NOTE: This disposition is nonprecedential. United States Court of Appeals
for the Federal Circuit
APPLICATIONS IN INTERNET TIME, LLC,
Plaintiff-Appellant
v.
SALESFORCE, INC.,
Defendant-Appellee
2024-1133
Appeal from the United States District Court for the District of Nevada in No. 3:13-cv-00628-RCJ-CLB, Senior Judge Robert Clive Jones.
APPLICATIONS IN INTERNET TIME, LLC,
Plaintiff-Appellant
v.
SALESFORCE, INC.,
Defendant-Appellee
2024-1685
2Appeal from the United States District Court for the District of Nevada in No. 3:13-cv-00628-RCJ-CLB, Senior Judge Robert Clive Jones.
Decided: October 10, 2024
MICHAEL DEVINCENZO, King & Wood Mallesons LLP, New York, NY, argued for plaintiff-appellant. Also represented by ANDREA PACELLI, CHARLES WIZENFELD; STEVEN C. SEREBOFF, SoCal IP Law Group LLP, Westlake Village, CA.
KEVIN P.B. JOHNSON, Quinn Emanuel Urquhart & Sullivan, LLP, Redwood Shores, CA, argued for defendant-appellee. Also represented by BRIAN C. CANNON, RAY ROBERT ZADO; GAVIN SNYDER, Seattle, WA; SAM STEPHEN STAKE, OGNJEN ZIVOJNOVIC, San Francisco, CA.
Before LOURIE, LINN, and STOLL, Circuit Judges.
Applications in Internet Time (“AIT”) appeals the grants of summary judgment of non-infringement and invalidity of the asserted claims of U.S. Patent Nos. 7,356,482 (the “’482 patent”) and 8,484,111 (the “’111 patent”). See Applications in Internet Time, LLC v. Sallesforce.com, Inc., 691 F. Supp. 3d 1223, 1228, 1230–31 (D. Nev. 2023) (“Summary Judgment Order”). Further, AIT challenges the district court’s construction of “automatically detecting” as requiring the use of intelligent agents. See Applications in Internet Time, LLC v. Salesforce.com, Inc., No. 3:13-CV-00628-RCJ-CLB, 2021 WL 5238767, at *13, (D. Nev. Nov. 9, 2021) (“Markman Order”). AIT also appeals the district court’s conclusion that its case was exceptional under 35 U.S.C. § 285 and the award of nearly3 $6.9 million in attorneys’ fees. See Applications in Internet Time, LLC v. Salesforce, Inc., 3:13-CV-00628-RCJ-CLB, 2024 WL 1199594, at *6 (D. Nev. Mar. 20, 2024) (“Fee Order”).
For the reasons that follow, we modify the district court’s claim construction of “automatically detecting,” vacate the district court’s grants of summary judgment of non-infringement and invalidity, and reverse its exceptional case determination.
BACKGROUND
In 2013, AIT sued Salesforce in the District of Nevada for patent infringement of the ’482 and ’111 patents. The ’482 and ’111 patents have substantially identical specifications. The asserted patents are directed to methods and systems for automatically detecting changes to business requirements and incorporating such changes into an application. Representative claim 1 of the ’482 patent recites:
1. A system for providing a dynamically generated application having one or more functions and one or more user interface elements; comprising:
a server computer; one or more client computers connected to the server computer over a computer network; a first layer associated with the server computer containing information about the unique aspects of a particular application; a second layer associated with the server computer containing information about the user interface and functions common to a variety of applications, a particular application being generated based on the data in both the first and second layers;
4a third layer associated with the server computer that retrieves the data in the first and second layers in order to generate the functionality and user interface elements of the application; and a change management layer for automatically detecting changes that affect an application, each client computer further comprising a browser application being executed by each client computer, wherein a user interface and functionality for the particular application is distributed to the browser application and dynamically generated when the client computer connects to the server computer.
’482 patent col. 32 ll. 9–34 (emphases added).
Relevant to this appeal, unasserted claim 8 recites: 8. The system of claim 1, wherein the change management layer further comprises one or more intelligent agents that detect changes that affect an application.
