purple_gear_header_logo.png

Blog Editors

Recent Posts

Analyzing Tech Regulation: Shapiro's Critique and Policy Recommendations

UPCOMING EVENT: Automated Decision-Making Technology and Artificial Intelligence

Free Speech and Trademark Law Clash in Vidal v. Elster

Using Technical Solutions to Address Issues in Privacy Law: A Talk by Professor Zubair Shafiq

An Introduction to Venture Capital

Archive

2013

To what Extent is Software Patentable Subject Matter?

Posted By Sofie Andersen, Apr 14, 2016

Does the Federal Circuit leave a door open for the patentability of software in DDR Holdings?

Ever since the test used in Mayo was affirmed by the Supreme Court in Alice, many commentators noted, that this could be the end of software patents. With DDR Holding, the Federal Circuit allowed a software patent for the first time since Alice, which makes for an interesting future in the area.

The reason that the test set forth in Mayo and Alice had everyone wondering if software was doomed, requires an explanation of the strict test (the “Alice Framework”). The Alice Framework consists of two parts. The first part of the test is to identify, whether the claims at issue somehow comprises an abstract idea (one of the three judge-made exceptions to patentable subject matter). If it does, the court should determine, whether the claims add “significantly more” than the abstract idea at issue.

The problems are somewhat evident: Almost every invention (especially software) somehow has its outset in an abstract idea. Isn’t every invention essentially an idea, involved into something more? So from the looks of it, most inventions will pass the first part of the test. At least the lower courts seem to have trouble identifying abstract ideas – the Federal Circuit in DDR Holdings couldn’t decide on which abstract idea it was, but decided that there definitely was one present.

The second part of the test is even more vague, because it requires “something more”, also referred to as “inventive concept”. But what is something more?

Here are some examples from case law: In Mayo, applying conventional steps at a high level of generality was not enough. In Benson, implementing a mathematical principle on a physical machine was not enough. In Flook and Alice, a conventional computer implementation of an abstract idea was not enough. In Diehr, a mathematical equation was employed, but in a process designed to solve a technological problem in a given industry, which was enough to satisfy “inventive concept”. The courts seem to want something less general and more specific, something more than conventional computer implementation, and some sort of real-world use.

The invention at issue in DDR Holdings, was a system designed to solve the problem of keeping customers on the “host webpage” when shopping online, as links from third party online shops would take them away. The system provided a composite website with the “look and feel” of the host website, but where customers could buy the third party products. In this case, the Federal Circuit decided–for the first time–that the invention did pass the Alice Framework. The court focused on the fact, that the invention solved an Internet related problem and was not just an Internet implemented solution to a pre-Internet world problem.

When analyzing the courts’ interpretation of the Alice Framework–based on the aforementioned cases–there is, however, some clarity to be found. As for the second part of the test, DDR Holding gives a somewhat better clue of what software might pass the test. There appears to be a great focus on whether an invention is just a computer implementation of a real world idea or if it is a solution to an Internet world problem, in which case it will pass. This approach gives a somewhat more conceptualized idea of what should and shouldn’t be patentable, but it still keeps a pretty strict limit on software patents.

When it comes to the first part of the test though, DDR Holdings hasn’t made us much smarter. It seems the interpretation of “abstract idea” is very expansive, which shows an (unfortunate) trend of expanding judge-made exceptions. The point of the judge-made exceptions is, that we don’t want to preempt the use of ideas, natural phenomena, and laws of nature.

When the ideas are so specific and implemented that there is no risk of preemption, why do we need to scrutinize them further under patentable subject matter by applying the Alice Framework? Wouldn’t the invention fail another part of the patentability analysis, like nonobviousness, anyway? Should the courts focus more on developing specific guidelines for patentable subject matter, or are flexible standards to be preferred given the difference in nature of the many industries involved in patent law?

The conclusion is, that it is still possible to get a software patent despite Alice, but the limits are pretty blurry. Is it only possible, if the invention solves an Internet related problem? In that way, we limit patents on software remarkably compared to other types of patentable subject matter. That might be what the Supreme Court intended as a compromise, since several of the Justices are against software patents per se.