue (the ARB patent) required a single computer to perform every method step. /d. at 975. The court relied on two earlier cases, Salazar and Nokia. Id. In Salazar, the court held the claim term “a microprocessor” and subsequent references to “said microprocessor” required the same microprocessor to perform each of the claim steps. Salazar v. AT&T Mobility LLC, 64 F.4th 1311, 1317 (Fed. Cir. 2023). In Nokia, the court held that “reciting “a computer’ that performs a function, and then further reciting that ‘the computer’ performs multiple additional functions, suggests that such ‘computer’ must be tied to all those functions.” Traxcell Techs., LLC v. Nokia Sols. & Networks Oy, 15 F.4th 1136, 1144 (Fed. Cir. 2021) (cleaned up). Applying the reasoning from Salazar and Nokia, the Federal Circuit found in Finjan that because the body of claim 1 of the ARB patent recited “a computer” and the following steps recited “the computer,” the same computer must perform all method steps. Finjan, 84 F.4th at 975. The Federal Circuit stated that it is “even more explicit” that Claim 22 of the ARB patent requires the same processor to perform all steps of the method. /d. Claim 22 includes reference to “a processor” in the preamble of the claim, with no subsequent reference to processor in the rest of the claim, which consists of the method steps. /d..

4 ARB claim 22 reads in whole:

A non-transitory computer-readable storage medium storing program code for causing a computer to perform the steps of: ' receiving an incoming stream of program code; determining any specific one of a plurality of programming languages in which the incoming stream is written;

Akamai argues that because claim | of the ’745 patent is like Claim 22 of the ARB patent in Finjan, claim 1 of the ’745 patent requires the same computer to carry out all the method steps. (D.I. 102 at 5). I agree. Like claim 22 of the ARB patent, claim | of the ’745 patent refers to “a [host] computer” in the preamble with no subsequent reference to “computer” in the rest of the claim. I find that Finjan mandates that there is at least one computer that carries out all of the method steps in claim | of the ’745 patent. While multiple computers can be used to implement the method, at least one of those computers must perform every step.” Equilibrium’s arguments do not convince me otherwise. Equilibrium argues that there is no claim language in claim | of the ’745 patent to which Finjan applies. (D.I. 110 at 1). Equilibrium contends that for Finjan to apply, there must be a subsequent reference such as “the computer” that refers to the antecedent “‘a computer.” (D.I. 110 at 6). In support, Equilibrium cites to Convolve, Inc. v. Compaq Computer Corp., 812 F.3d 1313, 1320 (Fed. Cir. 2016). In Convolve, the Federal Circuit found that claims reciting “‘a processor” in the preamble and subsequently referring to “the processor” in the claim body meant one processor must perform

instantiating a scanner for the specific programming language, in response to said determining, the scanner comprising parser rules and analyzer rules for the specific programming language, wherein the parser rules define certain patterns in terms of tokens, tokens being lexical constructs for the specific programming language, and wherein the analyzer rules identify certain combinations of tokens and patterns as being indicators of corresponding exploits, exploits being portions of program code that are malicious; identifying individual tokens within the incoming stream; dynamically building, while said receiving receives the incoming stream, a parse tree whose nodes represent tokens and patterns in accordance with the parser rules; dynamically detecting, while said dynamically building builds the parse tree, combinations of nodes in the parse tree which are indicator or potential exploits, based on the analyzer rules; and indicating the presence of potential exploits within the incoming stream, based on said dynamically detecting. > “Even if an infringing system can use ‘one or more computers,’ the plain language of the claims requires at least one of those computers to perform all the functions listed in the claims of the ARB Patent.” Finjan LLC v. SonicWall, Inc., 84 F.4th 963, 975 (Fed. Cir. 2023).

all claim steps. /d. at 1321. In contrast, apparatus claims 9 and 15, which recited “a processor” in the claim body with no subsequent reference to a [or the] “processor” in the process steps did not mean one processor must perform all steps. /d. Equilibrium contends that because claim | of the patent recites “a host computer” with no other references to a [or the] “host computer,” there is no requirement that one “host computer” perform all method steps. (D.I. 110 at 9-10). Equilibrium argues that the rule in Convolve “supersedes” Finjan because Convolve was decided before Finjan. (/d. at 10). Akamai replies that earlier Federal Circuit precedent controls only when two cases have “conflicting statements.” Johnston v. IVAC Corp., 885 F.2d 1574, 1579 (Fed. Cir. 1989). I do not think Convolve “supersedes” Finjan. There is no conflict between the two cases.’ In Convolve, the component at issue (“a processor”) was not recited in the preamble; it was recited only once in the body of the claims. In Finjan, the component at issue (a computer) was recited in the preamble of claim 22 with no subsequent reference to the computer. As Akamai argues, Convolve does not address the scenario that Finjan does: “a component recited in the preamble that is responsible for performing functional steps recited in the body of the claim (without explicit reference back to the preamble component).” (D.I. 111 at 4). Claim 1 of the ’745 patent is like claim 22 of the ARB patent in Finjan because it also mentions “a host computer” in the preamble with no subsequent reference to any computer. One “host computer” must perform all steps of the method.