’482 patent col. 32 ll. 59–61 (emphasis added).
In its Markman order, the district court construed “automatically detecting” as requiring the use of one or more intelligent agents, concluding that the specification made clear that intelligent agents were integral to the invention. Markman Order, at *6–7. Additionally, the district court afforded “changes that affect” its plain and ordinary meaning. See id. at *8.
Salesforce moved for summary judgment of non-infringement of all asserted claims. Summary Judgment Order, 691 F. Supp. 3d at 1227. Salesforce also moved for summary judgment of invalidity, contending that the claims were anticipated by U.S. Patent No. 6,249,2915 (“Popp”) and rendered obvious by the combination of Popp with an academic reference (“Amati”). Id. at 1237. AIT filed a motion for summary judgment of no anticipation, contending that Salesforce failed to prove that Popp disclosed an intelligent agent. Id. at 1227, 1242.
The district court granted summary judgment of noninfringement, reasoning that because AIT’s expert, Mr. Zatkovich, failed to compare the accused products to the specification’s description of intelligent agents, no reasonable jury could find in AIT’s favor. Summary Judgment Order, 691 F. Supp. 3d at 1234–35. The district court also granted summary judgment of invalidity, concluding that the asserted claims were anticipated by Popp and rendered obvious by the Popp–Amati combination. Id. at 1253. Determining that its Markman order ended any reasonable likelihood of AIT prevailing on the merits, the district court held that AIT’s continued litigation after the Markman order made this case exceptional and awarded Salesforce $6,890,328.28 in attorneys’ fees. See Fee Order, at *6. These appeals followed. We have jurisdiction under 28 U.S.C. § 1295(a)(1).
ANALYSIS
I. Standard of Review
“Claim construction is ultimately a question of law, decided de novo on review, as are the intrinsic-evidence aspects of a claim-construction analysis.” Intel Corp. v. Qualcomm Inc., 21 F.4th 801, 808 (Fed. Cir. 2021). “To the extent the district court, in construing the claims, makes underlying findings of fact based on extrinsic evidence, we review such findings of fact for clear error.” Williamson v. Citrix Online, LLC, 792 F.3d 1339, 1346 (Fed. Cir. 2015); see also Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318, 321–22 (2015). Claim terms generally carry their ordinary and customary meaning as understood by an ordinarily skilled artisan. Phillips v. AWH Corp., 415 F.3d 1303, 1312–13 (Fed. Cir. 2005) (en banc). “But where the6 inventor has clearly set forth a different definition of a claim term, or has manifested that the invention does or does not include a particular aspect, that intention is regarded as dispositive.” Techtronic Indus. Co. v. Int’l Trade Comm’n, 944 F.3d 901, 907 (Fed. Cir. 2019) (internal quotations omitted). Establishing disavowal requires “clear and unequivocal evidence that the claimed invention includes or does not include a particular feature.” Poly-Am., L.P. v. API Indus., Inc., 839 F.3d 1131, 1136 (Fed. Cir. 2016).
We review a grant of summary judgment under the law of the regional circuit. Adasa Inc. v. Avery Dennison Corp., 55 F.4th 900, 907 (Fed. Cir. 2022). The Ninth Circuit “review[s] the district court’s grant of summary judgment de novo, determining whether, viewing all evidence in the light most favorable to the nonmoving party, there are any genuine issues of material fact and whether the district court correctly applied the relevant substantive law.” Kraus v. Presidio Tr. Facilities Div./Residential Mgmt. Branch, 572 F.3d 1039, 1042–43 (9th Cir. 2009) (internal quotations omitted).
The determination of whether a case is exceptional under § 285 is a factual determination reviewed for clear error. Superior Fireplace Co. v. Majestic Prods. Co., 270 F.3d 1358, 1376–77 (Fed. Cir. 2001). The subsequent determination of a reasonable award is reviewed for an abuse of discretion. Id.
