by Dennis Crouch
Back in October 2024, I previewed the Federal Circuit case of Lynk Labs v. Samsung in a post titled Secret Springing Prior Art and Inter Partes Review. The court has now released its decision — holding that in IPR proceedings a published patent application is considered prior art as of its filing date. [Read it here] This solidifies the approach already taken by the USPTO and is significant because a substantial percentage of references relied upon to cancel patent rights represent “secret prior art” that were non-public at the time the patent was originally filed. The court’s analysis delves deep into statutory interpretation, legislative history, and the balance between different sections of patent law. Unfortunately, the Judge Prost opinion is almost unintelligible at its most critical point – where it explains how the legal term “printed publication” is date agnostic.
The central issue in Lynk Labs revolved around the interpretation of § 311(b)’s limitation that IPR challenges may be based only on “prior art consisting of patents or printed publications.” A patent application is typically published 18-months after its effective filing date and at that point clearly qualifies as a “printed publication.” The question though is what date to apply to the publication.
- On the one hand – A publication is typically defined by its date of publication, i.e., when it becomes publicly accessible.
- On the other hand – The patent statute provides a particular exception for patent applications that are published. Once published, the application is back-dated to its filing date. See 35 U.S.C. § 102(a)(2) and § 102(e)(1) (pre-aia).
The Federal Circuit’s analysis began with an examination of the term “printed publication” and its historical interpretation in patent law. Drawing from but ultimately distinguishing the seminal case In re Hall, 781 F.2d 897 (Fed. Cir. 1986), the court explained that while Hall case established the “public accessibility” standard for printed publications, the temporal requirement actually stemmed from § 102’s specific language rather than the term “printed publication” itself.
The “when” issue in these cases was controlled by other, temporal language in these subsections—in § 102(a), “described in a printed publication . . . before the invention thereof by the applicant for patent.
This distinction by the court proved crucial in its conclusion that patent applications should be treated under § 311(b). According to the court, the term “printed publication” just asks: was the material ever made publicly accessible? If so, it satisfies the 311(b) requirement.
I am struggling somewhat with the court’s reasoning. First, there is a fundamental tension in treating something as a “printed publication” before it was actually printed or published. While the court attempts to separate the timing requirement from the meaning of “publication,” the ordinary meaning of “publication” inherently involves making something public at a particular moment in time. When Congress used the phrase “prior art consisting of patents or printed publications” in § 311(b), it chose language that naturally suggests looking at when something became publicly accessible. The court’s interpretation effectively reads this temporal aspect out of the term “publication” without adequately explaining why Congress would have intended such an unnatural reading.
The court offers some textual analysis, but it is problematic. While claiming that the timing requirement comes from statutory language like “described . . . before,” the court never adequately explains how the word “described” creates this temporal requirement. “Described” appears to be doing substantive work about what must be disclosed, not temporal work about when it must be disclosed. If anything, the timing requirement seems to flow naturally from the word “publication” itself – something isn’t really “published” until it is made public. The court then attempts to tie the date to the critical date of the patent at issue, but that critical date is a deadline for publication — it does not tell us the prior art date of the publication itself. The court’s attempt to separate these concepts strains both ordinary meaning and statutory interpretation principles without sufficient justification.
The court’s reasoning could lead to absurd results that Congress surely did not intend. The court’s premise here is that the “prior art” and “printed publication” requirements of Section 311(b) are entirely separable. Under the holding, we need to know, is the reference being submitted a publication and separately is it prior art. Following this analysis sales and public uses could also become prior art if documentation of those events were later published. For instance, if someone publishes details about a public use years after it occurred, under the court’s logic that could constitute “prior art” as a sale and a “printed publication.” This would effectively eliminate Congress’s limitation of IPR proceedings to patents and printed publications.
In its analysis, the Court also drew significant support from Qualcomm Inc. v. Apple, which had previously examined Congress’s purpose in restricting IPR challenges to patents and printed publications. The Qualcomm decision emphasized that Congress intended IPRs to provide a cheaper and less time-consuming alternative for patent challenges that don’t require extensive discovery. This legislative intent proved crucial in understanding why published patent applications should be treated consistently across different forums. The court recognized that allowing published applications to serve as prior art from their filing dates in IPRs would further Congress’s goal of creating an efficient alternative to district court litigation. The alignment with congressional purpose provided additional justification for the court’s interpretation of § 311(b). I also do recognize that the court is working within a statutory framework where Congress explicitly created special rules for published patent applications in 102(a)(2)/102(e)(1). While there may be tension with the ordinary meaning of “publication,” Congress did specifically choose to give applications a different temporal treatment upon their publication. The court could argue it’s simply implementing this Congressional choice rather than redefining what constitutes a publication.
While the Federal Circuit’s decision in Lynk Labs focused specifically on pre-AIA 35 U.S.C. § 102(e)(1), the court’s reasoning would likely apply with equal force to post-AIA 35 U.S.C. § 102(a)(2). The post-AIA statute similarly treats publication of a patent application as cause to back-date the prior art status to the application’s filing date their filing dates. There is every reason to believe courts will apply the same analysis to permit IPR challenges based on § 102(a)(2) prior art.
The Federal Circuit panel consisted of Judges Lourie, Prost, and Stark, with Judge Prost writing the opinion. Lynk Labs was represented by Stephen Schreiner along with James Carmichael, Stephen McBride, and Minghui Yang all of Carmichael IP. Samsung Electronics was represented by Naveen Modi along with the team of Joseph Palys, Igor Timofeyev, David Valente, and Daniel Zeilberger of Paul Hastings LLP. The USPTO intervened in the case and was represented by Michael Forman, Peter Ayers, Amy Nelson, Maureen Donovan Queler, and Farheena Rasheed from the Office of the USPTO Solicitor.