© The way this rule is usually put is, “One panel cannot overrule another panel.” Thus, if there is a direct conflict between the holdings of two panels, the earlier decision is controlling on the would be hard to say there were “conflicting statements” even if the claims were essentially identical. The Convolve Court’s analysis was “bolstered” by the specification’s disclosure of an embodiment. 812 F.3d at 1321. That bolstering, by itself, takes the Convolve analysis out of the direct-conflict realm.

2. “the information indicating a first original media content, first content generation operations, and first transformation operations” (’745 patent, claim 1). a. Equilibrium’s proposed construction: “The information (for example information in the URL, user and/or browser characteristics, individual browsing conditions, and/or client connecting information) is used to determine the media requested, as well as the content generation and transformation operations to perform, if any. b. Akamai’s proposed construction: “indicating” means “pointing out” c. Court’s construction: No construction of “indicating” is necessary I ruled at oral argument that I will not include Equilibrium’s “for example” parenthetical. (Tr. 21:19-22:3). I also ruled I will not include the “if any.” (Tr. 39:15-17). With these portions removed, Equilibrium’s construction is, “The information is used to determine the media requested, as well as the content generation and transformation operations to perform.” The only dispute remaining is whether “indicating” means “pointing out,” as Akamai argues, or if it means “used to determine,” as Equilibrium contends. The point of this remaining dispute is obscure. I do not think either party’s proposal is a good construction. Akamai’s proposal is too narrow. “Indicating” is a broader concept than “pointing out.” Equilibrium’s proposal shifts the meaning from something the information is doing to how the information is being used. I reject the two proposed constructions. I think the parties have given me no reason to deviate from the generally broad plain meaning of “indicating.’”® Before I move on, I will discuss why I reject the parties’ arguments. Akamai argues construing indicating as “pointing out” is consistent with the plain meaning of “indicating” and with intrinsic evidence. (D.I. 79 at 17). Akamai contends that Webster’s Third New International Dictionary defines indicate as “to point out or point to or toward with more or less exactness; show or make known with a fair degree of certainty.” (D.I. 80 Ex. 6).

8 The parties are free to revisit the construction at the time of summary judgment motions.

The intrinsic evidence Akamai cites is portions of the specification that discuss how the system uses proprietary tags in URLs to carry out the method. (D.I. 79 at 18-19). For example, in the preferred embodiment the “URLs direct browsers to generate requests for media to the system. The system processes such URLs by interpreting the proprietary tags, executing the indicated image generation procedures on the original media, and returning derivative Web-safe media to client browsers via the Internet.” (°745 patent at 17:52-57). Akamai argues that the specification discloses that the media content and operations to be performed are specified in the request. The relevant portion states, “Upon receipt of the page, the browser contacts the system requesting media specified in the URL.” (’745 patent at 19:14-15). Therefore, Akamai argues, indicating should be construed as pointing out. Akamai also contends that the prosecution history supports that indicating should be construed as "pointing out." (D.I. 70 Ex. D). Akamai petitioned for inter partes review (IPR), arguing that the asserted claims’ first playback device and first content generation operations are found in the Tso prior art reference. (D.I. 79 at 19). Akamai mapped Tso’s network client and content characteristics’ to the ’745 patent’s first playback device and first content generation operations, respectively. (/d.). Akamai contends that at the IPR, Equilibrium argued that the content characteristics are not received from the network client, but from the Internet. (D.I. 79 at 20). Akamai argues that because Equilibrium distinguished the ’745 patent from prior art “in which the content generation operations are not received in a request from the client device,” Equilibrium cannot now make a conflicting argument. (D.I 79 at 21).