II. Claim Construction
Claims 1 and 21 of the ’482 patent recite “automatically detecting changes” and claim 13 of the ’111 patent recites a fourth portion of a server that is configured to “automatically detect changes.” ’482 patent col. 32 ll. 27–28, col. 33 ll. 52–53; ’111 patent col. 34 ll. 5–6. The district court concluded that the specification expressly disavowed automatically detecting changes without intelligent agents and the specification distinguished the invention from the prior art7 by its use of intelligent agents. Markman Order, at *7. Accordingly, the district court construed “automatically detecting” as limited to “detecting without human intervention through the use of one or more intelligent agents.” Id. at *13.
AIT contends that the statements the district court treated as express disavowals are merely exemplary and, thus, do not limit the invention as a whole. AIT also argues that because unasserted claim 8 expressly recites the use of intelligent agents, claim differentiation weighs against incorporating that limitation into the broader independent claim. Salesforce responds that the district court correctly construed “automatically detecting” to require the use of intelligent agents because intelligent agents are essential to the invention as a whole and the specification consistently describes the invention as including intelligent agents without providing any alternative.
We agree with AIT that the district court erred in construing “automatically detecting” as requiring the use of an intelligent agent. First, because claim 1 recites “automatically detecting changes” and claim 8 requires that the changes be detected by “one or more intelligent agents,” claim differentiation provides a strong reason not to limit claim 1 to intelligent agents. ’482 patent col. 32 ll. 27–28, 59–61; see Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 910 (Fed. Cir. 2004) (“[W]here the limitation that is sought to be ‘read into’ an independent claim already appears in a dependent claim, the doctrine of claim differentiation is at its strongest.”).
Second, the specification does not disavow “automatically detecting” without “intelligent agents.” The portion of the specification that describes “the invention” as detecting changes using intelligent agents is exemplary and not descriptive of the invention as a whole. ’482 patent col. 10 ll. 29–42; see Cont’l Cirs. LLC v. Intel Corp., 915 F.3d 788, 798 (Fed. Cir. 2019) (recognizing that descriptions of “the8 present invention” or “this invention” do not limit the claims where the intrinsic evidence does not support applying the limitation to the entire patent). The repeated use of intelligent agents in the description of the best mode merely indicates that intelligent agents may be the best way of implementing the invention, not that the invention cannot be implemented without intelligent agents. See Phillips, 415 F.3d at 1323 (“To avoid importing limitations from the specification into the claims, it is important to keep in mind that the purposes of the specification are to teach and enable those of skill in the art to make and use the invention and to provide a best mode for doing so.”). Moreover, “it is improper to read limitations from a preferred embodiment described in the specification—even if it is the only embodiment—into the claims absent a clear indication in the intrinsic record that the patentee intended the claims to be so limited.” Liebel-Flarsheim, 358 F.3d at 913.
The lack of any discussion of intelligent agents in the solution described by the Abstract, Background, and Summary of the Invention sections of the patent weighs against concluding that the inventor intended to limit the invention to the use of intelligent agents. See Cont’l Cirs., 915 F.3d at 798 (noting that the Summary of the Invention’s silence on desmear processes weighed against concluding “that the present invention ‘as a whole’ necessarily include[d] the repeated desmear process” (citations omitted)). Moreover, the Background section’s criticism of the prior art for failing to use intelligent agent routines similarly fails to limit the claim scope. “Mere criticism of a particular embodiment encompassed in the plain meaning of a claim term is not sufficient to rise to the level of clear disavowal.” Thorner v. Sony Comput. Ent. Am. LLC, 669 F.3d 1362, 1366 (Fed. Cir. 2012). Further, when discussing “[w]hat is needed” to address the problems in the prior art, the Background section merely explains that the solution9 requires automatically detecting and making changes, irrespective of whether that is done by intelligent agents. For these reasons, the district court erred in limiting the claims to require the use of one or more intelligent agents.
III. Non-Infringement
The district court granted summary judgment of noninfringement because “AIT failed to put forth evidence sufficient to raise a genuine dispute of material fact that the accused products use any such ‘intelligent agent’ to detect changes.” Summary Judgment Order, 691 F. Supp. 3d at 1233. Because we conclude that the district court erred in construing the asserted claims to require intelligent agents, we vacate the district court’s grant of summary judgment of non-infringement as to all asserted claims.
