Prof. Robert P. Merges, University of California at Berkeley School of Law
June 26, 2014
Those of us who sweat in the clammy gymnasia of patent law have been waiting – with a mix of excitement, dread, and cynical disregard – for the Alice v. CLS Bank decision. The idea was, when the Supreme Court took the case, that we would finally have an answer to the question of whether software can be patented under U.S. law. To say we did not get an answer is to miss the depth of the non-answer we did get. Reading the opinion reminds me of a famous passage in The Hitchhiker’s Guide to the Galaxy. Acolytes wait at the feet of a giant supercomputer, which 7.5 million years before had been asked “What is the meaning of life?” Finally, after eons of waiting, the computer spoke. Its answer was: “42.” The acolytes went forth, armed with this non-answer. And life went on. So it is with us, in the patent field. We have met our “42,” and its name is Alice. Now life must go on.
Not that this was unexpected. Alice said pretty much what most experienced (if not jaded) practitioners expected it to say: The claim in the patent at issue is invalid. It is not the type of subject matter that Congress wants patented under the Patent Act. But, as we all knew it would, the action lies not in the holding, but in the test (or “framework”) announced. The two-part framework announced by the Court in Alice was adapted from one of the Court’s earlier Section 101 cases, Mayo Collaborative Services, Inc. v. Prometheus, Inc., 132 S. Ct. 1289 (2012). According to the Alice Court, the Section 101 framework has two parts: (1) Determine if the claim at issue is directed toward an abstract idea; and (2) examine the elements of the claim to determine whether it contains an “inventive concept” sufficient to transform the abstract idea into a patent-eligible application. See Alice Corp. Pty. Ltd. v. CLS Bank Int’l, No. 13-298 (June 19, 2014), avail. at http://www.scotusblog.com/case-files/cases/alice-corporation-pty-ltd-v-cls-bank-international/, at 7, 11.
There is bound to be an enormous amount of argumentation over this framework in cases yet to come. But even at this stage a preliminary argument has broken out among patent people: Which of the two elements is more difficult to apply? Each side of the argument has its strong backers. Without choosing sides, let me illustrate some of the fun that lies ahead.
First is the problem of determining abstractness: When does a claim cover an abstract idea and when does it not? The starting point, of course, is the Court’s own precedent. Here we have some helpful guidance from the two recent cases, Bilski v. Kappos, 561 U.S. 593 (2010), and Alice itself. In the former the claim was essentially to the basic idea of hedging risk by making investments that pay off in the event some unhoped-for event takes place. And in Alice, the Court describes the claimed invention in these terms:
On their face, the claims before us are drawn to the concept of intermediated settlement, i.e., the use of a third party to mitigate settlement risk. Like the risk hedging in Bilski, the concept of intermediated settlement is “‘a fundamental economic practice long prevalent in our system of commerce.’”
Alice, at 9 (citing Bilski v. Kappos, 561 U.S. 593, 611 (2010)).
Thus the specific inventions in Bilski and Alice are the two most helpful data points in charting one’s course through the new landscape of Section 101. The fundamental flaw inherent in these claims is quite apparent: They are too broad, they cover well-known practices, and they do so in a comprehensive way. They have as their fundamental defect the absence of meaningful limitations. The claim in Bilski is very widely drawn to include the vast bulk of all hedging activity. Likewise, the claim in Alice covers all situations involving third party intermediaries who guarantee the integrity of a transaction, at least insofar as those intermediaries rely on periodic reports concerning the available funds at the disposal of the transacting parties.
So far, so sensible. Trouble arrives, however, in the form of past cases, which the Court tries heroically to fit into to its all-purpose two-part framework. Take for example the early case of Gottschalk v. Benson, 409 U.S. 63 (1972). That case involved a claim to a computer program that converted numbers from one format (binary-coded decimal) into another (true binary). The Supreme Court took the occasion to lay down some notoriously ill-considered tracks in the domain of software patents. The essence of the decision was a ruling that the mathematical algorithm underlying the computer program was discovered rather than invented – it was a preexisting mathematical relationship that the “inventor” merely harnessed for his patented invention. The resulting claim to the algorithm itself, was, the Court decided, tantamount to a claim to a law of nature. See Alice at 8 (citing Benson, 409 U.S. at 71-72.) This formulation of the claim misrepresents the nature of algorithms (which simply do not grow on trees), and has as a result spawned no end of confusion in the patent world. But now the Supreme Court has gone and assimilated the Benson holding into its new Bilski-Mayo-Alice framework, in a way that will surely bring future headaches. That’s because the algorithm in Benson did not cover the same kind of thing as the claims in these other recent cases; and it did not cover what it did claim nearly as broadly as the claims in the recent cases. By saying in effect that algorithms are a species of abstract idea, the Court invited all manner of mischief. An entire shelf full of discredited cases on the metaphysics of what is and is not an algorithm must now be dusted off. It would have been better to simply cite Benson as a “cf.” or the like, and rely much more heavily only on the recent cases. The Court missed a chance to limit the damage and put to rest some ill-considered chestnuts from the past.