Examples of content characteristics are “data type, type of encoding/compression, size, and dimension.” (D.I. 70 Ex. D at 38). 10

Equilibrium argues that one meaning of “indicating” is “suggest[ing] the probable necessity or advisability of...” (Webster’s Third New International Dictionary, 1150 “Indicating” def. 2 (2002)). Equilibrium contends that therefore construing “indicating” as requiring the information to point out the exact media content, generation operations, and transformation operations would improperly narrow the claims. (D.I. 79 at 26). Equilibrium argues nothing in the claim language, specification, or prosecution history supports this construction. (/d.). A valid way to read the claims, Equilibrium argues, is that the request could “indicate” the operations by providing information that is used to determine the necessary operations. (/d.). Equilibrium replies to Akamai’s argument about URL tags by contending that no part of the specification requires the tags to point out the exact media, generation operations, and transformation operations. (D.I. 79 at 27). Rather, the specification indicates the system uses the tags to determine the media, generation operations, and transformation operations. (/d.). One example Equilibrium puts forth is that browser characteristics can be used to determine the optimizations to perform. (/d.). Equilibrium cites the following portion of the specification as support: “the resulting media is passed to the user profile system where it is again modified to account for any specific user browser characteristics specified using the proprietary URL tags.” patent at 18:45-51). Equilibrium notes that the request may use client connection data, which is not an express instruction, to determine which operations are needed. (D.I. 79 at 28). Equilibrium argues the IPR dispute over the Tso prior art reference is not pertinent because the dispute was not about whether “indicating” means “pointing out,” but about whether the information in the Tso reference came from the user’s request or from the Internet. (D.I. 79 at 30).

1]

The IPR does not help decide how to construe the term “indicating.” The IPR dispute was not about whether the request indicated the content characteristics, which would have been pertinent to the present case. Instead, it was centered around which part of the system in the Tso reference sends the content characteristics to the transcoding server. In order to disclose the “information indicating” element of the ’745 patent, Equilibrium argues the Tso reference’s transcoding server would have to receive a request from the network client. (D.I. 70 Ex. D at 37). In its IPR response, Equilibrium argues that the transcoding server receives a request not from the network client, but from Internet: “But Tso does not disclose, and Petitioner fails to demonstrate, that Tso’s transcoding server receives the network characteristics in a request from the network client. On the contrary, Tso’s express teachings demonstrate that the content characteristics are received from the Internet, not from the network client.” (/d. (cleaned up)). The prosecution history is not helpful in construing “indicating.” Since neither party has proposed a useful and accurate construction of “indicating,” and since neither the specification nor the prosecution history show a reason to alter its ordinary meaning, I decline to construe “indicating.” It is not limited to “pointing out,” and it does not mean “is used to determine.” 3. “content generation operations” (’745 patent, claim 1) a. Equilibrium’s proposed construction: “procedures for generating derivative media content, the derivative media content being composed of one or more original media” b. Akamai’s proposed construction: “procedures for generating derivative media content” c. Court’s construction: “procedures for generating intermediate media content” The only difference between the parties’ constructions is that Equilibrium seeks to add the qualifier “the derivative media content being composed of one or more original media.”

Equilibrium presents two arguments in support of adding this phrase to the construction. First, Equilibrium argues that the derivative media in the claim does not need to be “derivative” in the sense that it is transformed in some way. (D.I. 79 at 38). Equilibrium relies upon the specification’s statement that the derivative image after the first content generation step is “composed of one or more original images acquired from the media repository” (°745 patent at 18:43-45), which Equilibrium refers to as a “copy.” (D.I. 79 at 38). Second, Equilibrium contends that without the qualifier, the generation step at issue in this claim term would become conflated with the later transformation step. (D.I. 79 at 39, ’745 patent at 1:15-17). Equilibrium’s argument seems to be that without the phrase “the derivative media content being composed of one or more original media,” the generation step would require some sort of transformation. (D.I. 79 at 39). Equilibrium argues the two steps have distinct names, “generation” and “transformation,” and if both were transformation steps, the claim would refer to a first and second transformation. (/d.). Equilibrium also cites the portion of the specification explaining that the generation step places the derivative image in the media cache in response to the user’s request. (745 patent at 18:34-67). Thus, Equilibrium contends no transformation occurs at the generation step. Rather, Equilibrium argues that the transformation step (°745 patent at 18:56-67) is when the image is altered according to the needs of the user. (D.I. 79 at 43). Akamai presents two responses to Equilibrium’s arguments. First, Akamai states that Equilibrium is importing an embodiment of the invention into the claims. (D.I. 79 at 40). Akamai contends that in order to do that, the qualifier “the derivative image is composed of one or more original images in the repository” would have to be a lexicographic definition, which it is not. (Id.; see CollegeNet, Inc. v. ApplyYourself, Inc., 418 F.3d 1225, 1231 (Fed. Cir. 2005) (“In

