October 2015
Intellectual Property & Technology Law Journal
By Doris Johnson Hines
Authored by Doris Johnson Hines and Andrew G. Strickland
The Federal Circuit recently issued an en banc1 holding in Williamson v. Citrix Online, LLC,2 (“Williamson”) that will likely affect the scope and validity of claims for tens of thousands of patents in a way unintended by the drafters of those claims. In short, Williamson broadens the number of patent claims falling under Section 112, paragraph 6 (Section 112(6)), or 35 U.S.C. § 112(f) since the September 2011 passing of the Leahy–Smith America Invents Act (AIA).
Section 112(6) allows inventors to functionally claim elements of their apparatus inventions. Using this format allows an inventor to claim a “means” or a “step” for performing a function without reciting the specific structure, material or acts that perform the function. As a simple example, consider a claim directed to shelf attached to a wall using an innovative screw-like part. An inventor might claim the shelf comprising “one or more screw-like parts,” or the inventor could use Section 112(6) and claim the shelf comprising “a means for attaching.”
While a claimed “means for attaching” appears broader than “one or more screw-like parts,” elements written pursuant to Section 112(6) are construed as “the corresponding structure, material, or acts described in the specification” and their equivalents. So, the “means for attaching” does not cover every conceivable structure for attaching the shelf to a wall, rather, it only covers the structures that attach the shelf to a wall disclosed in the specification and their equivalents.
Outside the context of Section 112(6), the law instructs courts not to limit claim language by the embodiments disclosed in the specification. In fact, reading limitations of the specification into the claims is considered a “cardinal sin” of patent law.3 This is not so for claim elements invoking Section 112(6), which explicitly requires construing those claim elements to the structure, material, or acts described in the specification. Because of this difference, whether Section 112(6) applies to a claim element can have a big impact on the scope of a claim. A term construed under Section 112(6) will likely have a narrower construction than one not construed under Section 112(6).
The implication of construing a term under Section 112(6) does not stop with claim scope. A construction under Section 112(6) can also open the door for a court to find the claim invalid as indefinite. In the Section 112(6) analysis, after the court determines the statute applies, the court turns to the specification to identify the disclosed structure, material, or acts described there and their equivalents. If the court cannot identify any corresponding structure, material, or acts in the specification, the claim term is indefinite under 35 U.S.C. § 112(b) (or paragraph 2 pre-AIA).4 This result makes sense—the scope of the claimed invention cannot be ascertained because the claim is limited to structure disclosed in the specification, yet the specification discloses no such structure.
Whether Section 112(6) applies to a claim term thus matters, and the Federal Circuit broadened the application of Section 112(6) in Williamson. After Williamson, courts will find more patent claims to fall under Section 112(6) than before. In fact, the withdrawn panel decision in Williamson previously found the claims to not invoke Section 112(6).5 It was not until after the en banc court articulated the new Williamson standard that the same panel found the claims to invoke Section 112(6).
While many patents may be impacted by the holding in Williamson, that impact will likely be strongest on software patents. Patent practitioners must carefully consider Williamson when drafting patent claims related to software and patent litigators should know Williamson’s impact when developing litigation strategy.
To understand why software patents may be more impacted by Williamson, consider the way courts interpret software patent claims under Section 112(6). Claims covering a device are directed to a tangible thing that inherently has structure. Using functional claim language is an alternative way to describe that structure by what it does as opposed to what it is. In the shelf described above, the screw-like part could be claimed structurally as a “screw,” or it could be claimed as a “means for attaching.” Software is a little different. While software can have structure in its design, it is typically thought of in terms of what it does as opposed to its structure. The inherent nature of software makes it more difficult to claim structural components.
To work around this difficulty, many early software patents claimed a “means for” performing the function of the software. A software claim directed to adding two values might recite a “means for receiving input values . . .” as an element. But, such a claim element must still have structural support in the specification under Section 112(6)—inventors are not free to claim a functional limitation without providing structural support. The proper construction of the “means for receiving input values . . .” is the structure, material, or acts described in the specification that perform that function. In the area of software, that could be the algorithm for performing the function.
