In Partial Reversal of Two IPRs, CAFC Says PTAB Erred in Prior Art Determination

Image taken from court documents

On August 17, the U.S. Court of Appeals for the Federal Circuit (CAFC) affirmed in part, reversed in part, vacated in part, and remanded two final written decisions from the Patent Trial and Appeal Board (PTAB or Board) involving inter partes review (IPR) proceedings that affirmed the patentability of two of Ironburg Inventions Ltd.’s patents.

Valve Corporation (Valve) filed two IPRs arguing that certain claims of both U.S. Patent No. 9,289,688 (the ‘688 patent) and U.S. Patent No. 9,352,229 (the ‘229 patent) are unpatentable.

In the first IPR proceeding, Valve petitioned for review of claims 1-3, 9, 10, 18-22 and 26-30 of the ‘688 patent on February 7, 2017. The ‘688 patent is directed to “an improved controller…for a games console that is intended to be held by a user in the same manner as a conventional controller that comprises additional controls…located on the rear of the controller.” Claim 29, in relevant part further discloses “the first elongate member is inherently resilient and flexible.”

The Board found that claims 1, 2, 9, 10, 20, 22, 27, 28, and 30 of the ‘688 patent were shown to be unpatentable as anticipated by U.S. Patent Application Publication 2015/0238855 (“Uy”). Additionally, the Board determined that claims 18, 19, 21, 26, and 29 were not shown to be unpatentable as obvious over a claimed prior art reference the “Burns article” in combination with other prior art references because the Burns article that Valve relied upon was never authenticated. The Board also found that claim 29 was not unpatentable because Uy did not teach an “elongate member” that was “inherently resilient and flexible.”

Valve appealed the PTAB’s determination that claims 18, 19, 21, 26, and 29 were not shown to be unpatentable, and Ironburg cross-appealed arguing that Board incorrectly determined claims 1, 2, 9, 10, 20, 22, 27, 28, and 30 were unpatentable as anticipated by Uy.

In the second IPR proceeding, Valve petitioned for review of all 24 claims of the ‘299 patent on August 9, 2017. Like the ‘688 patent, the ‘229 patent is directed to “a handheld controller for a games console, wherein the controller further includes at least one additional control located on the back of the controller, where the additional controls may be paddle levers.” The ‘299 patent, also like the ‘688 patent disclosed “an elongate member that is inherently resilient and flexible.”

The Board determined that none of the claims of the ‘229 patent were shown to be anticipated by Uy because Uy did not teach “an elongate member that is inherently resilient and flexible,” and because Valve failed to authenticate the Burns article which it relied upon as a prior art reference to show obviousness. Valve appealed the Boards decision.

What Constitutes Prior Art?

At issue in both of Valve’s appeals is whether an exhibit relied upon by Valve was shown to be a printed publication and thus prior art under 35 U.S.C. §102(a)(1). Valve relied upon the Exhibit to argue that claims 1, 2, 9, 10, 18, 19, 21, 26, and 28-30 of the ‘688 patent and claims 1, 2, 9-17, and 21-24 of the ‘229 are unpatentable as obvious in view of the Exhibit in combination with other references. Valve represented to the Board that the Exhibit was a printed copy of an online review of an Xbox 360 controller, titled “Review: Scuf Xbox 360 Controller” by Dave Burns, published on October 20, 2010. The Exhibit discloses a “controller” having “2 paddles to the back of the control pad which are made from polycarbonate.” Additionally, the Exhibit discloses an “elongate member that is inherently resilient and flexible.”

Valve argued that the Exhibit should be considered prior art because it was a printout of the same online article by Dave Burns that was cited and enclosed in the prosecution histories of both the ‘688 and ‘299 patents along with another one of Ironburg’s patents, U.S. Patent No. 8,641,525 (the ‘525 patent), and the prosecution histories of these patents are all prior art.

The Board rejected Valve’s showing on the ground that the Exhibit had not been shown to be the same as the asserted prior art documents in the prosecution histories, which is “an issue of authentication.” Ironburg contended that the documents in the prosecution history were not prior art because they had not been shown to be publicly accessible.

No Material Differences

The CAFC first addressed the Board’s determination that the Exhibit was not shown to be prior art because it was not shown to be the same as the documents in the prosecution history. Citing In re Grand Jury Subpoena Duces Tecum, the CAFC determined that “authentication by comparison is routine,” therefore, it would be proper to authenticate the Exhibit by a comparison with an authenticated specimen by an expert witness or trier of fact. In re Grand Jury Subpoena Duces Tecum Dated Oct. 29, 1992, 1 F.3d 87, 93 (2d Cir. 1993). Here, the CAFC noted that the Exhibit and the prosecution documents are virtually identical. The only difference that Ironburg was able to identify was the dates that the article was accessed. During the IPRs, the Board declined to compare the Exhibit with the prosecution histories, believing that they were not obliged to do so absent written testimony that the two are identical. However, the CAFC held that no such testimonial requirement exists, and therefore “the Board had an obligation to conduct the comparison[between the Exhibit and the prosecution histories in order to authenticate the Exhibit], as Rule 901(b)(3) contemplates and as the cases require.”