examining the specification for proper context, however, this court will not at any time import limitations from the specification into the claims.”)). Second, Akamai disputes Equilibrium’s assertion that derivative media must be a copy of the original media, without any transformation. (D.I. 79 at 40). Akamai argues that the word “generation” in “content generation operations” establishes that the derivative media is not the same as the original media. (/d.). In support, Akamai cites the claim language that the first intermediate media content is “created using the first original media content and the first set of content generation operations.” (’745 patent at 23:19-23). Akamai argues that neither the claim language nor the specification uses the word “copy.” (D.I. 79 at 40). Akamai disputes Equilibrium’s contention that the generation step will collapse into the transformation step without Equilibrium’s proposed “composed of” language. Akamai argues that the claim language makes it clear that there are two distinct steps. (D.I. 79 at 41). Akamai contends the generation step is performed on original media to create “intermediate media,” and the transformation step is performed on the intermediate media to create final “optimized media content.” (7745 patent at 23:15-37). Akamai also contends that the specification supports that the content generation and transformation steps are two distinct steps, even though the content generation step involves transforming the original media. (D.I. 79 at 45-46). In support, Akamai cites part of the specification: ““A content generation procedure is created with instructions on how the media is to be transformed...” (°745 patent at 19:5-7). The generation step and transformation step can involve transformation, Akamai’s argument goes, but are distinct because the first operates on original media and the second on derivative media. (D.I. 79 at 46). I will construe “content generation operations” as “procedures for generating intermediate media content.”

I do not think the claims or specification require the intermediate media in the generation step to be solely comprised of original media, so I reject Equilibrium’s proposal to add the phrase “the derivative media content being composed of one or more original media.” Equilibrium argues that the claims have two steps, a generation step and a transformation step. The portion of the specification that discusses the media generation step with respect to the preferred embodiment demonstrates that the derivative image may have transformations and need not only be composed of original media. (’745 patent at 19:1-47). The specification states that if the media specified in the URL does not exist in the cache, “a media generation occurs. In the case of media generation requiring dynamic content processing...” (/d. at 19:27-29). The generation step can therefore include dynamic content processing. Dynamic content processing involves transforming the intermediate media; the dynamic image content system “generate[s] a subsequent derivative media suitably modified for the needs of zooming, panning, or slice.” (/d. at 18:45-48). The specification makes clear that intermediate media may be transformed as part of content generation. I do not find persuasive Equilibrium’s argument that the generation step and transformation step will be conflated as a result of denying the proposal that the intermediate media is composed solely of original media. The steps of the claimed method put the generation (or creation) operations (or step) before the transformation operations (or step). I do not see why Equilibrium’s proposed limitation makes the amount of “conflation” any more or less. Referring to the media content involved in content generation operations as “intermediate,” rather than “derivative” as the parties both propose, seems more helpful to the jury. The claims do not use the word “derivative.” They refer to “intermediate.” (°745 patent at 23:19-39).

4. “transformation operations” (’745 patent, claim 1) a. Equilibrium’s proposed construction: “procedures for dynamically (e.g. on the fly) modifying the generated media content, for example to account for individual needs of website users” b. Akamai’s proposed construction: “procedures for dynamically (e.g. on the fly) modifying the generated media content” c. Court’s construction: “procedures for dynamically (i.e. on the fly) modifying the generated media content” The only dispute between the parties is whether to include “for example to account for individual needs of website users.” I ruled at oral argument that I will not include the example. (Tr. 54:21-23). I therefore construe “transformation operations” as “procedures for dynamically (i.e. on the fly) modifying the generated media content.”!° 5. “server” (667 patent, claims 1, 28, and 29) a. Equilibrium’s proposed construction: “a computer system, including one or more computers, that provides data to other computer systems over a network” b. Akamai’s proposed construction: “a computer system including one or more computers and/or programs, that provides services to other computer systems over a network” c. Court’s construction: “a computer system, including one or more computers and/or programs, that provides services to other computer systems over a network” All asserted claims require at least two servers: a “content server” and a “second server.” The parties have two disputes in construing the term “server.” The first is whether the server provides “data” or “services.” At oral argument, the parties agreed a server provides “services.” (Tr. 58:2-20; 66:24-67:5). The second dispute is whether a program can be a server. Akamai argues that a server can include programs. First, Akamai contends the specification discloses the option to implement the invention as a program. (D.I. 43 at 17). Akamai cites relevant portions of the specification in support. For example, Akamai cites, “The