While most software is executed using a processor or a computer, the Federal Circuit has consistently held that structural support in the specification for a computer-implemented claim invoking Section 112(6) must be more than a general purpose computer or microprocessor.6 Instead, the specification must disclose an algorithm for performing the claimed function. In a recent case, EON Corp. IP Holdings LLC, v. AT&T Mobility LLC, the Federal Circuit reiterated that the structure disclosed in the specification must be more than a general purpose computer or microprocessor and confirmed that the corresponding structure for a computer-implemented claim falling under Section 112(6) is the algorithm performing the claimed function.7 The specification can express the algorithm in any understandable way including as a mathematical formula, in prose, or as a flow chart, or in any other way that provides sufficient structure. Disclosing software, however, without providing detail about the way to accomplish the claimed function is not enough.8 In addition, “black box” disclosures are often too generic to provide corresponding structure for computer-implemented Section 112(6) limitations.9
A reason Section 112(6) can cause problems for software patents relates to how much a patentee can rely on the knowledge of one of ordinary skill in the art to support the claims. When a software claim does not fall under Section 112(6), a patentee can rely on the knowledge of one of ordinary skill to show that the claims are not indefinite.10 If, however, a court considers a claim element to fall under Section 112(6), the claim element must be supported in the specification by a specific algorithm and a patentee’s options to rebut an indefiniteness attack are more limited. The EON decision explains that in the context of a claim term invoking Section 112(6), when the specification discloses no algorithm, the skilled artisan’s knowledge is irrelevant, i.e., the knowledge of the skilled artisan cannot supplement the specification and provide structure when the specification provides none.11
When an inventor intends to invoke Section 112(6) in a software claim, it should come as no surprise that the specification must include adequate structural support in the form of an algorithm or other description of the way to perform the software function. An inventor has no grounds for complaint that a claim is invalid as indefinite for failing to have adequate structural or algorithmic support in the specification when intending to invoke Section 112(6). But, if the inventor does not intend to invoke Section 112(6), it can be an unpleasant surprise when a court determines that Section 112(6) applies to a claim element, determines the specification discloses no algorithm, and the knowledge of those skilled in the art is irrelevant. Since Williamson broadens the applicability of Section 112(6), such surprises may become more common.
While Williamson broadens when a court considers claim elements as falling under Section 112(6), its holding does not articulate a new rule of law. Rather, it rolls back a series of Federal Circuit decisions that strengthened the presumption that Section 112(6) did not apply when the word “means” was not used. And, that was the problem. Courts saw (or assumed) clever patent drafters were getting the benefit of functional claiming without Section 112(6)’s requirement of construing such functionality as that specifically disclosed in the specification.
Section 112(6) states that “[a]n element in a claim for a combination may be expressed as a means or step for performing a specified function.” Accordingly, a claim element’s use of the term “means” triggers a rebuttable presumption that Section 112(6) applies, and conversely, its absence triggers a rebuttal presumption that Section 112(6) does not apply.12 But, a claim element’s use of “means” is not dispositive; there are Federal Circuit decisions holding that Section 112(6) does not apply to claim elements including the word “means”13 and decisions in which Section 112(6) does apply even though the claim element does not include “means.”14 Use of the word “means” just introduces the presumption that Section 112(6) applies, but that presumption can be rebutted where the claim, in addition to any functional language, recites structure sufficient to perform the claimed function in its entirety.15 The question is whether the claim language recites particular structure or, instead, refers only to a general category of whatever may perform the specified functions. In the latter case, Section 112(6) applies.16
Before Williamson, a series of Federal Circuit decisions heightened the standard for determining whether the presumption associated with “means” was overcome, starting with Lighting World Inc. v. Birchwood Lighting, Inc.17 There, the Federal Circuit stated that the standard attached to the presumption was “a strong one that is not readily overcome.”18 The Federal Circuit raised the bar in Flo Heathcare Solutions, LLC v. Kappos, where it held that it was unwilling to apply Section 112(6) to a claim element unless that claim element is “essentially . . . devoid of anything that can be considered structure.”19 Later, in Apple Inc. v. Motorola, the Federal Circuit observed that the presumption is strong and not easily overcome, and noted that it seldom found Section 112(6) to apply to a claim limitation that did not recite “means.” Until Williamson, those arguing that Section 112(6) applied where a claim did not recite “means” faced a steep, uphill climb.