Within its proper province, however, the first step of the Alice framework can do some useful work. Take for example a fairly recent Federal Circuit software case, Ultramercial, Inc. v. Hulu LLC, 722 F.3d 1335 (Fed. Cir. 2013). The claim in that case was drawn to distributing copyrighted content over the Internet and requiring users to view advertising to gain access to the content. The Federal Circuit, in an opinion by then-Chief Judge Rader, held that this was not an abstract idea, as the detailed steps in the process called for complex and intricate computer programming in order to carry out the claimed process.
This holding would almost surely not stand post-Alice. The view-for-content claims were crafted at a level of abstraction that is uncomfortably close to the claims in Bilski and Alice. The steps of the process were claimed in very general terms, and it would be hard to argue that they did not cover the concept at the same degree of comprehensiveness as the claims covered in the two lodestar cases.
Yet there remains the problem that the Court’s insistence on carrying forward cases such as Benson makes it difficult to figure out just what it means for a claim to cover an abstract idea. To see this, consider as an example a celebrated claim issued to Larry Page, co-founder of Google. The Page patent claims what is known as the page rank algorithm: a way of weighting web pages by the density of links to them. Lawrence Page, “Method for Node Ranking in a Linked Database,” US Patent 6,285,999, Sept. 4, 2001. The idea is that when a web search turns up various web pages that include the search term, the more important pages will be those that show a density of links to and from them. This idea was key to the early success of Google in gaining a reputation for superior search results, and the rest, as they say, is history. My point is simply that it is conceivable that the claims in this patent could be characterized as too abstract under Alice. If a court were to become convinced that the Page patent claimed the abstract idea of “weighting,” it might be invalid. Likewise, if it were convinced that the page rank procedure was simply a claim to a weighting algorithm, it might invalidate the claim by analogy to Benson. In my opinion this is a patent that ought to survive Section 101 analysis; but the courts may be sorely tested in applying Alice to achieve that result.
I have not said much about the second step of the Alice framework. In general I do not think it will have the reach of the first step; novel hardware elements are simply not a prominent feature of many software patents. At the limit in fact, the second step can be read as saying “software is patentable as long as it is really just new hardware.” In addition, many patent experts are beside themselves over the fact that the “additional inventive concept” required by the second step of the framework seems duplicative or at least highly similar to the traditional novelty and nonobviousness tests of U.S. patent law. Even so, after thinking about it for a little while, I did come up with one case that I thought would be decided by the second step. The case of Ancora Techs., Inc. v. Apple, Inc., 744 F.3d 732 (Fed. Cir. 2014) involved a patent on a novel technique for verifying that a computer user has paid the required license fee to gain access to computer software. See Miki Mullor et al., “Method of Restricting Software Operation Within a License Operation,” US Patent 6,411,941, June 25, 2002. The patent in the case called for the output of an encrypted exchange to be planted in the BIOS section of a user’s computer. This section is hard-wired with very basic software, such as the “boot routine” that is invoked when a computer is turned on. It is tricky to work with, so using it for access verification makes it difficult for hackers to work around the access procedure. Arguably the patent here describes an abstract idea: user verification. But also arguably, the use of the BIOS for this purpose is non-conventional and non-generic. Hence it may well be said to transform an abstract idea into patentable subject matter by applying the idea using novel computer hardware elements.
To sum up: We now have our answer to the age-old question of whether software can be patented. The answer is in the two-part Alice framework. It is brief, yet somehow baroquely obscure as well. Does the claim merely cover an “abstract idea”? Is there an (additional) “inventive concept” that turns this idea into a patentable application of the abstraction?
The framework is simple, in its way, yet so very slippery too. Even so it will no doubt be made to do some useful work, as I have indicated. In the end, however, the new framework reminds me of an enigmatic quote I once heard from a baseball player, interviewed just after he found out he had been traded from the New York Yankees. “The Yankees are only interested in one thing,” he said, sagely; “and I still don’t know what it is.” Likewise, we might say, when looking at software patents, after the decision in Alice the Court only cares about two things. Yet we still don’t really know what they are.