think the parties use “e.g.” to be definitional, not as another way to say “for example.” Hence the switch to “i.e.” If Iam wrong, the parties should so advise. 16

above-described systems can be implemented in... software. The implementation can be as a computer program product (i.e., a computer program tangibly embodied in an information carrier.)” (667 patent at 14:50-54). Second, Akamai argues extrinsic evidence supports that a server can exist as hardware and software. (D.I. 43 at 18). Akamai cites to three computer and telecom dictionaries at the time of the invention. (/d.). For example, Akamai contends Newton’s Telecom Dictionary of 1999 has both a hardware and software definition of a server, with the software definition as follows: “A server is a program which provides some service to other (client) programs.” (D.I. 44 Ex. A). Equilibrium argues that construing “server” to allow it to be a program does not make sense in light of the claims. (D.I. 43 at 21). The asserted claims discuss a content server and a second server sending messages to each other. (°667 patent at 16:46-60). Equilibrium argues that under Akamai’s proposed construction, the claims would cover a content program and a second program sending messages to each other on the same computer. (D.I. 43 at 21). Equilibrium argues that sending messages in a single computer does not make sense, and that it would be meaningless to distinguish between the two programs. (/d.). Equilibrium argues that under this reading of the claims the two servers would become one server, which fails to give meaning to the claims’ recitation of at least two distinct servers. (/d. at 22). Equilibrium contends that the specification contemplates two servers that are separate computers remote from each other. (/d.). In support, Equilibrium points to a portion of the specification that states, “A client and a server are generally remote from each other end typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.” (°667 patent at 15:62-67). Equilibrium also points out that the

specification states the Content Insertion proxy and Web Server communicate through internet messaging protocols, which they would not need to do if the server were a single computer. (D.I. 43 at 22). Equilibrium advances an argument related to the purpose of the invention. Equilibrium argues that construing server to include program would ignore that “the whole point” of the invention is to “offload” content insertion and delivery from the Web Server to the Content Insertion Proxy. (D.I. 43 at 21, citing 667 patent at 8:3-7). Equilibrium contends that if the server was a single computer, it could not offload tasks to itself. (D.I. 43 at 21). Equilibrium alleges that Akamai’s construction would fail to fulfill another purpose of the invention, which is to enable “high performance, cost effective solution for the delivery of high quality video.” (667 patent at 6:8-11). Equilibrium contends the patent discloses this purpose would be fulfilled by using a “general purpose” computer for the web server and more specialized, scalable “special purpose media streaming components” for the content server. (°667 patent at 6:8-11, 8:3-7). Akamai responds in two parts. First, Akamai contends the purpose of the patent is fulfilled under Akamai’s construction because a program can offload tasks to another program. (D.I. 43 at 26). Second, Akamai argues that the purpose of the patent is not relevant to claim construction. In support, Akamai cites E-Pass Techs., Inc. v. 3Com Corp. for the proposition,“The court's task is not to limit claim language to exclude particular devices because they do not serve a perceived ‘purpose’ of the invention.” 343 F.3d 1364, 1370 (Fed. Cir. 2003). Akamai argues that Equilibrium fails to rebut Akamai’s arguments that a server can be a program. (D.I. 43 at 23-24). Akamai also argues that it is possible to have two software servers in a single computer and that this is contemplated by the specification. (/d. at 25). The two

servers, Akamai contends, do not become one just because they are in a single computer. (/d. at 26). I think that a server can include programs. The specification discusses the possibility of the method being implemented as a program: “Method steps can be performed by one or more programmable processors executing a computer program to perform functions of the invention by operating on input data and generating output.” (°667 patent at 15:1-4). Thus, the specification contemplates (or at least does not foreclose) that the “content server” and “second server” can be two computer programs in one programmable processor.'' To say otherwise would unjustifiably narrow the scope of the patent. There is no evidence in the record for Equilibrium’s argument that reading the claims to cover a content program and a second program inside one computer would render the claims meaningless. (D.I. 43 at 21). There is also no evidence in the record for Equilibrium’s argument that there would be no way to distinguish where one program ended and the other began. (/d.). On the contrary, the fact that there are dictionary definitions for server that include software suggests that it could make sense for multiple servers to exist on one computer. (D.I. 43 at 18). Both parties cite the part of the specification that discusses the client server relationship to bolster their argument. The specification recites: “A client and a server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.” (667 patent at 15:62-67). I do not think this portion of the specification clarifies whether a server can be a program. The specification