In Williamson, an en banc court returned the presumption to its pre-Lighting World strength, i.e., while a claim element’s recitation of “means” creates a presumption that courts construe the claim element under Section 112(6), that presumption is not a “strong one that is not readily overcome.” Instead, the Federal Circuit emphasized that “[w]hen a claim term lacks the word ‘means,’ the presumption can be overcome and § 112, ¶ 6 will apply if the challenger demonstrates that the claim term fails to ‘recite sufficiently definite structure’ or else recites ‘function without reciting sufficient structure for performing that function’” when viewed by persons of ordinary skill in the art.20 The court also explicitly overruled Flo Healthcare’s “strict requirement of ‘a showing that the limitation essentially is devoid of anything that can be construed as structure.’”21
Applying the holding of the en banc court, a panel of Judges Moore, Linn, and Reyna, first affirmed the district court’s holding that the claim element “distributed learning control module for receiving communications ... for relaying the communications ... and for coordinating the operation of the streaming data module” is subject to Section 112(6). Since the “distributed learning control module” claim element does not include the word “means,” the Federal Circuit recognized the presumption that that Section 112(6) did not apply, but found that the presumption was overcome. The “distributed learning control module” claim element was drafted in a format consistent with Section 112(6) claiming—it merely replaced the word “means” with “module,” followed by the functions the module performs. Further, since the specification and file history provided no support that a module, or more specifically a distributed learning control module, has sufficiently definite structure, the court held that Section 112(6) applied. After doing so, it affirmed the holding of the district court that the claim was invalid because the specification failed to disclose structure corresponding to the functions of the distributed control module.
Williamson’s biggest impact on software patents may be from the panel’s characterization of “module.” Divorced from the other claim language, the court stated that “module” is a well-known “nonce word” that operates as a substitute for means and was a generic term, similar to “mechanism,” “element,” and “device,” which rarely connotes structure. In particular, the word “module,” the Federal Circuit stated, “sets forth the same black box recitation of structure for providing the same specified function as if the term ‘means’ had been used.”22
Interestingly, in the withdrawn panel decision, Judges Moore, Linn, and Reyna previously considered whether the “distributed control module” was governed by Section 112(6), and Judges Moore and Linn arrived at a different result.23 That earlier result was based on the Lighting World “strong presumption” standard and the Flo Healthcare “devoid of structure” analysis. In addition, Judges Moore and Linn arrived at their earlier decision by considering technical dictionaries not cited by the parties to conclude that a “module” had “understood dictionary meanings as connoting either hardware or software structure to those skilled in the computer arts.”24 They also stated that in light of the specification, a “distributed control module,” connoted structure. The specification described that the distributed control module was part of a distributed learning server, and the “interconnections and intercommunications” between it and other systems and modules, connoted structure to one of ordinary skill in the art.” Judges Moore and Linn further stated that “[w]hile the supporting specification describes the claimed expression ‘distributed learning control module’ in a high degree of generality, in some respects using functional expressions, it is difficult to conclude that it is devoid of structure.”25
The phrase “module for” appears in tens of thousands of issued patents.26 While not all claim elements with this phrase will be construed to fall under Section 112(6), many more will after Williamson. In some cases, the result will be similar to Williamson. Some patentees will find that claims drafted without the intent of invoking Section 112(6) will have that section applied to their claims, and the claim found invalid for lacking structural support in the specification. This result will be especially troubling for patentees who drafted claims reciting a “module” after Lighting World, but before Williamson. In any event, Williamson has opened the door to increased challenges based on Section 112(6) in the area of computer-related patents.
