The Status of Software Patents in the Post-Alice Era

Nathan Hakimi | November 20, 2016

The Post-Alice Era

Since the 2014 Supreme Court decision of Alice Corp. v. CLS Bank,1 which will not extensively be discussed in and of itself herein, commentators have argued whether Alice was a good decision, and whether, in its wake, software patents are now a thing of the past.2 Yet, as will be discussed, litigation on the issue is still very much alive, as is public commentary, demonstrating that Alice hardly eliminated software patenting, but merely clarified it.3

In the post-Alice era, cases from the lower courts are necessary to clarify the Supreme Court’s ruling. Two contrasting cases from the Federal Circuit will demonstrate that, far from dead, software patents are clearly permissible, merely limited by a specific (and obvious) constraint. As will be shown, in spite of pessimistic hullabaloo, the lower courts are remaining true to Alice‘s fundamental proposition that software is patentable, provided that the underlying process or function being computerized is sufficiently specific and novel itself, whereas, naturally, merely computerizing a known process does not yield patentable technology.

Following a brief introduction laying out the historical and theoretical background relating to software patents, these two Federal Circuit cases will be explained and their implications synthesized.

Software Patents: A “Conceptual” Dilemma

Computer programs became a known technology in the first half of the 20th century,5 and questions regarding their patentability followed not long thereafter. The main issue underlying the difficulty of patenting software is in the nature of software itself. A computer program may come in a box or (once upon a time) a CD, but on its own the program is itself an “abstract” entity. The program itself has no physical components to construct or manufacture whatsoever. Hence the distinction between “hardware” and “software.” Programs are called “software” because they contain no tangible or physical components. Code is literally just language, i.e. “programming language” such as C or Java, which can be written or typed out. A program is thus, essentially, considered just a set of instructions for a machine to carry out a process.4

And so the nature of software interacts with the nature of law, wherein traditional structures concerning patents generally dictate that Sec. 101 of the patent code7 precludes “abstract ideas” from “patent-eligible” status.8 The fact that “code” does nothing more than to make explicit the steps of a preconceived procedure in a given language,4 10 makes it seem like code is itself always an intrinsically abstract thing and therefore inherently unpatentable. Unlike physical inventions which are fundamentally transformed in the transition from blueprint to prototype – and in fact a reasonable pathway to physical manifestation is requisite to patentability – code achieves virtually no new ontological status from description to implementation because by definition code is a form of description anyway.11

Therefore, the earliest Supreme Court case on the subject, Gottschalk v. Benson, dating back to 1972, seems to have made the bleak suggestion that an “algorithm,” which is merely the expression of a “mathematical formula,” cannot and would never be patentable.6

So, seemingly, pursuant to basic logic and to Supreme Court precedent going back several decades, a program can never be patented, due to the definition of patent-eligible subject matter and the nature of a program, vis a vis “abstract ideas.”

This problem was apparently drastic enough that a former USPTO director went so far as to call for the repeal of Section 101 entirely (the patent code forbidding “abstract ideas”) in order to fully open the doorway for patenting code.12 In this leader’s opinion, such a move would as a matter of policy better facilitate American biotech, logistics, cybersecurity, and other types of innovation which rely on software, which would in turn help us compete with competitor systems in Asia and Europe who are hard at work developing the same technologies and apparently better protecting their innovators in this regard.12

Conceptual Problems Require Conceptual Solutions

But, as we shall see, it didn’t take such a drastic step to solve this legal/policy problem, since the fact that we find ourselves in a “digital era,” where most of the technology we use every day consists of no more than code, is not lost on (most) jurists. The judges, policymakers, media, and public all know and take notice and a ready solution is not far behind.13

So, far from a “death knell” to software patents (in spite of the language of one of its interpreting jurists) (see below), the Supreme Court in Alice actually clarified and mitigated Gottschalk, by holding that Gottschalk actually did not foreclose the possibility that “applications of [abstract] concepts ‘to a new and useful end,’ [may] remain eligible for patent protection.”14 Essentially, Alice stands for the proposition software is not all necessarily unpatentable, but rather that merely taking a process which is already well-known or obvious, and implementing that process for the first time in the form of a computer program, does not make the software automatically patent-worthy.9

This is such a natural and obvious conclusion that it seems it shouldn’t have taken a unanimous panel of our nine greatest jurists to determine it. But, the beauty of it is in that in disguising their decision in this commonsense and perfectly agreeable principle, this generation’s Supreme Court was able to avoid some of Gottschalk’s more unsavory (if not outright primitive) axioms without explicitly “overturning” or rejecting that vintage opinion; rather, they formulated the outcome in the form of the ever-famous rhetorical tool known well to all law students as a “distinction.”