Akamai cites many other portions of the specification that establish a server can be a program. (D.I. 43 at 16-17). I will not repeat them here. 19

contemplates that the client and server must be on separate computers. But it does not address whether the content server and second server can exist on a single computer. The portion of the specification discussing multiple servers is instructive. It states: “The system can include, for example, a plurality of servers (e.g. web server can represent a server farm or other group of servers logically and/or physically grouped together) which process the requests and/or reply to the request.” (°667 patent at 8:9-12). Servers can be grouped together logically or physically. “Logical” must refer to software, particularly given the later statement, “The above-described systems and methods can be implemented in... software.” (667 patent at 14:50-52). There is no reason that a logical grouping of servers could not exist as programs on one computer. The parties dispute whether construing “server” to include programs would fulfill the purpose of the invention. The specification does not lay out any explicit, overriding purpose. Rather, the specification states that there can be advantages to having separate servers: The policy-based content insertion techniques described herein can provide one or more of the following advantages. An advantage is that the streaming of data separated from the processing of a request for the data enables the processing and the streaming to be scaled independently of the other. Another advantage is that the decoupling of the control and application processing from that of media delivery provides for efficient scalable media streaming. patent at 5:1-9). Iam not convinced by Equilibrium’s argument that construing server as program would miss the “whole point” of the invention because I do not think offloading is the “whole point” of the invention. Rather, offloading is one advantage that some embodiments of the invention can offer. But I will not limit the scope of the claims to the invention’s purported advantages. “An invention may possess a number of advantages or purposes, and there is no requirement that every claim directed to that invention be limited to encompass all of them.” E-

Pass Techs., 343 F.3d at 1370. | am persuaded by Akamai’s argument that courts do not “limit claim language” based on “the perceived purpose of an invention.” /d. The claims do not foreclose that a server can be a program. Claim 1 states that a computing device sends a request and Claim 2 states that a client transmits the request. (°667 patent at 16:46-63). Thus, the client is a computer. But the claims never state that the servers are computing devices. Because the specification allows for the servers to be implemented as software programs, and the claims do not foreclose it, I will construe “server” as “a computer system, including one or more computers and/or programs, that provides services to other computer systems over a network.” 6. “content insertion instruction comprising data indicative of content to transmit to the requesting computing device” (°667 patent, claims 1, 28, and 29) a. Equilibrium’s proposed construction: “instruction which does not itself contain media content for transmission to the user, but which indicates to the content server what media content the content server should transmit to the user” b. Akamai’s proposed construction: “indicative of’ means “pointing out” c. Court’s construction: no construction of “indicative of” is necessary Akamai argues that this claim term needs no construction as it is readily understood by a jury. (D.I. 43 at 29). Akamai contends it is proposing construing “indicative of’ as “pointing out” for clarity, as it is the same construction Akamai proposes for “indicating” in the ’745 patent. (D.I. 43 at 30).!" Akamai argues that Equilibrium’s construction should not be used because the construction inserts an unjustified “negative limitation” into the claim. (D.I. 43 at 30-31). The negative limitation at issue is that the instruction “does not itself contain media content for

!2 The two patents are unrelated to each other. Each is at most extrinsic evidence in relation to the other. Thus, the constructions in one patent are essentially irrelevant to the determination of the constructions in the other. 21

transmission to the user, but which indicates to the content server what media content the content server should transmit to the user.”’ Negative limitations are not preferred. They are only appropriate in the rare circumstances. Two such circumstances are when a negative limitation is required by lexicography or by disclaimer. See Omega Eng’g, Inc. v. Raytek Corp., 334 F.3d 1314 (Fed. Cir. 2003). Akamai argues that neither the claims, the specification, nor the prosecution history support that the negative limitation should be read into the claim. (D.I. 43 at 31-35). The claims recite “content insertion instruction comprising data indicative of content to transmit.” (667 patent at claims 1, 28, and 29). Akamai argues that nothing in the wording of the claims suggests that the content insertion instruction cannot include content. (D.I. 43 at 31). Akamai contends that the specification has no disclaimer or lexicography prohibiting the content insertion instruction from having content that needs to be transmitted. (/d.). Rather, Akamai argues the specification recognizes the possibility that content can be transmitted in the instruction. (/d.). Akamai cites the portion of the specification that states “technology which does not and/or cannot separate the processing of control plane and media delivery can utilize the PCIP [policy-based content insertion proxy”’] to implement a scalable streaming solution.” patent at 14:7-12). Akamai argues a POSITA would know this means the web server can deliver content that is in the content insertion instruction. (D.I. 43 at 32). As for the prosecution history, Akamai argues that the statements Equilibrium cites that relate to the motivation behind the invention—for example, to offload content delivery from a web server (D.I. 36 Ex. C at 927)—4do not rise to the level of “clear and unmistakable” statements of disavowal or disclaimer needed to import a negative limitation into the claim. (D.I. 43 at 34; CUPP Computing AS v. Trend Micro Inc., 53 F.4th 1376, 1382 (Fed. Cir. 2022) (“[A]

