16 May, 2014

Retrospective - Software Patents in the US

With the Alice Corporation v CLS Bank Supreme Court decision looming in the horizon, and the patentability of software in the US faces its most recent judicial challenge, this writer thought it was best to look at the origins of the current stance of such patents in the United States. Software patents are incredibly lucrative, generating an estimated 20 billion dollars yearly to tech giants such as Google and IBM. Regardless of this their patentability can be questioned, and an answer that still stands today was provided by the US Supreme Court 4 years ago, which is now effectively being challenged by Alice v CLS.

The case in question is Bilski v Kappos, decided in mid-2010. It dealt with Bernard Bilski and Rand Warsaw's application for a patent (found on page 79) for a method which hedged risks in the selling of commodities over time. The method, roughly explained, would have a third party selling the commodities in question to parties at a fixed price, mitigating any spikes in price should a single entity buy those commodities in large quantities at once. This also ensures that the providers of the commodities would not suffer either, as the third party purchaser would safeguard them from a lull in demand after a big buyer would take their share, ensuring that the price of the commodities would not dip below the third party's fixed price. The patent application for this method, more specifically described under Claim 1 and Claim 4 (both under contention), the latter of which puts the former in a mathematical equation, were initially rejected by the patent examiner. The decision was subsequently appealed, and through an en banc decision of the Court of Appeals, finally ended in the US Supreme Court.

What the Court was faced was an assessment of whether the patent contravened 35 USC § 101, under which one cannot patent fundamental principles, abstract ideas, laws of nature or natural phenomena. Could a process such as Bilski's one be patented as a process under section 101, or would it be merely patenting an abstract idea? The Supreme Court would have to assess Bilski's method under the machine-or-transformation test to determine if it is a process or not: "...(1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing". What this could be is the method's usage through a computer for example, making the method a more specific one as opposed to just a pure idea, being a process and thus patentable. The machine-to-transformation is not, according to the US Supreme Court, the only definitive test, but more of "...a useful and important clue, an investigative tool, for determining whether some claimed inventions are processes under §101".

Gavin always thought in the abstract
The majority's decision hinged heavily on its previous decisions of Gottschalk v Benson, Parker v Flook and Diamond v Diehr. The three decisions quite heavily dismiss the possibility of patenting processes very similar to the one which Bilski was attempting to patent, as the Court saw in each case that those processes would merely be patenting abstract ideas, leaving them outside the remit of patentability. Although Benson and Flook failed, Diehr succeeded as the process which was attempted to be patented was an actual process, unlike in the former two which attempted to patent formulas rather than real processes, leaving the door open for such attempts should they sufficiently fall within the Diehr precedent's line.

In the end the majority rejected Bilski's appeal, stating that "The concept of hedging, described in claim 1 and reduced to a mathematical formula in claim 4, is an unpatentable abstract idea, just like the algorithms at issue in Benson and Flook. Allowing petitioners to patent risk hedging would pre-empt use of this approach in all fields, and would effectively grant a monopoly over an abstract idea". This can be said to be right based on the precedents used by the Court in their decision. Bilski attempted to patent an idea of how to hedge risks, lacking a secondary component which would make the method a process aside from just a principle under which a business would operate, clearly making it just an idea, not a process.

What the Bilski case establishes does have an impact on software patents, but can be argued to be distinguishable from the forthcoming Alice case to an extent. What Bilski attempted to patent was just an idea of a process; however should a software patent sufficiently establish a process akin to what was seen in Diehr, it could very well be patentable. On the face of it the Alice patent would seemingly fall under the Benson and Flook precedent, potentially causing it to be not eligible for patentability; however this remains to be seen once the Supreme Court presents its decision on the case.

No comments:

Post a Comment