The Federal Circuit has held that a patent claim term construed under 35 U.S.C. § 112 is indefinite where, although the claim term encompassed a concept that was well known in the art, that concept was embodied by a class of algorithms, and therefore it was unclear what specific algorithm was contemplated. Triton Tech of Texas, LLC v. Nintendo of America, Inc., 2013-1476 (June 13, 2014). Coming on the heels of the U.S. Supreme Court's decision in Nautilus, Inc. v. Biosign Instruments, Inc., this decision reinforces the advisability of erring in the direction of defining claim terms, no matter how well known or understood they would seem to be.
U.S. Patent No. 5,181,181 included claims reciting in part "integrator means associated with said input device for integrating said acceleration signals over time to produce velocity signals for linear translation." The district court had found that a broad reference to "numerical integration" in the disclosure was not by itself "a sufficient disclosure," reasoning that "[n]umerical integration" was "not a single algorithm, but rather a whole class of algorithms that can be used to calculate definite integrals.” The Federal Circuit agreed that this was a basis for finding the claims indefinite.
The patent owner "argue[d] that merely using the phrase “numerical integration” is sufficient disclosure of an algorithm because numerical integration was well known to those skilled in the art." However, the Federal Circuit, in an opinion by Judge Moore, agreed with the district court that "numerical integration" referred to "an entire class of different possible algorithms used to perform integration." Accordingly, the disclosure did "not limit the scope of the claim to the 'corresponding structure, material, or acts' that perform the function, as required by section 112." That the concept of "numerical integration" was well known did not save the claims.
The Federal Circuit refused to consider the patent owner's argument that the "patent discloses a two-step algorithm that consists of sampling and accumulating," because this argument was not raised before the district court. However, one wonders whether this argument would have saved the day.
One could dismiss this case as narrowly applicable to claims drafted under 35 U.S.C. § 112, paragraph 6 (post AIA, 35 U.S.C. § 112(f)). Even so, and even if means-plus-function language is eschewed in the initial claim drafting process, one would do well to define claim terms just in case a means-plus-function claim needs support at some point. But more broadly, especially in light of Nautilus' loosening of the standard for indefiniteness, one could read this case as suggesting that claim terms be defined. For example, what if the disclosure of the '181 patent had referred to "numerical integration," and then given a short definition and/or said "examples of numerical integration algorithms include A, B, and C?" Even if a claim term is not governed by the specific strictures applied to means-plus-function language, particularly in today's environment, an undefined claim term risks, at best, an uncertain claim construction and, at worst, a holding of indefiniteness.