patentee will only be bound to a disavowal [that was] both clear and unmistakable.)). Akamai cites to previous statements before the Patent and Trademark Office to contend that the prosecution history supports that the content insertion instruction should not have a negative limitation precluding it from including content. (D.I. 43 at 35, citing D.I. 36 Ex. C at 969). Specifically, the Applicants argued that the Ross prior art reference does not disclose the “content insertion instruction” because it “buffers, delays, or drops packets without adding content or data.” (D.I. 36 Ex. C at 969). Akamai argues that because the ’667 patent was distinguished from the prior art on the basis that the prior art could not add content or data, there is no basis to add a negative limitation to the claims preventing the addition of content. The prosecution history, Akamai argues, is consistent with Akamai’s interpretation that the content insertion instruction should not have a negative limitation. (D.I. 43 at 34-35). Equilibrium argues that the whole point of the invention is to offload “from the web server to the content server the tasks of content insertion and content delivery.” (D.I. 43 at 36). Equilibrium contends that allowing the content insertion instruction to include the content would disregard the point of the invention, and that the Applicants used this statement to distinguish its invention from the prior art.!? (/d.). Equilibrium cites to cases in which the Federal Circuit has found a disavowal of claim scope based on statements made during prosecution history.'* (D.1. 43 at 45). Equilibrium notes that what is claimed is the “content insertion instruction.” Equilibrium argues that if the instruction can include content, then the instruction would not be an instruction.

'3 Equilibrium made a similar argument about the purpose of the invention for the construction of “server.” In making that argument, Equilibrium only cited to the specification. (D.I. 43 at 21). Here, Equilibrium cites to the prosecution history, so I consider the argument again, this time based on the prosecution history. '4 discuss these below. 23

(D.I. 43 at 36, Tr. 77:12-17). Equilibrium contends that therefore, the plain language of the claims makes clear that the “content insertion instruction” cannot include content. (D.I. 43 at 36). Equilibrium replies to Akamai’s argument that there has been no express disclaimer by arguing there need not be an unmistakable disavowal in order to justify the negative limitation. (D.I. 43 at 37-38). Equilibrium cites Indacon, Inc. v. Facebook, Inc., 824 F.3d 1352, 1357-58 (Fed. Cir. 2016), for the proposition that, “Because the [] claim terms lack a plain or ordinary meaning in the art,” the court “need not determine whether the patentees' statements during prosecution rise to the level of clear and unmistakable disclaimer.” Equilibrium argues that Akamai mischaracterizes the way the Applicants distinguished the Ross prior art reference during prosecution. (D.I. 43 at 46). Equilibrium contends that the portions Akamai cited were not discussing adding “content or data” to the content insertion instruction, as Akamai argues, but to the content request. (D.I. 36 Ex. C at 929-930). “Ross simply does not teach or suggest a content request, let alone a content request that is modified by adding data. At best, Ross describes dropping packets or buffering and delaying packets.” (D.I. 43 at 46, citing D.I. 36 Ex. C at 929). Equilibrium contends that this part of the prosecution history does not comment on whether a content insertion instruction can include content. (D.I. 43 at 46). At oral argument (Tr. 81:3-7), Equilibrium pointed to the specification: “The content server includes a storage system and the content insertion module is further configured to associate content located on the storage system with the content request based on the content insertion instruction.” (°667 patent at 4:47-50). Equilibrium argued that “this description of architecture is consistent throughout” the specification. (Tr. 81:10-12). Equilibrium’s argument

