I know you want me to explain what the recent Bilski decision [PDF] means, because you are writing to me. I've shown you some of the apoplectic reactions from the patent bar. But what does it mean to us, to FOSS?
First, in a simplified nutshell, here is what Bilski means to everyone: You can't get patents any more on a pure mental process. You can no longer patent a process that you can think through all in your mind. In other words, abstract ideas are not patentable. There has to be either a particular machine or a transformation in the process. So pure "ideas" or "mental processes" are over. That means most business methods patents are no longer valid because they are outside the parameters of what is eligible for patenting. In simple terms, it means this:
The End for the stupidest of the stupid patents. Yay! It means that the tide is turning.
There could still be an appeal of Bilski, and even without one, there are ways to chip away at this decision's new standard for patentability, to try to get over the new turnstile, so to speak, and strategies on how to do that have begun already. I've spent the days since the decision issued researching for you, so I can explain Bilski to you. There is too much material for just one article. So, I'll break it up into parts. My purpose is to make sure you understand fully, so you can be helpful with your ideas and so you can explain this issue to others, so they understand what is involved for FOSS. If there are parts you don't understand, ask. If I don't know the answer, I can ask someone.
The most important reaction to Bilski, to me, is Red Hat's, which you can now read in full here. It's the most important to us, because what you and I want to know is: what about FOSS? What does Bilski mean to us? So, in this series of articles on the Bilski ruling, let's start there.
Rob Tiller, Vice President and Assistant General Counsel, IP, Red Hat, whose name you will find along with Richard Fontana's on the Red Hat amicus brief filed in the case, tells us what it means for FOSS:
Last week the Federal Circuit issued a major decision, In re Bilski, concerning the subject matter limits of patent law. The case presented questions relating to software patents, an issue of great significance to the free and open source software community, and so Red Hat filed a brief in Bilski to educate the appeals court about FOSS and its problems with the software patents. In the new opinion, the court cited Red Hat’s brief, but declined to settle the issue of when, if ever, software based inventions should be patentable. Even so, the new test in Bilski will probably limit the patentability of software. The war is not over, but the odds of success for FOSS just got better.
As you no doubt recall, Red Hat argued in its amicus brief that abstract ideas are not patentable. That argument won out. But Red Hat also argued that abstract ideas are not patentable *just because there is a computer involved*. The court didn't go that far. In a way, it found the opposite, that a machine has to be involved. But what it said was that as far as software is concerned, future cases will have to decide exactly where the line is. And what does "a particular machine" and "transformation" mean? Ah! That's where it gets more complex, and it is what will have to be clarified going forward. What is definitely out is the Bilski patent, and presumably all that are just like it. There was no computer or other machine in that picture, which is one reason the court decided not to use that case to clarify, on top of essentially wanting to throw the hot potato about software patents to the Supreme Court. But what happens now, in the meantime?
As we explained to the court in our brief, everyone recognizes that the purpose of patents is to foster innovation, but the patent system tends to hinder innovation for FOSS. History shows that software innovation happens without patents: prior to the 1990s, there were few software patents, but software developers produced programs that were not only innovative, but world changing. The Federal Circuit decisions in the late 1990s that allowed software patents produced a land grab that has led to more than 200,000 software patents today. Many of these patents are of poor quality, and many have vague boundaries. There is no practical way to “clear” any given program by examining all the possible patents that might apply to the features or functionality of the program. This means that the risk of lawsuits is a fact of life for the software innovators. Plainly this risk has not stopped innovation, but it is an inhibiting factor.
Bilski does not cure this problem, but it is a step in the right direction. The Federal Circuit decided to abolish the “useful, concrete, and tangible result” test that had been applied to allow dubious software patents. It substituted a new test: to be patentable, a process must be either “tied to a particular machine or apparatus” or must “transform a particular article into a different state or thing.”
The discussion of “transformation” indicates that this option probably will not cover typical kinds of software-based inventions. The question for future cases is whether “tied to a particular machine” will be interpreted broadly to favor software patents based on the mere tie between software and a general purpose computer on which it is supposed to run.
