Initial thoughts on Bilski | Opensource.com

Initial thoughts on Bilski

Posted 28 Jun 2010 by 

Rob Tiller (Red Hat)
Rating: 
(8 votes)
Image by : 

opensource.com

submit to reddit

The Supreme Court finally issued a decision in the Bilski case today [PDF]. For those troubled by the problems surrounding software patents, the opinion will be disappointing, because it does not resolve those problems. But it would be a mistake to view the opinion as a victory for the proponents of expanding software patents. In fact, there are some aspects of the opinion that auger well for the future.

To be sure, the Court did not establish a definite limitation on software patents. From listening to the oral argument of the case, I got the impression that the Court understood that its decision would have an impact on software patents, and it seemed possible that it would address the issue. It's disappointing that it didn't do so.

But the Court's silence shouldn't be interpreted too broadly. The issue before the court concerned a business method patent, rather than a software patent. The Court's traditional jurisprudence calls for deciding only the case before it on narrow grounds, rather than speaking to related legal issues. Here it followed the traditional methodology, and addressed only the issue of business method patents. Its discussion of that issue should not be understood as any kind of approval of software patents.

In fact, there are several points in the Court's discussion of business method patents that are positive. The Court didn't entirely reject the Federal Circuit's machine-or-transformation test. This test has been used by some lower federal courts to reject some software patents. Although the Bilski Court found that the test was not the exclusive test of patentable subject matter, it also found that it was a “useful and important clue, an investigative tool for determining whether some claimed inventions are processes under Section 101.”

The Court also made clear that the the Federal Circuit's pre-Bilski approach is no longer valid. Justice Kennedy's majority opinion made explicit that the Court was not endorsing the “useful, concrete, and tangible result test” of State Street, and the Justice Stevens's concurring opinion stated that following that test would be “a grave mistake.” Thus the Federal Circuit's test that contributed to opening the floodgates on software patents is no longer operative.

Even more encouraging, the rationale for invalidating the Bilski patent is one that could easily be applied to void some software patents. The Court found the business method patent at issue was an unpatentable abstract idea. It reaffirmed the validity of its prior case law rejecting attempts to patent abstract ideas, including mathematical algorithms. As Ben Klemens has explained in Math You Can't Use, software is properly viewed as consisting entirely of algorithms.

Finally, the Court took note of the purpose and function of patent law in a way that could be important in a future case. Justice Stevens (joined by Justices Ginsburg, Breyer, and Sotomayor), acknowledged that patents were not always necessary to encourage innovation. He wrote that business innovation does not entail the high costs and risks of some other types of innovation. Citing James Bessen and Michael Meurer, Stevens noted the business methods are made of intangible steps and involve problems of vagueness. He acknowledged that business method patents could stifle competition and hinder innovation.

Each of these points apply to software patents. As explained in Red Hat's amicus brief to the Court [PDF], there is abundant evidence that the existing system does not at all work to encourage software innovation, but rather discourages it. The Bilski decision does not fix the problem, but it points toward the possibility of a solution in a future case.

 

submit to reddit

5 Comments

stites
Open Source Champion

I think that we need to force the United States Supreme Court to face the software patent issue head on, eyeball to eyeball. One way to do that would be to fight a software patent case all the way to the Supreme Court where the defendent argues that all software patents are invalid including the one they are accused of violating. A less effective, but much cheaper, way to do that would be to submit an an amiae curia with somebody else's software patent case that is being appealed to the Supreme Court.

Another area where should attack software patents is in the U.S. Congress. Congress has been looking at patent reform for several years now. Patent reform is so contentious that no bill seems to be able to make it out of committee. Neverless we should continue to lobby the House Judiciary and Senate Judiciary Committees to include a clause in the patent reform bill to abolish computer software patents. Several years ago I spent some time lobbying Congress to abolish software patents. One of the conclusions that I came to from that experience is that it would be extremely helpful to hire a professional lobbyist to present our position to Congress.

-------------------
Steve Stites

Vote up!
0
Vote down!
0
Florian Mueller

First off, I'm glad to see you as Red Hat's assistant general counsel take this position because there was someone in that same position at Red Hat a few years ago who wanted to get rid only of business method patents but not all other software patents, which he appeared to like and defended in his lobbying in the European Parliament and presumably also on Capitol Hill (if he said the same on both sides of the pond).

In my top ten list of Bilski losers on the FOSS Patents blog, I have listed Red Hat because I believe your business model is particularly adversely affected by the ruling.

In my opinion, there can be no doubt that the Supreme Court would, if it had to, draw a line somewhere between Bilski and a software-implemented equivalent. A business method patent disguised as a "signal processing patent" but effectively also monopolizing a business method would not be affected by far and away. Those are the business method patents granted over here in Europe, and US patent attorneys could draft their claims similarly if they (ever) had to.

I believe your assumption that software patents could be invalidated on the grounds of being abstract mathematical ideas is far too optimistic. If this worked, it would affect a marginal minority of all software patents already granted and patent attorneys could easily and would certainly draft around that test subsequently.