IV. Anticipation
Anticipation is a question of fact that may be resolved on summary judgment only when there are no genuine issues of material fact. Trintec Indus., Inc. v. Top-U.S.A. Corp., 295 F.3d 1292, 1294 (Fed. Cir. 2002). For the reasons detailed below, we disagree with the district court’s analysis and vacate the district court’s summary judgment order of anticipation as to all asserted claims.
A. First Layer
The district court concluded that Popp discloses the claim limitation “a first layer associated with the server computer containing information about unique aspects of a particular application,” because that phrase does not exclude data used by an application, such as the data contained on the websites in Popp. Summary Judgment Order, 691 F. Supp. 3d at 1239–40 (noting that “aspects” is broad enough to include “data” and that dependent claim 23 expressly adds a limitation that “the first layer comprises . . . data”).
10The district court erred because it did not consider the limitation as a whole. The limitation at issue claims a particular kind of information; namely, that which is “about unique aspects of a particular application.” ’482 patent col. 32 ll. 16–17 (emphasis added). Other information, such as information unrelated to a particular application, would not necessarily satisfy this limitation, even if “data” is considered a kind of information. The question for anticipation should have been whether the data disclosed in Popp is about unique aspects of Popp’s application. The district court did not ask, much less answer this question, because it erroneously treated its conclusion that “information” may include data as dispositive of the anticipation inquiry.
The district court’s reliance on the testimony of Salesforce’s expert, Dr. Bederson, was therefore misplaced. Although Dr. Bederson opined that Popp’s application fetches all the data from the database such as “all of the models of cars” within a shopper’s price range, J. App’x at 2075–76,1 Dr. Bederson does not explain how “models of cars” (i.e. “data”) stored on a database and subsequently displayed on a webpage are “about a unique aspect” of the webpage.
Because the district court’s summary judgment order fails to explain why a reasonable jury could not find that Popp fails to disclose the “first layer” limitation, we vacate that portion of the district court’s order.
B. Changes that Affect an Application
At the Markman stage, the district court afforded the term “changes that affect” its plain and ordinary meaning. Markman Order, at *8. At summary judgment, the district court purported to apply that construction, holding that “a skilled artisan would find the claim language broad enough11 to encompass any ‘changes’ related to an application,” and relying heavily on Dr. Bederson’s opinion that Popp’s “detection of a name entered in a field to be a change that affects an application under at least . . . AIT’s interpretation.” Summary Judgment Order, 691 F. Supp. 3d at 1243.
AIT argues that the district court’s adoption of Dr. Bederson’s opinion on anticipation was not based on the plain and ordinary meaning of “changes that affect,” but on a new and overly broad construction that fails to appreciate the limitation as a whole, which requires the detected changes to be “changes that affect an application” rather than just “related to an application.” See id. (emphasis added). AIT also argues that the district court failed to view all inferences in AIT’s favor (as it was required to do on summary judgment) by baselessly rejecting Mr. Zatkovich’s testimony that the information entered into Popp’s data field leaves the application “entirely unaffected.” J. App’x 2904 at ¶ 255.
Salesforce argues that the ordinary meaning of “changes that affect” is not limited to changes affecting the functionality of an application, but can encompass changes to the input data on which the application operates. Further, Salesforce contends that without such a limitation, there can be no genuine issue of material fact that Popp discloses automatically detecting changes to such inputs.
We agree with AIT that the district court failed to apply the proper standard for summary judgment. Rather than viewing the evidence in the light most favorable to AIT, the non-moving party, the district court adopted the perspective of Dr. Bederson, and rejected the testimony of Mr. Zatkovich without explanation. The district court did not explain why any reasonable jury must accept Dr. Bederson’s testimony and reject Mr. Zatkovich’s. Moreover, the district court analyzed anticipation under “Dr. Bederson’s understanding of AIT’s interpretation” of the12 claim language, rather than the claim construction it previously adopted. This was error.
