Last week the summer was winding down and the Federal Circuit was gearing up for its September argument session. But the Court still found time to hand down a number of decisions—17 in total. Below we provide our usual weekly statistics and our case of the week—our highly subjective selection based on whatever case piqued our interest.
Precedential opinions: 9
Non-precedential opinions: 8
Rule 36: 0
Longest pending case from argument: Baxalta Inc. v. Genentech, Inc., No. 19-1527 (206 days)
Shortest pending case from argument (non-Rule 36): Tie between OMG, Inc. v. United States, No. 19-2131; Midwest Fastener Corp. v. United States, No. 19-2226; and Simpson Strong-Tie Company v. United States, No. 19-2403 (all 24 days)
Case of the week: Egenera, Inc. v. Cisco Systems, Inc., No. 19-2015 (Aug. 28, 2020)
Panel: Chief Judge Prost and Judges Reyna and Stoll, with Chief Judge Prost writing the opinion.
You should read this case if: You have a matter involving means-plus-function claiming, correction of inventorship, or judicial estoppel.
Our case of the week would make a great law school exam fact pattern, since it involves several interlocking (and interesting) issues running the patent-law gamut—from claim construction to statutory interpretation to estoppel.
Egenera owns a patent covering a virtual network that can be quickly reconfigured as computing power needs change. An outside network may use different protocols from the virtual network, so there must be some way to modify outbound messages to be compatible with the external network. Thus, the claims recite a control node with “logic to modify said received messages to transmit said modified messages to the external communication network” (emphasis added). All the issues on appeal can be traced back to this “logic to modify” term.
Egenera sued Cisco for patent infringement, and Cisco requested an inter partes review of the asserted patent. Egenera responded by petitioning the PTO to remove one of the listed inventors, Peter Schulter, as an inventor. In the Court’s words, it was “apparent that at least part of Egenera’s motivation” was to help Egenera swear behind a prior-art reference in the IPR: Mr. Schulter’s status as an inventor would have undermined Egenera’s asserted conception date, which preceded the start of Mr. Schulter’s employment there. While the inventorship petition was pending, the Board declined to institute Cisco’s IPR. It assumed that the asserted reference was prior art but concluded that Cisco had not established a reasonable likelihood of prevailing on the merits. The PTO subsequently granted Egenera’s petition and removed Mr. Schulter’s name from the patent.
Back in district court, Cisco argued for the first time that that the “logic to modify” limitation should be construed as a means-plus-function term under § 112(f). The district court agreed and determined that the corresponding structure in the specification was a certain “tripartite structure.” Based on that construction, Cisco contended that the claims were invalid for improper inventorship under pre-AIA § 102(f), as Mr. Schulter allegedly invented the tripartite structure, but the patent no longer listed him as an inventor. Egenera responded by arguing that, if Mr. Schulter was an inventor, the patent’s inventorship should be corrected under § 256(b), which provides that “[t]he error of omitting inventors . . . shall not invalidate the patent” if the court can correct the list of inventors instead.
After a bench trial on inventorship, the district court found that Mr. Schulter had conceived the tripartite structure and thus was an inventor. And it concluded that Egenera was judicially estopped from invoking § 256(b) to restore Mr. Schulter’s inventorship. Thus, it held the patent invalid. Egenera appealed the claim construction and the application of judicial estoppel.
The Federal Circuit affirmed the claim construction but overturned the judicial-estoppel determination. To start, the Court held that the district court correctly construed “logic to modify” as a means-plus-function term. As our readers likely know, after the Court’s 2015 en banc decision in Williamson v. Citrix Online, a claim term without the word “means” is presumed not to be means-plus-function. But that presumption can be overcome if the claim term “fails to recite sufficiently definite structure,” as is the case for generic terms or abstractions such as “mechanism,” “element,” and “device.” The Court held that presumption was rebutted here. Even assuming that “logic” connotes software, firmware, or circuitry, as Egenera had argued, the Court concluded the term was still just a “black box” that did not provide sufficient structure for performing the associated modification function. And it rejected Egenera’s argument that “logic” is structural because it is part of the “control node,” reasoning that “[m]ere inclusion of a limitation within a structure does not automatically render the limitation itself sufficiently structural.”
But the Court went on to hold that the district court erred in applying judicial estoppel to prevent correction of inventorship and thus invalidate Egenera’s patent. As a threshold matter, the Court rejected Cisco’s contention that inventorship could not be corrected, even absent judicial estoppel, because Egenera’s initial removal of Mr. Schulter was a “tactical ploy” and not an “error” within the meaning of § 256(b). “‘Error’ is simply the incorrect listing of inventors,” the Court explained, and includes not only “unintentional inaccuracy” but also “considered acts” and even “deceptive intention.” And Mr. Schulter’s omission was “error” because, under the district court’s claim construction and findings at trial, he was an inventor.
On the merits of judicial estoppel, the Court concluded that the district court’s analysis of all three relevant factors was legally erroneous, and it vacated the invalidity judgment premised on the judicial-estoppel determination.
First, the Court determined that Egenera had not advanced “clearly inconsistent” or “mutually exclusive” positions before the PTO and the district court. When Egenera petitioned the PTO to remove Mr. Schulter’s name, the district court had not yet determined that “logic to modify” was a means-plus-function term. Indeed, the Court observed, “it may well be that Mr. Schulter would not be an inventor” were that term not construed as means-plus-function, as Egenera had contended. So “at least due to the intervening claim construction,” Egenera’s successive requests for correction of inventorship were not premised on “mutually exclusive” positions.
Second, the Court determined that Egenera had not “succeed[ed] in persuading a court or court-like tribunal to accept its” earlier position. It noted that the PTO examines a request for correction of inventorship “only for the presence of supporting statements and the required fee” and “does not consider the substantive adequacy of the petition.” Thus, the Court held, a § 256 petition, “without more,” does not count as “persuasion” of a “court” for judicial-estoppel purposes. But the Court emphasized that this holding was “narrow”: it acknowledged that positions taken before an administrative tribunal can give rise to judicial estoppel in some circumstances. And it left open the possibility that judicial estoppel could apply to statements made during “substantive prosecution, ex parte reexamination, or other quasi-adjudicatory proceedings,” and that other theories of estoppel could apply based on even “ministerial filings” with the PTO.
Third, the Court concluded that “Egenera would gain no unfair advantage, and Cisco would suffer no unfair prejudice, if judicial estoppel were not applied.” Had Egenera succeeded in swearing behind the prior art based on its earlier § 256 petition, the Court stated, “[t]hings might be different.” There would be “clear” prejudice and inconsistency in the potential for two decisions based on contradictory inventorship statuses. But the Court pointed out that Egenera’s earlier position did not end up affecting the IPR, as the Board denied institution even without considering Egenera’s priority arguments.