is that the content is already in the storage system; the content is not in the content insertion instruction. (Tr. 81:13-82:15). Equilibrium also contends that no part of the specification describes the content insertion instruction as including the content to be transmitted. (D.I. 43 at 47). Instead, the Abstract and every example states that the content server, not the web server, transmits content. (/d.). I will not import “which does not itself contain media content for transmission to the user” into the construction for “content insertion instruction comprising data indicative of content to transmit to the requesting computing device.” It is a negative limitation. Equilibrium only relies upon two theories. One, that there has been disclaimer. Two, that that since “content insertion instruction” is not a term of art, it cannot be construed any broader than the disclosure. In Indacon, the claim construction issue did not involve a negative limitation. ndacon, 824 F.3d at 1357-58. It is thus not pertinent. I conclude that there is no support in the claim language, specification, or prosecution history for finding a clear and unmistakable disavowal precluding a content insertion instruction from containing content. I do not think the words of the claim lend support to importing the negative limitation into the claim. The claim simply states there is a “content insertion instruction comprising data indicative of content to transmit.” (°667 patent at claims 1, 28, and 29). Nothing indicates that the instruction cannot include content. Equilibrium argues if the instruction included the content to be transmitted, it would not be an instruction anymore, but the content itself. Not so. An instruction that says, insert this content, which can be found at xyz, and a second instruction that says, insert this content, which is attached, would both be understood as instructions about the

content to be inserted. The claim language does not foreclose the possibility that the instruction might include the content to be transmitted. (See Tr. 86:24-87:15). It is true that no description of “content insertion instruction” in the specification states that the instruction could include content. But I do not think this means I should exclude the possibility of a content insertion instruction including content from the claim scope. “[T]o disavow claim scope... [i]t is ... not enough that the only embodiments, or all of the embodiments, contain a particular limitation.” Aventis Pharma S.A. v. Hospira, Inc., 675 F.3d 1324, 1330 (Fed. Cir. 2012). “To disavow claim scope, the patentee may demonstrate intent to deviate from the ordinary and accustomed meaning of a claim term by including in the specification expressions of manifest exclusion or restriction, representing a clear disavowal of claim scope.” /d. (cleaned up). After considering Applicants’ statements during prosecution, I do not think they rise to the level of a clear and unmistakable disavowal of claim scope. Equilibrium cites to Applicants’ statements that “the insight that inspired the inventors to realize th[eir] [patented] messaging protocol” was “the desire to offload, from a web server, the bandwidth intensive task of content delivery[.]” (D.I. 36 Ex. C at 927). Equilibrium concludes Applicants distinguished the prior art reference, Ross, using “this exact aspect” of the invention. (D.I. 43 at 44, citing D.I. 36 Ex. C at 928). I do not think the desire to offload is what Applicants used to distinguish their invention from the Ross prior art reference. Applicants stated that it is the messaging structure of the invention, “rather than any specific [routing or ingress] ‘policy,’” that distinguishes the invention from the Ross prior art. (D.I. 36 Ex. C at 926, 927). The desire to offload “inspired” the messaging protocol, but the messaging protocol is discussed as including a content request, a

modified content request, and a reply to the modified content request. (/d.). The discussion of purpose does not constitute clear and unmistakable disclaimer. The Ross prior art reference concerned the messaging protocol, which deals with content requests. The relevant portion of the response to the Office action at issue states, “Ross does not teach or suggest a modified content request because Ross does not modify any packets by adding data.” (D.I. 36 Ex. C at 969). The term in dispute, however, is the content insertion instruction. I do not think Applicants’ statements distinguishing Ross shows any clear and unmistakable disclaimer that the content insertion instruction cannot include content. I quickly dispose of Equilibrium’s second argument. I do not interpret Jndacon or similar cases as endorsing “negative limitations” on the theory that the inventors used a claim term— content insertion instruction—that does not have a plain and ordinary meaning (because it is not a term of art). The “content insertion instruction” is part of larger term—content insertion instruction comprising data indicative of content to transmit to the requested computing device— that includes “data indicative of content to transmit to the requested computing device.” It does not say what it does not include. Because it “compris[es] data” etc., the claim language does not prohibit it from including other things. It would be wrong to include a negative limitation when the intrinsic record does not require a negative limitation. Because I do not think Applicants made a clear and unmistakable disclaimer in regard to including content in a content insertion instruction, and because | do not think there is any other basis to include a negative limitation, I do not include “instruction which does not itself contain media content for transmission to the user” in the construction for “content insertion instruction comprising data indicative of content to transmit to the requesting computing device.”

IV. CONCLUSION Within five days the parties shall submit a proposed order consistent with this Memorandum Opinion.