Besides the Supreme Court's positions on legal tests, which are permissive and expansive, the overall tone of the decision shows that a majority of the justices follows the approach that in case of doubt, an invention is patent-eligible, and only legislation can change that. Given the widespread support for those patents and the absence of meaningful resistance (anywhere in the world and especially in the US), restrictive legislation on substantive rules is a non-starter. Over here it was hard enough to prevent an exacerbation, and the EU software patent directive was defeated although (one of) your predecessor(s) would have preferred for a law that would have been only a little bit more restrictive than the Bilski ruling.

Vote up!
0
Vote down!
0
amicus_curious

I have been awarded several patents myself that essentially involve using some new or enhanced method of doing something with a computer via software. My employer insists that each developer submit patent disclosures twice a year to a review committee that decides if the idea expressed should be sent on to the patent department. We are rewarded to the tune of a couple of thousand dollars for each disclosure accepted for patent processing and a hundred dollars for each disclosure submitted. The notion is that our company should have a broad portfolio of applicable technology patents that can be traded with any other company that might complain of some infringement of their own patents. That seems to be the way that business is done today.

The result is a hodge-podge of patents on very narrow issues that doubtless have simple work-arounds for anyone who would actually go to the trouble to understand what has been patented and want to avoid the issue. Since patent attorneys always want to broaden the issues, these patents end up all claiming the same broader issues and these broad claims are what stir up the technicians who demand the death of all software patents.

I think that the courts should consider the actualized value to a plaintiff in a patent suit of the patent. If the plaintiff is a troll and not actually in the business of implementing the technology claimed, the patent can be considered to be of no value. Similarly, if it can be shown that a substitute method exists that does not violate the patent, the value of the patent is restricted to the costs involved in avoiding it. For most software patents, I think that the amount of available damages would be so low as to preclude any effort to assert the patent in the first place.

Vote up!
0
Vote down!
0
Unidentified

Big companies who have lots of patents and do software themselves (like Microsoft) are worst kind of trolls. By removing patent trolls, you are removing distraction for big companies and then they lose last reason why they might have interest in removing patents altogether.

But by removing trolls, you make little difference to small companies because troll usually aim where money is. Biggest problem for small companies are established competitors with lots of patents who frivolously protect their market. Not patent trolls... unless patent troll is funded by that same big competitor.

Best way would be either to remove patents altogether, or stay quo until there is consensus to remove patents.

And how we reach consensus? There is no better defense than obvious offense. Thing called Defensive Patent License is being developed. Pool of patents and trolls around it, just like MPEG LA. But trolls just sue proprietary companies and leave FLOSS alone. When expenses for IBMs and Microsofts of the world become to high, they will say: "You FLOSS guys were right! Lets abolish patents!".

Vote up!
0
Vote down!
0
William F. Bos, P. E.

I believe that in all of these discussions and proceedings, the primary purpose of a patent has been lost. My understanding is that the primary purpose is to encourage inventors to disclose the details of their inventions. This is so that others can examine the patents and get ideas for better solutions for the betterment of society,

A patent gives an inventor a monopoly on his invention for a limited period of time in return for disclosure. The disclosure is contained in the patent document which is made available to the public. During the limited time of the patent, no one can use the patent without the inventors permission, but anyone can use the information contained in the patent to create a superior invention. In other words, the quid pro quo is supposed to be the monopoly for disclosure.

The theory is that the offer of a patent with its limited monopoly will encourage inventors to apply for patents rather than keep their inventions as
trade secrets.

Monopolies work as a detriment to our free competitive economy. Consequently, they are discouraged in our society. This is why a monopoly granted under a patent.is only for a limited time. It is assumed that society will gain more from the improvements in technology that are based on the information in the disclosures than it will lose from the time limited monopolies.

Over the years, the patent and copyright laws have been modified and interpreted in favor of the holder at the expense of society. I believe that in the case of software, we are no longer getting enough quo for our quid. These laws just are no longer appropriate for our rapidly expanding technical businesses such as the software business. For example, consider the time limits on patents and copyrights. Why should there be any proprietary rights for anything in this business over five years old? Why should Microsoft still have proprietary rights for DOS? It takes nearly five years to get patent filed and accepted. The lives of patents and copyrights have continually been extended. They have not yet been extended to the end of time, but they now may be good until the first coming.

It is quite obvious from reading this article and the excellent AMICUS brief from Red Hat, the courts just do not have the technical capability to deal with this matter. They shouldn't have to. All of the controversy indicates that the law is inappropriate. They are trying to apply it to something it was not designed for. It is quite obvious that the rate and volume flow of the pertinent technical information is so great that it is just impossible to keep track of it and evaluate it. The remedy is not in the courts, but in the legislature. They could remedy the situation very simply. Just exempt all software from the patent laws and all software except that which is used in place of printed material from the copyright laws. Software development proceeded very well before patent and copyright laws were applied to it. See Bill Gates and Professor Donald E. Knuth's remarks on pages 12 and 13 of Red Hat's amicus curiae.

William F. Bos, P. E.

Vote up!
0
Vote down!
0

Rob Tiller is vice president and assistant general counsel for Red Hat, where he manages patent, trademark, and copyright matters. He is a frequent speaker and writer on open source legal issues. Before coming to Red Hat, he was a partner with the law firm of Helms, Mulliss & Wicker, PLLC, where he specialized in commercial and IP litigation. He is a graduate of the University of Virginia School of Law, and a former clerk for Justice Antonin Scalia of the U.S. Supreme Court, and Judge

Holiday gift guide promo