Therefore, Alice did nothing other than rescue software patents from their precarious stature. Yet the unenlightened public has reacted to this glorious opinion with nothing short of a hysteria.2 Granted, the proof is in the pudding so the implementation must remain to be seen in the lower courts. Thus, evidence for this optimistic conclusion will be garnered from two recent Federal Circuit rulings that have come down since Alice, both in September this year. They will be referred to herein as “Symantec” and “Bandai.”15 These cases came out with opposite but consistent conclusions, thereby illustrating the contrast between software that is patentable and software that isn’t, and demonstrating the bottom-line that software is patentable.

Intellectual Ventures v. Symantec (“Symantec”)       

In Symantec, at issue were three (3) patents dating back to 1997, which had been purchased by Plaintiff Intellectual Ventures (an infamous “patent troll”16). Intellectual Ventures claimed infringement by defendant Symantec (maker of Norton antivirus) of their newly-acquired patents.17 Two of the patents were for email junk-filtering and virus screening, respectively.17

In the District Court of Delaware, infringement claims on these two patents were successfully carried to verdict. After the verdicts came back, Defendant Symantec moved the judge to strike both patents as containing ineligible subject matter pursuant to § 101.18 They successfully struck down the email filtering patent, but the judge upheld the virus screening patent.19 Intellectual Ventures appealed the striking of one patent, and Symantec cross-appealed the upholding of the other.

The Court of Appeals for the Federal Circuit ultimately held that both patents were invalid, as containing ineligible subject matter pursuant to §101 and Alice.20 Just as Alice had held that software automating a well-established finance-sector procedure was invalid, so too, email-filtering and virus-screening were not novel enough functions to create patent-eligibility for software code implementing them on a computer.21

We see in Symantec a software patent getting defeated, but that doesn’t make Symantec a negative harbinger for software patentability in general. Rather, it is a clear and easy application of Alice which holds essentially that the particular functions being performed by the software at issue was not itself novel, therefore the transference of that function to code was not patentable either.

Nonetheless, the opinion raised some worried eyebrows,22 perhaps primarily due to a fiery concurrence written by Judge Haldane Mayer. In it he stated that “this court [should] acknowledge that Alice sounded the death knell for software patents.”23 Despite various fretting in the media over this statement, some even going so far as to call for Judge Mayer’s resignation,24 his opinion in concurrence is actually harmless (if not nonsensical). It makes the rather inventive claim that computer programming is a form of linguistic expression deserving of First Amendment protection, and therefore is de facto undeserving of patent protection.23 This is a unique view which cannot be said to represent either the view of the majority of the Federal Circuit in Symantec, nor the Supreme Court in Alice, nor that of the community at large. So, florid as it was, it was a legally and logically toothless concurring assertion.

So, all things said and done, it can be fairly said that while Symantec dealt a blow to a specific piece of software, it did not injure the status of software in terms of potential patentability. Rather, it stayed true to the precedent of Alice, holding simply that virus protection and spam filtering and not novel and patentable functions – not that all software is necessarily unpatentable because it is too “abstract.” As we will see from the next case, the Federal Circuit has certainly not abandoned software patents in the wake of Alice altogether.

McRO v. Bandai (“Bandai”)

Indeed, in the very the same month that the Federal Circuit brought down Symantec, it upheld a software patent in McRO v. Bandai, much to the interested community’s relief, to the extent of the opinion being praised as having “breathed new life” into the issue of software patentability.25

In Bandai, at stake were patents on a particular function of a video game. The code “automatically animated lip synchronization and facial expressions of three-dimensional animated characters.”26   The District Court for the Central District of California had struck down the patents on a §101 motion. The Federal Circuit reversed.

The Federal Circuit analyzed the patents extensively, detailing many aspects of the technology involved.27 Just to give a broad overview, the claimed technology involved replication of facial motions during speech of animated 3D characters.28 It did this by employing a system called “morph models”. This itself employed such mathematical structures as “vertices,” “delta set of vectors,” “morph weights,” “keyframes,” and “time-aligned phonetic transcription” to essentially teach a computer how to model and portray any given “phoneme” of human speech in a 3D animation of a talking face.29

The defendants had argued before the District Court that the mapping of phonemes to facial motions was a “’preexisting fundamental truth’ about how a human mouth looks while speaking,” and hence was ineligible subject matter pursuant to §101.30 The District Court had actually initially disagreed, holding that the subject matter itself was patentable.31 The only problem it found was that  “the novel portions of [the] invention [had been] claimed too broadly.”31 In other words, the idea was patent-eligible subject matter, but it was not a specific enough implementation to render the software patent-worthy in the second step of the analysis.