Williamson provides some important lessons for those drafting claims directed to software inventions and also to litigators asserting, or defending against, software patent claims. Through an understanding of Williamson’s holding, and with careful drafting, patent practitioners can minimize the risk that a court will apply Section 112(6) to their claims when it was not intended. Meanwhile, litigators should understand the impact Williamson has on the applicability of Section 112(6) so they can effectively develop invalidity and claim construction strategy.
Perhaps Williamson’s most obvious lesson is that patent drafters should draft claims in a way that avoids Section 112(6) if they do not intend for the section to apply. Drafters should avoid words like “device,” “module,” “component” and other terms indicative of a black box. For non-method claims directed to software inventions, one solution might be to claim a processor executing instructions that when executed performs a function of what would have been claimed as a module. Similarly, litigators should be on the lookout for claim elements reciting “means for” substitutes followed by a recitation of function.
Another approach for patent drafters seeking to potentially avoid Section 112(6) is leveraging claim differentiation by explicitly include claims reciting a “means for” performing the functions of the software invention. By doing so, patentees will at least have an argument that by choosing “means for” in one claim, there was no intent to invoke Section 112(6) in another. This strategy has the additional advantage of providing a failsafe: by choosing to include a claim with “means for” elements, the drafter should be forced to consider the structural support for those elements and add it to the specification, if needed. So, even if a court applies Section 112(6) to claims not drafted with “means for” elements, the claims containing those Section 112(6) elements have a better chance of surviving an invalidity challenge because they will have specification support. Similarly, litigators should consider the entire claim set, including non-asserted claims, to determine the viability of an argument that Section 112(6) will apply to a claim element not reciting a “means for” performing a claimed function.
Patent drafters should also consider that for claims not reciting a “means for” performing a function, the claim language alone does not necessarily answer whether Section 112(6) applies. The Federal Circuit has frequently reviewed the specification and file history of a patent to determine whether a challenger to a claim has overcome the presumption that Section 112(6) does not apply to a claim term.27 When the intrinsic record conveys structure for the term allegedly serving as a “means for” substitute in a claim element, the presumption of Section 112(6) inapplicability is not overcome. Sound drafting practice includes providing a structural description of claimed “modules,” “devices,” “units,” “mechanisms” and the like in the specification. For litigators, the lesson learned is similar—when assessing the strength of an argument applying Section 112(6) to a claim term that does not recite a “means for” performing a function, the specification’s description of the element performing that function should be considered.
A court may still find that Section 112(6) applies to a claim term even though the inventor or drafter did not intend that. Effective drafting of patents related to software, therefore, should include the assumption that any “module for,” “component for,” “logic for,” etc. claim element could be construed as a non-structural term, and the specification should be drafted accordingly. Since for software inventions the algorithm performing the function is the structure, patent drafters should include as much of a description of the algorithms performing the claimed functions as possible. Good practice might include preparing a flow chart and associated description for each element of the claim so there is algorithmic, structural support for each claim element. A related practice is to avoid an empty “black box” for a claimed function and to describe the functions within the black box as much as possible and with as many alternatives as possible. For example, when a claim recites “a module for calculating inputs,” a simple black box in a diagram called “input calculator,” may not provide sufficient structural support for the “module for calculating input.” Even without the threat of Section 112(6) application, drafters of software patents should err on the side including more flowcharts and descriptions of algorithms.
While Williamson is no doubt an important case for software patents, it does not set out brand new law. Rather, it rolls back holdings concerning the presumption attached to Section 112(6) applicability. While Williamson’s impact may be significant now and lead to some uncertainty with claims that issued after Lighting World and before Williamson, the impact may be felt less as time goes on and patent drafters employ best practices to minimize the risk of inadvertent applicability of Section 112(6). However, drafting best practices for handling Williamson’s holding may boil down to traditional patent drafting best practices: avoid “nonce” words in claims, provide detailed description of the invention in the specification, and leverage claim differentiation.