On cross appeal, Ironburg pointed out that some of the images in the ‘525 Burns article and the Exhibit were missing in the copies of the Burns article in the ‘688 and ‘229 patent’s prosecution histories. However, the CAFC was unconvinced by these differences, noting that the differences appear to simply be due to the manner in which the copies were downloaded and printed and “does not amount to affirmative evidence challenging the Exhibit or the ‘525 Burns article’s material facts.” Accordingly, the CAFC concluded that the Exhibit is substantively the same as the Burns article of the prosecution histories of the ‘525 patent and the ‘688 and ‘229 patents.

Printed Publications

The Board did not determine whether the copies of the Burns article were printed publications. Ironburg argued that Valve failed to show that the copies of the Burns article in the prosecution histories were prior art because they were not shown to be publicly accessible before the priority date. However, the CAFC was unconvinced, noting that “there is overwhelming evidence that at least the ‘525 Burns article is prior art, and thus the Exhibit, which is substantively the same, is also prior art.” The determination of whether a reference qualifies as a printed publication under Section 102 is a legal conclusion based on underlying factual findings. The earliest priority dates for the ‘688 patent and the ‘229 patent are April 14, 2021 and May 22, 2013, respectively. Accordingly, the Exhibit would be prior art for both patents if it were publicly accessible prior to May 22, 2013. Several pieces of evidence indicated that the ‘525 Burns article was publicly available prior to the critical date. First was the testimony of Mr. Burgess, a co-inventor of the ‘525, ‘688, and ‘229 patents, in which he explained that he facilitated the publication of the Burns article in order to promote his business. Second, the patent examiners found the Burns article was published on October 20, 2010, more than two years before the critical date. Third, the applicants during prosecution, including Mr. Burgess, did not dispute the October 20, 2010 publication date of the Burns article. Lastly, the patent examiner for the ‘525 patent independently located the Burns article no later than May 15, 2013. Therefore, the CAFC reversed the Board’s determination that the Exhibit was not prior art, vacated the Board’s determination that claims 18, 19, 21, 26, and 29 of the ‘688 patent and claims 1, 2, 9-17, and 21-24 of the ‘229 patent were not shown to be unpatentable, and remanded for the Board to consider Valve’s arguments that relied on the Exhibit as to those claims.

Obviousness

Separately, Valve appealed the Board’s determination that Valve failed to prove claim 29 of ‘688 and claims 1-15 and 18-24 of the ‘229 patent were unpatentable as anticipated by Uy or as obvious over a combination of Uy and U.S. Patent No. 5,989,123 (Tosaki). Valve argued that the Board erred in determining that Uy did not teach an additional control located on the back of the case of the controller comprised of an “elongate member that is inherently resilient and flexible.”

Valve relied on two levers disclosed by Uy to argue that Uy taught the “elongate member.” Additionally, Valve argued that Uy taught the “inherently resilient and flexible” limitation for the elongate member because “the resilience and flexibility may be provided by a biasing spring” which is disclosed by Uy. However, the Board determined that because the springs and levers disclosed by Uy are formed separately, Uy does not teach levers alone that are inherently resilient and flexible. Valve reasons that the Board erred in construing the claims to exclude any elongate member made from multiple parts. However, the CAFC did not find this argument persuasive. As a result, the CAFC affirmed the Board’s decision of no unpatentability for claims 3-8 and 18-20 of the ‘229 patent which Valve relied solely upon Uy. Valve additionally relied on the Exhibit, either alone or in combination with another reference, to meet the limitation for claim 29 of the ‘688 patent and claims 1, 2, 9-15, and 21-24 of the ‘229 patent. As to those claims, the CAFC remanded to the Board to consider Valve’s grounds that relied upon the Exhibit.

Cross Appeal

Ironburg Cross appealed the Board’s determination that Uy anticipates claims 1, 2, 9, 10, 20, 22, 27, 28, and 30 of the ‘688 patent. Ironburg argued that Uy does not teach two limitations pertaining to the “elongate member.” The parties disputed the meaning of “disposed proximate” as it relates to the first limitation, “a first surface disposed proximate an outer surface of the case.” Ironburg contends that “disposed proximate” means that “ the first surface of the elongate member is positioned to face the outer surface of the case.” Whereas Valve argued that “disposed proximate” means “close to the outer surface without having to face the outer surface.” The Board did not resolve the disputed interpretation because they concluded that even with Ironburg’s narrower interpretation, Uy would anticipate the claims in question. The CAFC agreed, affirming the Board’s decision, noting that Valve’s claim construction was correct.

In the alternative, Ironburg argued that Uy fails to teach an elongate member with “a second surface opposing the first surface.” Instead, Ironburg argued the correct construction of the limitation is a “long thin member with substantially flat opposing surfaces.” According to the Board’s interpretation, the limitation’s “a second surface opposing the first surface” language “refers to two surfaces that face generally opposite directions” and does not require the surfaces to be substantially flat. The CAFC concluded that the Board’s interpretation was the correct one, reasoning that the plain meaning of “opposing” does not mean substantially flat, nor does the specification provide a basis for this interpretation. Therefore, the CAFC affirmed the Board’s determination that claims 1, 2, 9, 10, 20, 22, 27, 28, and 30 of the ‘688 patent are unpatentable as anticipated by Uy.