The Bilski Case - Impact on Software Patents

Posted by Prasad Krishna at Aug 24, 2010 06:45 AM |
The Supreme Court of the United States gave its decision in Bilski v Kappos on 28 June, 2010. In this case the petitioners’ patent application sought protection for a claimed invention that explains how commodities buyers and sellers in the energy market can protect, or hedge, against the risk of price changes. The Court in affirming the rejection by the Court of Appeals for the Federal Circuit also held that the machine- or-transformation test is not necessarily the sole test of patentability. The Court’s ruling of abstract ideas as unpatentable and its admission that patents do not necessarily promote innovation and may sometimes limit competition and stifle innovation have provided a ray of hope. In the light of the developments, the Bilski decision as far as patentability of software is concerned may not be totally insignificant, says Krithika Dutta Narayana.

The United States Supreme Court’s much awaited decision of last month in Bilski v. Kappos (2010) (Bilski), a case that was touted as a potential watershed in the debate surrounding patentability of software, was disappointing, even though it was not without any impact. While the Supreme Court affirmed the rejection by the Court of Appeals for the Federal Circuit (CAFC) of a patent claim for a business method, it failed to define with clarity, any test for patentability which might have constituted a precedent for future cases involving patentability of software or business method. At the same time, it held that the “machine- or- transformation” test which was the test followed by the CAFC in rejecting the claim, was not the sole test to determine patentability, thus effectively providing no guideline to determine patentability of software or business methods in future cases.

The Supreme Court in Bilski, affirmed the rejection by the CAFC in In re Bilski (2008) of a patent claim involving a method of providing insurance against fluctuating energy prices due to changes in weather. The applicants, Bernard L. Bilski and Rand Warsaw filed a patent application for such a method of hedging risks – essentially a claim for a business method – under Section 101 of US Patent Act before the United States Patent and Trademark Office (USPTO). The examiner at the USPTO rejected the claim on the ground that the claim was not for patentable subject matter and that “the invention is not implemented on a specific apparatus and merely manipulates (an) abstract idea and solves a purely mathematical problem without any limitation to a practical application, therefore, the invention is not directed to the technological arts”. The Board of Patent Appeals and Interferences (BPAI) took a re-look at the examiner’s decision and held that the “machine or apparatus” test was in itself insufficient to determine patentability since a claim that included transformation of a physical object from one state to another would also be patent eligible subject matter. The BPAI also struck down the requirement of the invention to be a “technological art”. Thus, it rejected the Bilski claim on the ground that it did not cause transformation of a physical object from one state to another, since transformation of financial liabilities and risks does not constitute transformation of physical matter.

In its decision on October 30, 2008, the CAFC affirmed the ruling of the BPAI and laid down the machine or transformation test for patentability and held that Bilski’s claim was neither tied to any machine or apparatus to derive the result nor did it cause transformation of any physical object from one state to another and is hence, unpatentable subject matter. The Court reasoned that the “machine or transformation” test was crucial for determining patentability as it ensured that the claim based on a fundamental principle did not preempt all other uses of the principle. This test was the first test since the US Supreme Court’s decision in Diamond v. Diehr (1981) – which held that laws of nature, mathematical formulae and algorithms are not patentable – that had a huge potential for laying down definitive rules for patentability including declaring software and business methods to be outside the realm of patentable subject matter. If this test was upheld in the Supreme Court, that would effectively put an end to the rise of software patents since software, in most cases, did not cause transformation of physical object from one state to another. Thus, the decision of the Supreme Court had huge stakes for both sides of the software patent debate.

In light of the same, the Supreme Court’s ruling holding that the machine or transformation test is not the sole test for determining patentability and at the same time, failing to provide any other test on which to determine patentability, was a sore disappointment. Though, it affirmed the rejection of Bilski’s patent claim on the ground that the subject matter claimed was abstract and thus not a patentable “process” under section 101, its core decision was only limited to this particular claim and it did not lay down a concrete and definitive guideline for future claims. However, one must not be too quick to dismiss this decision as either going against the interests of open society and free software or as a completely inconsequential case that simply maintains status quo. There are important takeaways for the patentability of software in the Bilski decision – The Court did not totally reject the machine or transformation test relied on by the CAFC. It only held that the machine or transformation test is not the sole test on basis of which the patentability of a subject matter of a claim can be decided. The Court, in fact, held that the “machine or transformation test” was a “useful and important clue, an investigative tool for determining whether some claimed inventions are processes under section 101.”  This leaves open the possibility of using the test to determine patentability in future cases and this is good news for opponents of software patents since software (an algorithm designed to be operated upon by a computer) is merely an abstract idea which, in most cases, does not involve transformation of a physical object from one state to another.

