Red Hat Assistant General Counsel Analyses Supreme Court's Patent Ruling

ectoman (594315) writes The U.S. Supreme Court issued a groundbreaking decision concerning software patents, claiming that abstract ideas are not by themselves patentable. The ruling was a cause for celebration among those opposed to software patent abuse, like Red Hat's Vice President and Assistant General Counsel, Rob Tiller. Here, Tiller analyzes and offers some context for the Court's ruling, which "uses the traditional common law methodology of comparing one case to previous similar cases and harmonizing with those most similar."

Comments

Patent the invention

[tomhath]

As I read it, this says an invention is patentable - but implementing the invention on a computer doesn't necessarily add anything. That seems to be very good news for all those silly patents that only claim "on a computer" or "on the internet".

Re:Patent the invention

[Trepidity]

Yes, despite the tech press's focus on the impact on software patents, imo this is really a case about business-method patents, and that's where it'll have the main impact. The test case in the decision is about someone who tried to patent the idea of intermediated financial settlement, by essentially adding the words "on a computer" to it. The court more or less just clarified that: 1) the basic idea of settling contracts through an intermediary is not patentable; and 2) merely doing it on a computer does not transform the nonpatentable idea into a "machine for intermediated financial settlement" that constitutes an invention.

What the court didn't comment on in this case is whether a patent making specific software-implementation claims would be valid. Their problem with this patent was precisely that it didn't claim any specific technical implementation as constituting the invention, only the really general idea of "this method... on a computer".

Re:Patent the invention

[harperska]

This case was specifically about whether adding "on a computer" would make something patentable that is otherwise not, but it does have wide implications beyond software patents, including possibly business-method patents.

The things that are by their very existence unpatentable are abstract ideas, and things preexisting in nature (possibly among other things that I am not remembering). This ruling was actually more wide reaching than the red hat article suggested, because it establishes tests specifically to be used in the future and not making it so narrow as they like to do, so that the ruling would only apply to the case at hand.

This ruling doesn't just apply to software patents. The common law rule now is that if anything is not already patentable such as an idea or thing of nature, you can't patent a method based on that thing if the steps of that method are themselves well known or obvious to the industry to which they apply. In this case, an abstract idea combined with an implementation on a generic computer is considered unpatentable, and the precedent cited was from Mayo v. Prometheus where a biological function (i.e. naturally occurring) combined with a common medical procedure to measure that function was considered unpatentable by the same logic. Since Clarence Thomas relied so heavily on Mayo for this decision, that rule now seems to apply not only to stupid software patents, but anything in any industry that seems obvious to those in that industry.

In a way, they did comment on whether specific software-implementation claims would be patentable by pointing out that this claim specifically did not further the state of computing technology, suggesting that software that was truly innovative that did advance the technology and didn't just use methods "well known" and "long in use" may themselves be patentable. Otherwise, why point out that this particular software was specifically unpatentable because it was "well known" and "long in use"?

Interestingly enough, Sotomayor wrote her concurring opinion specifically to make a statement about how she thinks business methods are themselves unpatentable. I don't think concurring opinions have any common law teeth like the primary opinion does, so we don't have any specific precedent regarding business method patents. But we do know how 3 of the justices feel about them (since Ginsburg and Breyer joined Sotomayor's concurring opinion).

This is NOT a land breaking ruling

[Stan92057]

This is NOT a land breaking ruling, why in gods good name would the poster say this?? Fact: Abstract Ideas were Never patentable. It has always been the law, so the headline is totally wrong. The main number 1 problem that needs to be cleaned up is the Patent Office they are creating this unnecessary lawsuits, wasted billions because they haven't an idea what an Abstract Idea is? I just don't believe that they are that stupid. This is all IMO but the fact remains Abstract Ideas were never patentable.

Re:This is NOT a land breaking ruling

[harperska]

It is a landmark because even though abstract ideas were never patentable, it was never established that joining abstract ideas to generic implementations was also not patentable. Previous law suggested that an implementation, even if it seems obvious, was transformative enough to make an abstract idea patentable. This case clarifies that once and for all.

Software Patents Are Just Wrong

[mschwanke97402]

If software patents were around back in the 80's then IBM would have patented the hell out of the PC and its BIOS. Only IBM would be making them to this day and they'd cost $5000 for a base model. Only major corporations and well heeled geeks would own them. The Internet would likely still be DarpaNet and I could go on and on. If anyone recollects, IBM's BIOS was reverse engineered, legally, as it was only copied by copyright, as is reasonable and proper. The people that read the code described it and people who had never seen the IBM code wrote their own code to behave similarly, again, all very reasonable and legal. The processor your code runs on is the invention. Your code is your art or writing that makes the processor sing. If I can make my code sing a similar song without copying yours then that's as should be. Code should be copyright-able as are words in a book. Just because you write a clever whodunit does not mean that no one can ever write another!