To the extent the district court’s holding was based on a modification or clarification of its “plain and ordinary meaning” claim construction, this too was error, because the district court failed to provide any reason for any such modified construction. Moreover, to the extent that the district court resolved a dispute between the parties about the proper application of the court’s claim construction, the district court erred by resolving a factual issue on summary judgment.
C. AIT’s Impeachment Evidence
AIT additionally argues that the district court erred by relying on Dr. Bederson’s report, even though AIT presented credible evidence that Dr. Bederson’s report did not accurately represent his personal expertise, but was copied from Dr. Schmidt’s declaration. From this, AIT argues that a reasonable jury could have reasonably disbelieved Dr. Bederson’s report and testimony, and thus the district court erred in granting summary judgment based so heavily on his evidence.
We also agree with AIT that the district court erred in holding that summary judgment should not be denied simply because AIT asserted that Dr. Bederson’s testimony is not to be believed. Summary Judgment Order, 691 F. Supp. 3d at 1251 (citing TypeRight Keyboard Corp. v. Microsoft Corp., 374 F.3d 1151, 1158 (Fed. Cir. 2004) (“Summary judgment should not be denied simply because the opposing party asserts that the [movant’s] witnesses are not to be believed.”)). AIT did not merely “assert[] that the movant’s witness[ is] not to be believed.” See id. Instead, AIT reasoned that Dr. Bederson’s opinions as to Popp were questionable based on the fact that much of his testimony was identical to that of Dr. Schmidt—Salesforce’s expert in a separate ex parte reexamination proceeding—and that Dr. Bederson could not specify how much of his report he13 wrote or edited. These allegations provide specific bases for doubting the credibility of Dr. Bederson’s testimony and go to the kinds of facts that TypeRight recognized could preclude summary judgment. See 374 F.3d at 1159 (“[S]ummary judgment is not appropriate where the opposing party offers specific facts that call into question the credibility of the [movant’s] witness[].”). The district court here failed to explain why a reasonable jury was required to accept Dr. Bederson’s testimony in light of AIT’s allegations.
D. Claims 23–26
The district court concluded that there was no genuine issue of material fact that Popp anticipates claims 23–26 of the ’482 patent because they are substantially identical to claims 14–17 of the ’111 patent. Summary Judgment Order, 691 F. Supp. 3d at 1248.
AIT argues that the district court erred in so concluding because claims 23–26 include additional limitations— “a business content database” in claim 23 and its dependent claims and “a metadata database” in claim 25—not present in claims 14–17. AIT argues that Dr. Bederson provided no opinions as to whether Popp anticipates claims 23–26 and that his opinions as to claims 14–17 do not compel a determination of anticipation as a matter of law for claims 23–26.
Salesforce defends the district court’s grant of summary judgment on the grounds that “[t]he differences in claim language are minor” and that Popp discloses “an enterprise’s corporate database.” Salesforce also disputes AIT’s claim that Dr. Bederson did not opine that Popp anticipates claims 23–26, arguing that Salesforce filed an opposed motion to file a supplemental invalidity report including Dr. Bederson’s opinions as to claims 23–26 that the district court denied as moot.
14We agree with AIT that the district court erred. First, the district court failed to grant every reasonable inference in AIT’s favor. The “business content database” and “metadata database” elements recited in Claims 23–26 have facially different scopes than the “business knowledge” and “metadata” elements recited in claims 14– 17. Therefore, a showing that Popp anticipates the latter does not necessitate a conclusion that Popp likewise anticipates the former. Second, the anticipation inquiry is a search for strict, not substantial, identity. See Trintec Indus., 295 F.3d at 1296. Here, the district court acknowledged that claims 23–26 of the ’482 patent differ from claims 14–17 of the ’111 patent. Summary Judgment Order, 691 F. Supp. 3d at 1248 (“The only differences between these two sets of claims identified by AIT are the ‘business content database’ (claim 23) and ‘metadata database’ (claims 25–26), but claims 14–17 of the ’111 patent already require ‘information associated with one or more predetermined business applications’ (claim 14), ‘business knowledge’ (claim 15), and ‘metadata’ (claims 16–17).’”). The district court failed to explain, however, why the scope of claims 23–26 is necessarily congruent with the scope of claims 14–17.