Endnotes
1 The Federal Circuit sat en banc only for the portion of the opinion regarding the applicability of Section 112(6), Part II.C.1.
2 No. 13-1130 (Fed. Cir. June 16, 2015).
3See e.g., SciMed Life Systems, Inc. v. Advanced Cardiovascular Systems, Inc., 242 F.3d 1337,1340 (Fed. Cir. 2001).
4 See e.g., Bosch v. Snap-on, Inc., 769 F.3d 1094, 1097 (Fed. Cir. 2014).
5 Williamson v. Citrix Online, LLC, 770 F.3d 1371, 1378 (Fed. Cir. 2014) (Williamson (withdrawn)).
6 See e.g., Noah Sys., Inc. v. Intuit, Inc., 675 F.3d 1302, 1312 (Fed. Cir. 2012).
7 EON Corp. IP Holdings LLC, v. AT&T Mobility LLC, 785 F.3d 616, 621 (Fed. Cir. May 6, 2015), citing WMS Gaming, Inc. v. Int'l Game Tech., 184 F.3d 1339, 1348-49 (Fed. Cir. 1999).
8 Noah Sys. 675 F.3d at 1312.
9 TecSec, Inc. v. Int'l Bus. Machines Corp., 731 F.3d 1336, 1349 (Fed. Cir. 2013).
10 See Nautilus, Inc. v. Biosig Instruments, Inc., 134 S.Ct. 2120, 2129 (2014).
11 EON Corp., 785 F.3d at 621.
12 Bosch v. Snap-on, Inc., 769 F.3d 1094, 1097 (Fed. Cir. 2014).
13See e.g., Phillips v. AWH Corp., 415 F.3d 1303, 1311 (Fed. Cir. 2005) (“means disposed inside the shell for increasing its load bearing capacity” not a Section 112(6) claim term)
14 See Bosch, 769 F.3d at 1099 (Fed. Cir. 2014).
15 Altiris, Inc. v. Symantec Corp., 318 F.3d 1363, 1375 (Fed. Cir. 2003).
16 Bosch, 769 F.3d at 1099.
17 382 F.3d 1354 (Fed. Cir. 2004).
18 Id. at 1358.
19 697 F.3d 1367, 1374 (Fed. Cir. 2012).
20 Williamson Slip Op. at 16.
21 Id.
22 Id. at 18.
23 See Williamson (withdrawn), 770 F.3d at 1378.
24 See id. at 1379.
25 See id. at 1380.
26 As of August 8, 2015, a search in the Thompson Innovation database for the term “module for” in the claims of issued patents yields 42,396 results.
27 See e.g., Inventio AG v. ThyssenKrupp Elevator Ams. Corp., 649 F.3d 1350, 1357 (Fed. Cir. 2011).
Originally printed in Intellectual Property & Technology Law Journal. This article is for informational purposes, is not intended to constitute legal advice, and may be considered advertising under applicable state laws. This article is only the opinion of the authors and is not attributable to Finnegan, Henderson, Farabow, Garrett & Dunner, LLP, or the firm's clients.
June 10-12, 2024
San Francisco
Lecture
Patent Protection for Software-Related Inventions in Europe and the USA Training Course
June 5, 2024
Hybrid
10th Annual Georgia Asian Pacific American Bar Association Gala
May 29, 2024
Atlanta
Due to international data regulations, we’ve updated our privacy policy. Click here to read our privacy policy in full.
We use cookies on this website to provide you with the best user experience. By accepting cookies, you agree to our use of cookies. Please note that if you opt not to accept or if you disable cookies, the “Your Finnegan” feature on this website will be disabled as well. For more information on how we use cookies, please see our Privacy Policy.
Finnegan is thrilled to announce the launch of our new blog, Ad Law Buzz, devoted solely to breaking news, developments, trends, and analysis in advertising law.