The Circuit Court, on de novo review, concurred that the software carried out a novel task. It also took the extra step in finding that the software’s methods were in fact “specific” enough to render them patent-eligible.32 Citing Alice, the Court opined that “the automation goes beyond merely ‘organizing [existing] information into a new form’ or carrying out a fundamental economic practice.”33 And so it upheld the patent.

It should be noted that Bandai is not the only case from the Federal Circuit or in many District Courts where software patents have been upheld. It has been brought here only as a recent representative example, standing for this article’s proposition that the potential of software patenting is hardly a lost cause in spite of histrionics to the contrary.

Conclusion

Bandai and Symantec represent two conflicting sides of the current software patent jurisprudence: the mere implementation of a known process on software is not patentable. This is evidenced by the holding in Symantec. On the other hand, if an idea is genuinely novel, but the only actual manifestation of that concept is through software implementation, that concept can still be patented even though the implementation is arguably inherently “abstract.” Therefore, despite various flurries over the so-called controversy regarding software patents in the media and in the lay public’s opinion,3 ultimately the bottom-line in the post-Alice world is that legally speaking, software is patentable.

In truth, it should be obvious that software is patent-eligible from the veritable bevy of actual companies who do in fact to this day hold valid and active patents on software which have not been challenged or whose challenges have been literally denied: see, for example, patents on “color correction” functions in Adobe Photoshop,34 a plethora of patents on various functions in Apple’s video editor Final Cut Pro,35 and numerous patents on various video games besides the ones addressed in Bandai.36 These are all active and enforceable patents and software. Software patenting is hardly a lost cause.

Yet, the practical reality for a would-be inventor or applicant may still unfortunately be a little more precarious. The Federal Circuit, at least, is on the same page as the – correctly interpreted – Supreme Court. Some suggest, however, that the Patent Office is at odds sometimes with the Federal Circuit on this issue. So the road for an actual applicant may be rockier than it should. The Supreme Court has for some reason declined to take up another appeal on the issue yet to bring down further guidance, probably because they don’t understand why Alice was so unclear. Congress has apparently promised to take some sort of action to address the issue but it hasn’t. There are many good blogs out there with instructions for how to make your software more patent-worthy at the USPTO realistically. Suffice it to say, at a purely legal level, it’s not a foregone conclusion.

 

[1] Alice Corp. Pty Ltd. v. CLS Bank Int’l, 134 S.Ct. 2347 (2014).

[2] Michael Borella, Section 101 and the Growing Alice Backlash, Patent Docs (May 15, 2016), http://www.patentdocs.org/2016/05/section-101-and-the-growing-alice-backlash.html.

[3] See, e.g., John Conley, Are Software Patents Dead?—Alice’s Implications for Life Sciences, Genomics Law Report, (July 21, 2015) http://www.genomicslawreport.com/index.php/2015/07/21/are-software-patents-dead-implications-for-life-sciences/; Jason Sanders, Software Patents are Not Dead, Enterprise Technology (November 20, 2015), https://mabr.com/software-patents-are-not-dead/; Courts and Patent Litigants Continue to Wrestle With Scope of Supreme Court’s Alice Test, M&R Blog (July 11, 2016), http://www.mrllp.com/blog-courts-and-patent-litigants-continue-to-wrestle-with-scope-of-supreme-court-s-alice-test; Robert R. Sachs, Two Years After Alice: A Survey of the Impact of a “Minor Case”, Bilski Blog (June 16, 2016), http://www.bilskiblog.com/blog/2016/06/two-years-after-alice-a-survey-of-the-impact-of-a-minor-case.html; The Elusive Meaning and Impact of Alice, Technology Academics Policy (April 28, 2016) https://www.techpolicy.com/Blog/Featured-Blog-Post/Elusive-Meaning-and-Impact-of-Alice,-The.aspx.

[4] Definition: “Program,” Techtarget (March 2007), http://searchsoftwarequality.techtarget.com/definition/program.

[5] Timeline of Computer History, Computer History Museum (2016), http://www.computerhistory.org/timeline/computers/.

[6] Gottschalk v. Benson, 409 U.S. 63 (1972) (holding that a “process claim directed to a numerical algorithm,” was not patentable because “the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.”).

[7] 35 U.S.C. § 101.

[8] Alice Corp., 134 S.Ct. at 2347 (citing Mayo Collaborative Servs. v. Prometheus Labs., Inc., 132 S.Ct. 1289, 1293 (2012)).

[9] Alice, supra, at 2357 (quoting Bilski v. Kappos, 561 U.S. 593, 611 (2010).

