If a claimed means requires programming to carry out a specified function, then a patent specification must disclose an algorithm to support the claimed means. Absent such disclosure, means-plus-function claim elements are indefinite, and render a patent claim invalid under 35 U.S.C. § 112, according to the court in Microsoft Corp. v. Motorola Inc., No. C10-1823 JLR (W.D. Wash. Feb. 7, 2013). The court’s opinion will likely get attention from many commentators as an interesting pro-Microsoft development in this long-running and high-profile patent litigation. This post simply summarizes the basis for the court’s indefiniteness holding, which, although not citing the Federal Circuit’s recent Eplus v. Lawson decision, amplifies but does not seem to deviate from that case or earlier precedent.
The patents-in-suit, which all rely on a common specification, are United States Patent Nos. 7,310,374, 7,310,375 and 7,310,376. These patents are directed, “at a high-level,” to “systems and methods for encoding and decoding a bitstream (or sequence) of digital video data.” Microsoft brought a summary judgment motion of invalidity directed at certain claims that included recitations of “means for decoding” and “means for using said plurality of decoded smaller portions to construct a decoded picture.”
Looking at the first step of means-plus-function claim construction, determining the claimed function, the court noted that the parties agreed that all “means for decoding” recitations claimed the function of decoding. Likewise, the parties agreed that the “means for using” recitations all identified the claimed function in the language following the “means for using” term. Further, concerning the second step of means-plus-function claim construction, the parties agreed that the structure corresponding to each of the recited means was a decoder disclosed in the specification. However, the parties disagreed about whether the decoder was adequately disclosed.
Recent Federal Circuit cases have clearly stated that means-plus-function claim recitations directed to a special-purpose computer must do more than disclose a general-purpose computer; these cases require disclosure of an algorithm. Motorola argued that these cases did not apply because here, Motorola asserted, the disclosure of a “decoder” would have been adequate for one skilled in the art.
Looking at the Federal Circuit cases, including Aristocrat Techs. Austl. Pty Ltd. v. Int’l Game Tech., 521 F.3d 1328 (Fed. Cir. 2008), Harris Corp. v. Ericsson Inc., 417 F.3d 1241 (Fed. Cir. 2005), and WMS Gaming, Inc. v. Int’l Game Tech., 184 F.3d 1339 (Fed. Cir. 1999), the court here concluded that “where the disclosed structure corresponding to a means-plus-function element is a computer-implemented algorithm, the algorithmic structure must be disclosed.” Although Motorola’s expert had submitted a declaration to support Motorola’s position, nowhere did that declaration state that additional programming of the decoder would not have been required. In contrast, Microsoft “provide[d] substantial evidence by way of citations to technical journals and dictionaries that each of the examples is nothing more than a computer chip that must be programmed or designed to perform the desired function.”
Further, analyzing each of the specific claim recitations at issue, the court found that the specification disclosed at most methods of encoding, but not decoding, video data. In this case, Motorola could not show that any disclosed structure could be used to meet the claim requirements without additional modification. Thus, this case was distinguishable from cases cited by Motorola because here “the decoder is defined broadly and functionally to mean anything that can decode.” The court concluded that “the “means for decoding” limitations claim all corresponding structure under the sun by expansively defining the function in the specification as anything that decodes digital data. This definition renders the ‘means for decoding’ limitation invalid for indefiniteness.”