What Changes After Fed. Circ.’s Williamson Decision
Posted on Jul 10, 2015 in Articles
This article originally appeared in Law360 on July 10, 2015.
When the Federal Circuit issued the en banc decision in Williamson v. Citrix [1], the court changed the guidelines regarding the adjudication of patent claims with functional claim language. Many blogs, articles and newsfeeds embraced the decision as yet another positive step towards curbing frivolous litigation based on overly broad software patents. Only time will tell if Williamson accomplishes that goal. The decision, however, is interesting because it refines an important test for interpreting functional claim language, while at the same time it preserves aspects of the precedent.
Williamson was originally decided by a three judge panel of the Federal Circuit (hereafter Williamson I). At issue was the question of whether the phrase “distributed learning control module” invoked 35 U.S.C. 112(6) [2] [3]. Appellees argued that the phrase was drafted in the same format as a means-plus-function limitation (which thus invoked 112(6)), and since the patent specification did not disclose an algorithmic structure for implementing the claimed function, the claim was therefore indefinite.
The majority in Williamson I concluded, based on the lack of the word “means” in the claim, that 112(6) did not apply: “[F]ailure to use the word ‘means’ in a claim limitation created a rebuttable presumption that 35 U.S.C. 112, para. 6 did not apply … This presumption is “a strong one that is not readily overcome.” Citing Lighting World Inc. v. Birchwood Lighting Inc., 382 F. 3d 1354 (Fed. Cir. 2004).
Judge Jimmie Reyna wrote a strong dissent for Williamson I in which he argued:
The claim language explains what the functions are, but does not disclose how the functions are performed ….[T]he term “module” [that appears in the claims] is “a “nonce” word…a generic word inherently devoid of structure … that can operate as a substitute for “means” in the context of 112, paragraph 6 … The following is a list of non-structural generic placeholders that may invoke … 35 U.S.C. 112, paragraph 6: “mechanism for,” “module for,” “device for,” “unit for,” “component for,” “element for,” “member for,” “apparatus for,” “machine for,” or “system for.”
Therefore, he argued, 112(6) did apply to the claim, and, since the specification did not disclose sufficient structure corresponding to the claimed function, the claim should have been held invalid as indefinite.
Evidently, the en banc court ultimately agreed with Judge Reyna’s dissent. In the Williamson en banc decision (hereafter Williamson II), Williamson I was reversed as the court concluded that “we should abandon characterizing as ‘strong’ the presumption that a limitation lacking the word ‘means’ is not subject to 112, para. 6” and that “[t]he standard is whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure.”
Even though Williamson II has been described as changing the guidelines for interpreting functional claim language, Williamson II is also consistent with aspects of prior case law.
Apple Inc. v. Motorola Inc., 757 F.3d 1286 (Fed. Cir. 2014) (hereafter “Apple”) was decided seven months before Williamson I. Apple came to the Federal Circuit on appeal after the district asserted that (1) the word “means” was not in the claim; (2) 112(6) was applicable to the claim even though the word “means” was not in the claim; (3) the claim should have been interpreted based on an embodiment disclosed in the patent specification because 112(6) was applicable to the claim; and (4) based on the interpretation of the claim there was no infringement. The Federal Circuit reversed the district court and concluded that 112(6) was not applicable to the claim, and therefore an interpretation of the claim based on a disclosed embodiment was incorrect. Summary judgment of noninfringement was thus vacated for one of the asserted patents.
Of notable interest is the fact that one and a half pages of the Federal Circuit’s opinion cited examples from Apple’s patent describing the structure that performed the claimed feature. Not only did the decision provide details, column numbers and examples from Apple’s patent, but three figures from Apple’s patent were reproduced. The court determined that 112(6) should not be invoked because the claim recited a term “with a structural definition that is either provided in the specification or generally known in the art.”
Robert Bosch LLC v. Snap-On Inc., 769 F.3d 1094 (Fed. Cir. 2014) (hereafter “Bosch”) was decided one month before Williamson I. Like Apple, the word “means” did not appear in the claims. The Federal Circuit decided, however, that 112(6) did apply to the claim. Notable in this case was the court’s observation that the patent specification “does not contain a single reference to the structure of the [device] itself” and “one of ordinary skill in the art could not find in the specification a definition of the terms as referring to a particular structure.” The patent was found to be invalid.
Williamson I can be difficult to reconcile with Apple and Bosch for a number of reasons. Williamson I acknowledged that the description of the claimed expression in the patent specification was with “a high degree of generality,” and the court needed to look at dictionaries to determine whether terms in the patent denoted structure. Surprisingly, Williamson I concluded that the patent in suit, with functional language and “a high degree of generality,” did not invoke 112(6). In fact, Judge Reyna’s dissent specifically cited Bosch for its holding that claim language invokes 112(6) when it fails to recite specific structure. In some ways, Williamson II is more consistent with Apple and Bosch than Williamson I. The table below summarizes how these cases compare:
The main problem with the “strong” presumption standard of Williamson I (i.e., that 112(6) did not apply when “means” was absent from the claims) was the vagueness of the word “strong.” Williamson II acknowledged this very point: “[C]haracterizing as ‘strong’ the presumption that a limitation lacking the word “means” is not subject to 112, para. 6 … is UNCERTAIN IN MEANING AND APPLICATION” (emphasis added). Thus, while the “strong” presumption yielded results in Apple and Bosch that on their faces seemed appropriate, the presumption resulted in a finding of a valid claim in Williamson I despite functional claim language and a supporting specification that (in even the majority’s opinion) described the claimed expression “in a high degree of generality.”
Finally, the standard went from a “strong” presumption against invoking 112(6) if the word “means” was not present (Williamson I), to “whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure” (Williamson II). Evidence that claim language denotes structure to one of ordinary skill in the art, however, could always have been a “strong” argument so that the old presumption was not overcome; prior case law supports that conclusion. In Apple, for example, the Federal Circuit concluded that the proper inquiry is “whether the limitation, read in light of the remaining claim language, specification, prosecution history, and relevant extrinsic evidence, has sufficiently definite structure to a person of ordinary skill in the art.”
The Williamson II decision changes the articulated test of whether 112(6) applies, but the decision seems more motivated by a questionable interpretation (in Williamson I) of the word “strong” (as evidenced by Judge Reyna’s dissent) than other precedent. The decision, however, should be helpful. Now, instead of arguing over whether a “strong” presumption has been overcome to determine if 112(6) applies, it will be necessary to provide evidence based on the understanding of persons of ordinary skill in the art. It is unknown whether Williamson II will increase the number of software patents that will be held invalid, but the guidelines to make that decision are now certainly more precise.
[1] Williamson v. Citrix Online LLC, 2015 U.S. App. LEXIS 10082 (Fed. Cir. June 16, 2015),
[2] 35 U.S.C. 112(6): An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
[3] The America Invents Act replaced 35 U.S.C. 112(6) with 35 U.S.C. 112(f). The two paragraphs are identical.
– by Lawrence Ashery