The Bilski litigation was about whether patents can be stretched to cover a disembodied idea, an abstract idea that has no machine implementing anything. The court said no, that there has to be a particular machine in the process or a transformation, but what it left unexplained is what exactly is meant by a particular machine. If you want to know what Microsoft's thinking appears to be on that subject, assuming as I do that they very much want "tied to a particular machine" to mean any old desktop running Microsoft software, read their amicus brief [PDF]. Also, you might want to review this article that explains what the issues in Bilski were. Tiller on the "particular machine" requirement and what it probably will mean:
There are good arguments against a broad interpretation of “tied to a particular machine.” The Supreme Court’s Benson decision, which was a primary authority for the Bilski court, involved claims covering an algorithm implemented in software running on a general purpose computer, and the Supreme Court concluded that the process claims were not directed to patentable subject matter. Moreover, a recent PTO Board of Patent Appeals decision, Ex parte Langemyr, found that a general purpose computer was not a “particular machine” for purposes of determining whether a claimed process is patentable. Here's the Langemyr ruling. And here's Benson. Ben Klemens stressed that next battle about "particular machine":
But some patent lawyers will undoubtedly argue the other way. Patent trolls and other beneficiaries of the status quo will battle fiercely to reverse Bilski if the Supreme Court reviews it, or else they will promote narrow interpretations in future cases. Failing that, they will seek new protection through federal legislation. They will lobby influential members of Congress with all the resources at their command
On the key question of when information becomes a machine, the ruling does provide some clues: “First, [...] the use of a specific machine or transformation of an article must impose meaningful limits on the claim's scope to impart patent-eligibility. [...] Second, the involvement of the machine or transformation in the claimed process must not merely be insignificant extra-solution activity.” For more on extra-solution activity, have a look at my law review article (PDF) that focuses heavily on the idea.
You can't just add the words "on a computer" to your patent, probably, then, and hope it will fly over the USPTO's net.
So, that is the overview. If you want to see something intriguing, look at this chart, showing the percentage of patent applications approved over the years. Note the direction of the arrow. That's why some are so unhappy with the very trends that you and I find -- at last! -- so encouraging. Here's an example of the distress some patent lawyers are now feeling:
Astonishingly, the court seems to be making its own new law of patentability setting out that a software or business method patent is valid only if it (a) is tied to a particular machine or apparatus, or (b) transforms a particular article into a different state or thing. (Diamond v. Chakrabarty)
Despite the fact that Congress had earlier noted that “anything” created by humans deserves patent protection, the Federal Circuit decided that software and business methods have a new, higher standard.
I confess that I'm astonished too. But in a good way. I never expected a change like this from the US Court of Appeals for the Federal Circuit, the folks who gave us State Street. But it's here. There is more to do, helping folks to understand the new factor -- FOSS and its development model. The old ways of viewing patents when they were covering cotton gins and machinery don't fit so well in the world of software. We've made it over the first hurdle. But there are more to go.
How in the world, you ask, could anyone think that you could patent an idea or software in the first place? I think a comment the other day on Groklaw captured well the geek point of view:
Basically, we understand that programs are "speech", but
the general public thinks of programs as "things". For
the most part, the patent office and courts have suffered
from the same sort of ignorance.
Ask the man in the street about patenting a certain type
of plot line in a book, and he'll laugh the idea away as
silly, assuming he understands the difference between
patents and copyright. For example, no one would think it
reasonable if J.K. Rowling were able to demand royalties
from anyone writing any story about witches and wizards.
But if you ask the same person whether a patent on file
locking should be allowed, he may think of a "file lock"
as if it were a Master combination lock, and thus think a
patent to be OK.
So part of it is that some simply don't understand the tech very well. There are two other elements, in my view: One is greed, of course. Some simply don't care about harm to others, if there is money to be made for them. And you can make a lot of money from patents.
And there is this simple truth: Folks in the world of patents genuinely believe that patents are a good thing, that they help the economy and benefit the public by encouraging innovation. They used to, after all, when patents were about machinery. So why shouldn't the same principles apply to software and business methods, they reason? State Street was an attempt to do just that. Timothy B. Lee, in ars technica, explains:
The patent at issue in State Street was both a software patent and a business patent, and language in that decision has been used to justify both types of patents. The court's newfound skepticism about business method patents will almost certainly translate into increased scrutiny of software patents as well. That's why everyone is so excited about Bilski, even though it didn't go all the way. It marks a clear shift, and by the very court that gave us the State Street decision.
The US Court of Appeals for the Federal Circuit is all about patents. It's their area of specialty, and it would be hard to find any group anywhere that loves patents more. That is who decided State Street. It's also the same court that just gutted it. Lee:
The Federal Circuit's rejection of Bilksi's application is especially striking because in the quarter century the court has been in existence, it has been a consistent supporter of broader and stronger patent rights. A series of decisions during the 1990s abandoned earlier restrictions on software and business method patents, creating the flood of new patents that we have covered extensively here at Ars. Until recently, software patent critics have regarded the Federal Circuit as the primary villain on the patent scene. I'd like to write a bit more about that in the next article. If we are going to explain this issue to others effectively, it's important to see the other side's point of view, not in a black and white way, but with empathy and comprehension that goes beyond the surface. So let's use the dissenting opinions in Bilski to take a look at that point of view.