Bilski’s claim was essentially interpreted to be a patent for a business method. The Supreme Court was completely silent on the issue of patentability of software in its decision and stuck to only the narrow issue in hand – that of the patentability of a particular business method. This means that the “machine or transformation test”, whose applicability was ruled out in this particular case, may still be applicable for software patents. Nothing in this case precludes an opponent of a software patent from urging the courts to use the “machine or transformation test” to rule on patentability. Thus, the very fact that the Supreme Court only dealt with the narrow issue in hand ensures that the “machine or transformation test” is not altogether dismissed.

The main ground on which Bilski’s claim was rejected was that the patent claim was for an overly abstract idea which was not patent-eligible. The Court held that the basic concept on which the claim was based – the concept of hedging risks against risk is an unpatentable abstract idea. Further, some of the claims are constituted by equations and are purely mathematical in nature and are abstract and thus not patentable. This means that basic concepts and use of mathematical formulae constitute abstract ideas which are unpatentable. This test can strike down many software patents as these are simply algorithms executed by a computer and incorporate very fundamental and basic concepts which are abstract in nature and are thus, not patentable. This test for determining patentability on the basis of the claim being abstract as laid down in Bilski reaffirms the patentability test laid down in Diamond v. Deihr which kept laws of nature, mathematical formulae and algorithms outside the scope of patentable subject matter. This may serve as an important test to determine and especially, limit the patentability of software in coming years.

Notwithstanding the fact that Bilski’s claim has been interpreted to be one of a business method patent, when examined in detail, the claims indicate that the ‘method’ cannot be implemented without a computer. Certain claims for calculating probability (and risk), although mathematical or algorithmic in nature, have too many variables to be executed in any way other than by using a computer.1 Such algorithms which can be executed only by a computer fall under the category of software and the patent is thus, also, a software patent. That being said, the ruling of the Court that the claim is for an overly abstract idea and thus not patentable lends credence and indicates that software patents can be validly claimed to be abstract ideas not falling under the scope of patentable subject matter.

Another important outcome of the Supreme Court’s ruling was the invalidation of the 1998 CAFC decision in State Street Bank v. Signature Financial Group2 which opened the floodgates for software patents by holding that a practical application of an algorithm or formula to produce “useful, concrete and tangible result” was sufficient to constitute patentable subject matter. The State Street test was too broad and afforded an opportunity for many frivolous patent applications to be admitted. In fact, Justice Stevens, in his concurring opinion, has stated that it would be a “grave mistake” to follow the test. By clearly striking down and dismissing such a test to determine patentable subject matter, the Court in Bilski has precluded future software patent claims for taking recourse to this test and has effectively, to an extent, made it that much harder for a software to be granted patent. The test in State Street Bank which opened the floodgates for software patents was definitively dismissed.

The Court in the 1978 case of Parker v. Flook, had rejected patent for a mathematical algorithm on the ground that an algorithm was a law of nature although its use was limited to a specific field in this case (the “field of use” test) and added an insignificant post solution activity (“post solution activity” test). The test laid down in Flook had been subsequently questioned and thus, subtly dismissed by the Court in Diehr in 1991. The Court in Bilski emphasized on the test for patentability laid down in Flook and opined that the two tests may well come in handy in future challenges or oppositions to a patent claim while determining if the claim pertained to an idea that was abstract and hence, not patentable. Thus, this test can be used in future for invalidating software patents which are characterized by broad claims adding insignificant post solution activity.

It is heartening to note that the Court looked at the importance of patent law while recognizing that patents are not always necessary to encourage innovation. It noted that patents could also limit competition and stifle innovation. They can have ill effects such as increasing prices while slowing progress and could actually be deterrent to free flow of information within society. By recognizing and validating this, the ruling not only helped increase awareness about the debate surrounding software patents but also showed that the Courts are open to such an approach to patent law in future. This can only be good news for busting software patents.

For further reading

  1. Claim 4 of Bliski's claims is as follows -  “perform a Monte Carlo simulation across all deals at all locations ... over the last 20 years of weather patterns and establish the payoffs from each deal under each historical weather pattern “ Such a simulation would involve multiple parameters such as deals, locations, weather patterns, to establish a payoff.

  2. 149 F.3d. 1368.