- * * *
For these reasons, we vacate and remand the district court’s grant of summary judgment of anticipation of all asserted claims.
V. Obviousness
The district court also granted summary judgment of invalidity, concluding that all the asserted claims were obvious over the combination of Popp and Amati. This was error. The district court’s obviousness determination depended in some measure on both its erroneous conclusion, discussed above, that Popp discloses each of the asserted claims’ limitations as a matter of law, and on its erroneous claim construction requiring an “intelligent agent,” also15 discussed above, that led to its conclusion that if Popp did not disclose an “intelligent agent,” Amati did, and a person of ordinary skill would have been motivated to combine the references. Further, the district court rejected AIT’s evidence of commercial success and technical advantages of the accused product as objective indicia of non-obviousness given its finding of non-infringement. The district court’s determination of non-infringement, however, was based on the absence of intelligent agents in the accused product, an absence that no longer leads to non-infringement in light of our holding on claim construction, supra.
Because the district court’s grant of summary judgment of obviousness depended on its erroneous claim construction, non-infringement, and anticipation determinations, we vacate the district court’s grant of summary judgment of obviousness.
VI. Fees
The district court determined that AIT’s case was exceptional under 35 U.S.C. § 285 because AIT continued to litigate “after claim construction end[ed] any reasonable likelihood of prevailing on the merits.” Fee Order, at *6. Because the district court’s determination relied on an incorrect claim construction, and because we have now vacated the district court’s summary judgments of noninfringement, anticipation, and obviousness, we reverse as moot the district court’s finding that AIT’s case was exceptional under § 285.
Additionally, “[w]hether or not a district court ultimately finds a case exceptional on motion for attorney fees, it is important that the court provide some indication of the reasoning underlying its decision to provide a basis for meaningful review.” Superior Fireplace, 270 F.3d at 1377. The district court, however, failed to explain why its claim construction order ended any reasonable likelihood of AIT succeeding on the merits. Specifically, the district court did not articulate whether its claim construction order16 prevented AIT from proving infringement, avoiding invalidity, or both. Because infringement and invalidity cannot be assessed by looking to the claims alone, the district court needed to specify how its claim construction order prevented AIT from prevailing on the merits in view of the nature of the accused product or Salesforce’s proffered prior art references.
The district court’s assessment of the reasonableness of Salesforce’s attorneys’ fees was similarly threadbare. The district court simply concluded that “Salesforce is entitled to the $6,890,328.28 incurred after the Court’s Claim Construction Order” because “Salesforce has satisfied its burden under the Court’s Local Rules to provide a ‘reasonable itemization and description of the work performed,’ and the Court finds that the requested amount is reasonable considering the context of the case and the substantial amount in dispute.” Fee Order, at *6. But the district court provided no explanation as to why it held Salesforce’s attorneys’ fees to be reasonable, even though the narratives for all of Salesforce’s time entries were redacted.
Should the district court encounter these questions again on remand, consistent with our discussion above, the district court should articulate the factual findings underpinning its assessment of exceptionality and explain its rationale as to why the fees awarded are reasonable.
CONCLUSION
For the foregoing reasons, we modify the district court’s claim construction; reverse the district court’s exceptional case determination; vacate the district court’s grants of summary judgment of non-infringement, anticipation, and obviousness; and remand for further proceedings consistent with this opinion. Because “denials of summary judgment are ordinarily not appealable,” M. Eagles Tool Warehouse, Inc. v. Fisher Tooling Co., 439 F.3d 1335, 1344 (Fed. Cir. 2006), we do not address AIT’s17 passing arguments that the district court erred by denying summary judgment of no anticipation.
REVERSED-IN-PART, VACATED-IN-PART, AND
REMANDED
COSTS
Costs are awarded to AIT.
Footnotes
- ↩ 1 “J. App’x” refers to the appendix filed under appeal No. 2024-1133.