Patenting Software - Still Crazy After All These Years

By:  Bruce W. Stratton

A version of this article was published by LexisNexis in The Lawyers Weekly, 20 November 2009

_____________________________________________________________________________________________________________

Decades after computer software has become both ubiquitous and essential, courts and patent offices continue to tie themselves into knots trying to decide whether patents should be granted on computer software. There are cases currently pending in both Canada and the United States whose eventual outcomes will have enormous impacts on patent coverage for software-related inventions. No matter the end result of these decisions, however, there will remain many years of debate before a reliable approach to the patenting of software-related inventions is likely to emerge.

In Canada, the leading case on the patentability of software hails from the early 1980s. In the Schlumberger case (Schlumberger Ltd. v. Canada (Patent Commissioner) [1982] 1 F.C. 845, 56 C.P.R. (2d) 204, 38 N.R. 299), the Federal Court of Appeal refused patent protection for a computer-implemented method - effectively put a legal hurdle in the way of software developers seeking patent protection.

Although the Canadian Patent Office is bound by the Schlumberger decision, since the 1980s the Patent Office has developed and revised its own internal guidelines for patenting computer software and its technological cousin “business methods” (effectively software-related inventions in the financial or business context). These guidelines, unlike the Federal Court of Appeal’s Schlumberger decision, were relatively lenient and effectively permitted software-based inventions to be patented in Canada.

However, the Canadian Patent Office is currently considering adopting stricter new guidelines on the patentability of software and business methods and earlier this year the Patent Appeal Board released the Amazon decision. The Patent Appeal Board, in Amazon (Commissioner’s Decision 1290, March 2009), rejected a software-based patent application which it considered to cover a business method and, in doing so, has again shifted the standard used to determine patentability of software-related inventions.

The applicant in the Amazon case has appealed the decision of the Patent Appeal Board to the Federal Court (Federal Court No. T- 1476-09). This closely-watched proceeding will provide a chance for the Canadian law on patentability of software and business methods to enter the Twenty-First Century. In the meantime, the Canadian Patent Office continues to use one set of internal guidelines, while considering adopting a new set that will be potentially out of date once the Amazon case makes its way to a final resolution. The result is that the standard for patentability of software in Canada remains unsettled and unpredictable, despite software-related patent applications having been filed and granted to patent in Canada by the droves over the years.

In contrast to the few decisions on point in Canada, in the United States, the jurisprudence on patentability of software and business methods lies thick on the ground. The high-profile decision of the Court of Appeals for the Federal Circuit in the State Street case (State Street Bank & Trust Company v. Signature Financial Group, Inc., 149 F.3d 1368 (Fed. Cir. 1998)) appeared to place almost all software or business method inventions into the category of eligible subject-matter. This approach had the advantage of simplicity but such “bright line” legal standard did not last.

The Supreme Court of the United States has just heard argument in the storied Bilski case (oral argument was scheduled to be heard on November 9, 2009). In this case, the court below, the Court of Appeal for the Federal Circuit, came up with a new amalgam of previous legal standards and, in the result, refused a patent to the applicant (see In re Bilski, 545 F.3d 943, 88 U.S.P.Q.2d 1385 (Fed. Cir. 2008)).

The current climate in the United States is relatively anti-patent and it is quite likely that the Supreme Court in that country will use the Bilski case to put some limits on the patentability of software and business methods. However, no matter what conclusion the Court comes to on the patentability of the specific invention that is before it, the opinion or opinions of the Court will almost certainly introduce a new approach that will be different again from that which is now established in the appellate court below.

The effect of a United States Supreme Court decision on patent law ripples through the court and patent systems of the U.S. and, indeed of Canada, for many years. The U.S. Patent Office will arrive at its own view of the law, based on these other court decisions. Some, but not all, of the United States law will be reflected in the approach of the Canadian Patent Office. Our Patent Office looks for guidance not only to the U.S. cases but to the approach taken in the United Kingdom and in the European Patent Office. Since European law on this point has its own inconsistencies and uncertainties, the guidance from overseas is no panacea.

For the computer software industry, the allowable extent of patent coverage for innovation is a crucial factor that influences investment and business planning. In a system where it can take several years to have a patent application granted by the patent office, it is important for applicants to know that patent protection is indeed available for software-related inventions but that the standards for patentability are in flux.

Although they will be essential reading when released, because of the existing high level of uncertainty that exists on the question of patentability of software and business methods, both the Amazon and Bilski decisions will be, at best, merely the beginning of an end to the current confusion in the law.