[10] See B. Jack Copeland, Diane Proudfoot, Alan Turing: Father of the Modern Computer, The Rutherford Journal (2001), http://www.rutherfordjournal.org/article040101.html.

[11] Meaning, the code written out in any given programming language is still essentially the same “type of thing” (i.e., ontological category) as the abstract description outlining the logical steps such code would take. See The Philosophy of Computer Science, Stanford Encyclopedia Of Philosophy, (Aug 20, 2013) http://plato.stanford.edu/entries/computer-science/, (esp. Sec. 5, “The Ontology of Programs”).

[12] Dave Kappos Calls for Abolition of §101, Nat’l Law Review (April 14, 2016), http://www.natlawreview.com/article/dave-kappos-calls-abolition-section-101.

[13] Steven Shaw, Why we need software patents, and yes, I’m smarter than you, zdnet (June 13, 2012), http://www.zdnet.com/article/why-we-need-software-patents-and-yes-im-smarter-than-you/.

[14] Alice, 134 S.Ct. at 2354 (quoting Gottschalk, 409 U. S. at 67).

[15] Intellectual Ventures I LLC v. Symantec Corp., No. 2015-1769, 2016 WL 5539870 (Fed. Cir. Sept. 30, 2016); McRO, Inc. v. Bandai Namco Games Am. Inc., No. 2015-1080, 2016 WL 4896481 (Fed. Cir. Sept. 13, 2016).

[16] Daniel Nazer, Mega-Troll Intellectual Ventures Hits Florist With Do-It-On-A-Computer Scheduling Patent, Electronic Frontier Foundation (March 31, 2016) https://www.eff.org/deeplinks/2016/03/stupid-patent-month-mega-troll-intellectual-ventures-hits-florist-do-it-computer.

[17] Intellectual Ventures, 2016 WL 5539870, at *1.

[18] Id.

[19] Id.

[20] Id at *10-11.

[21] Id at *4.

[22] Jeff John Roberts, Here’s Why Software Patents Are in Peril After the Intellectual Ventures Ruling, Fortune (October 3, 2016), http://fortune.com/2016/10/03/software-patents/.

[23] Intellectual Ventures, 2016 WL 5539870, at *13 (J. Mayer, Concurring).

[24] Gene Quinn, It is time for Judge Mayer to Step Down from the Federal Circuit, IP Watchdog (October 6, 2016) http://www.ipwatchdog.com/2016/10/06/judge-mayer-step-down-federal-circuit/id=73567/.

[25] Michael P. Visconti, III, MCro Latest Federal Circuit § 101 Decision Breathes New Life into Software Patents, IP Law Bulletin (September 19, 2016) http://www.iplawbulletin.com/2016/09/mcro-v-bandai-latest-federal-circuit-%C2%A7-101-decision-breathes-new-life-into-software-patents/ (“McRO joins a growing list of Federal Circuit cases that find computer-implemented inventions to be non-abstract, including DDR HoldingsEnfish, and BASCOM.”).

[26] McRO,., supra, 2016 WL 4896481 at *1.

[27] Id at *1 – 3.

[28] Id at *1.

[29] Id at *1-3.

[30] McRO, Inc. v. Sony Computer Entm’t Am., LLC, 55 F. Supp. 3d 1214 (C.D. Cal. 2014), rev’d and remanded sub nom McRO, Inc., 2016 WL 4896481  (Fed. Cir. 2016). [Henceforth McRO I and McRO II, respectively].

[31] McRO II at *4, quoting McRO I at 1230.

[32] McRO I at *10.

[33] Id at *9 (citing Digitech Image Techs., LLC v. Elecs. for Imaging, Inc., 758 F.3d 1344, 1351 (Fed. Cir. 2014) and Alice, 134 S.Ct. at 2356).

[34] Peter Kang, Jury Clears Adobe In Photoshop Patent Row With Everyscape, Law360 (January 23, 2015) http://www.law360.com/articles/614701/jury-clears-adobe-in-photoshop-patent-row-with-everyscape (regarding patents for “color correction” in Adobe Photoshop).

[35] Alex Gollner, Apple’s ‘Magnetic Timeline’ Final Cut Pro X patent, Alex4D (October 29, 2014), http://alex4d.com/notes/item/apple-magnetic-timeline-fcpx-patent (see diagram of all the various patents on Apple’s flagship video editing software, laid over a screenshot of the interface).

[36] Steve Chang, Ross Dannenberg, The Ten Most Important Video Game Patents, Gamasutra (2016), http://www.gamasutra.com/view/feature/130152/the_ten_most_important_video_game_.php?print=1.