lightbulb gear

Indefiniteness has for decades enjoyed a reputation as the Rodney Dangerfield of patent defenses: It gets no respect. The U.S. Supreme Court’s decision to hear an appeal in Nautilus Inc. v. Biosig Instruments Inc. may attract new attention to indefiniteness. But regardless of the outcome, the defense is already catching on when applied to “means-plus-function” terms, especially with respect to computer-implemented inventions, which make up the lion’s share of business-method patents. 715 F.3d 891 (Fed. Cir. 2013), cert. granted, Docket No. 13-369 (U.S. Jan. 10, 2014).

Already, more than 70 percent of patents reviewed under the America Invents Act postgrant review reforms implicate electrical or computer patents. The ax most frequently raised to invalidate such patents is the patentable subject-matter challenge under 35 U.S.C. 101, with prior-art invalidation close behind. The standard for patentable subject-matter under Section 101 has enjoyed considerable attention in recent years, and is set to gain still more when the Supreme Court hears the latest in a recent string of Section 101 cases later this year. Meanwhile, lurking quietly in the claims of many of these patents is a defect that is garnering little attention; namely, indefiniteness due to unsupported means-plus-function claiming.

At its core, the law requires a patent to distinctly point out and particularly claim the invention. 35 U.S.C. 112(b). In Nautilus, the Supreme Court will grapple with whether claims must be proven “insolubly ambiguous” to be susceptible to invalidation on the basis of indefiniteness. Docket No. 13-369 (Jan. 10, 2014). This is because a claim will be indefinite if one skilled in the relevant art cannot determine the metes and bounds of the invention from the claim alone.

However, when a claim is drafted using means-plus-function language, the specification must explicate the limits of the claimed means. Essentially, means-plus-function claiming represents a statutorily permitted shortcut whereby a patent drafter can avoid reciting complicated details in a claim by instead describing a claim limitation as a means or step for performing a particular function. If the patent drafter chooses to take this route, the structure, material or acts for performing the recited function must be described in the specification or the patentee risks invalidation for failure to comply with 35 U.S.C. 112(b) and (f).

In the mid-1990s, the heightened standard associated with Section 112(f) was reinforced in an en banc U.S. Court of Appeal for the Federal Circuit decision. In re Donaldson Co., 16 F.3d 1189 (Fed. Cir. 1994). Donaldson resulted in fewer drafters risking “means for” language. In its place, drafters began using what the U.S. Patent and Trademark Office (PTO) dubs nonstructural terms like “system for” or “computer for” more often because there is a strong, though rebuttable, presumption that such language does not implicate Section 112(f). MPEP 2181.

The Federal Circuit and the newly formed Patent Trial and Appeal Board (PTAB) in the PTO have taken two approaches to the strict requirements of Section 112(f): Apply Section 112(f) and strike patent claims that fail to disclose adequate structure in the specification for means-plus-function claim limitations; or refuse to apply Section 112(f) and strike claims for failure to meet 112(b) particularity requirements.

For computer-implemented patents, which often use shorthand, the frequency of 112(f) defenses is likely to rise. By some estimates, millions of patent claims could be implicated, especially when nonstructural terms are included. The greatest challenge when it comes to computer-implemented patents is the analysis for the sufficiency of structure necessary to satisfy 112(f). In the case of a mechanical or chemical invention, the analysis of whether structure corresponding to a means-plus-function claim term is disclosed is generally straightforward. Computer-implemented inventions, however, often blur the line between structure and function.

The Federal Circuit addressed this problem in In re Katz Interactive Call Processing Patent Litigation, 639 F.3d 1303, 1316 (Fed. Cir. 2011), in which it clarified the rule that when a general-purpose computer is the structure identified to perform a special function, thus converting it into a special-purpose computer, there must be an algorithm disclosed for performing that function. However, when a general-purpose computer is used without any special function, there is no need for an algorithm because there is no need for special programming. See Noah Systems Inc. v. Intuit Inc., 675 F.3d 1302, 1312 n.8 (Fed. Cir. 2012). In Katz, the functions included means for processing, receiving and storing and only discussed a general-purpose computer as the structure capable of performing these functions. 639 F.3d at 1316.

When an algorithm is required, it may still be unclear whether one exists, because the Federal Circuit has provided limited guidance. While it is apparent that the disclosure of software is insufficient without providing some detail as to how the software performs the claimed function, the level of detail necessary is unclear. See Noah Systems, 675 F.3d at 1312. It is likely computer code is not necessary, but the broad allowance of “any understandable terms including as a mathematical formula, in prose, or as a flow chart, or in any other manner that provides sufficient structure,” may leave a practitioner scratching his or her head. Id. One reason for this is that the necessary level of disclosure is tied to the structures known in the art, which do not need to be disclosed. See Function Media LLC v. Google Inc., 708 F.3d 1310, 1318 (Fed. Cir. 2013).


Despite some petitioners’ insistence that the Federal Circuit misapplies 112, the PTAB has consistently followed the Federal Circuit’s guidance on the subject in regularly affirming examiner’s rejections of claims. See, e.g., Ex parte Sashikanth Chandrasekaran and Gary D. Young, 2013 WL 3323626, at *2 (PTAB May 23, 2013). For its part, the PTAB has shown little difficulty in applying the foregoing analyses to computer-implemented claims, and likewise has shown little reluctance to strike down means-plus-function terms on indefiniteness ground. See, e.g., Ex parte Jeff Bodner and Randy Bierwerth, 2014 WL 296409 (PTAB Jan. 23, 2014). The PTAB has also displayed a willingness to treat claims written as “computer for” or the like as means-plus-function terms, despite the presumption that a claim term lacking the word “means” is not subject to Section 112(f). E.g., Ex parte Aaron Smith, 2013 WL 1341109 (PTAB March 12, 2013).

There are many reasons the PTAB would be more agile with Section 112 analysis, including that PTO examiners are often those with the relevant level of skill in the art. See Ex parte Martin W. Masters, 2014 WL 174463, at *3 (PTAB Jan. 13, 2014). For instance, the PTAB cites computer dictionaries sua sponte to define an algorithm as “a finite sequence of steps for solving a logical or mathematical problem or performing a task.” MPEP 2181; see also Ex parte Bodner, 2014 WL 296409, at *2. The PTAB also is not hampered by a presumption of validity, instead applying a broadest-reasonable-interpretation standard that can be fairly forgiving.

Despite the PTO’s firm stance on enforcing the requirements of sections 112(b) and (f) in rejecting claims and reviewing such rejections, the PTAB has not canceled any claims through covered business method review on Section 112 grounds and has been reluctant to grant such review on these grounds. Instead, the focus of many covered business method reviews are on amorphous patentable subject-matter ground under Section 101. Notably, the review of abstract subject matter under Section 101 is not unlike the review of functional claiming under Section 112: Both grounds require that claims provide meaningful limitations to those skilled in the art. It would thus behoove a practitioner to raise both defenses, particularly before the PTAB and when computer-implemented patents are involved.

With the continuing confusion over Section 101 standards and the pervasive use of means-plus-function and similar nonstructural claiming in computer-implemented inventions, indefiniteness is poised to emerge as an essential tool for judging patentability. Both litigators and prosecuting patent attorneys should take care in assessing the use of means language and the adequacy of supporting algorithms. The simple enforcement of these already existing standards can strengthen the patent system by focusing on a central premise of patentability: notice.

Douglas Nemec is a partner and Hoda Rifai-Bashjawish is a law clerk in the patent and technology litigation group in the New York office of Skadden, Arps, Slate, Meagher & Flom.