On October 30, 2008, the penultimate patent court in the United States delivered a much-anticipated decision in the matter of In re Bilski (see In re Bernard L. Bilski and Rand A. Warsaw, United States Court of Appeals for the Federal Circuit No. 2007-1130 (October 30, 2008)). As expected, the decision has far-reaching consequences for process patents, which include patents directed to many software and business method innovations. If recent history is any indicator, Bilski may also influence the development of policy in Canada concerning the patentability of business method and software subject matter.
In the Bilski decision, the Court of Appeals for the Federal Circuit attempted to clarify what constitutes a patentable process. Under U.S. law, a process is not patentable if it is a mere law of nature, natural phenomena, abstract idea or mental process. According to the Court, these are “the basic tools of scientific and technological work” and thus attempts to broadly pre-empt the use of such fundamental principles are impermissible. However, “particular applications” of fundamental principles are nevertheless patentable if they satisfy a test first developed by the U.S. Supreme Court in Gottschalk v. Benson (409 U.S. 63 (1972)).
Under the Benson test, a process is patent-eligible if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing. The Bilski court declined to define what constitutes a “particular machine” and instead focused on the second branch of this so-called machine-or-transformation test.
Not surprisingly, physical “things” such as raw materials are considered particular articles. The Bilski court also recognized that the “raw materials” of the information age are “electronic signals and electronically-manipulated data”. Significantly, however, it drew a distinction between data that represents physical articles (e.g. X-ray images of bones) and data that represents more abstract concepts such as legal obligations or business risks. Specifically, a process that transforms or manipulates data representing physical articles is patent eligible. However, a process that manipulates data representing abstract concepts may be ineligible if not tied to a particular machine.
The Court was careful to note that many software and business method innovations remain patentable under the current test. Despite this, defining acceptable claims for such methods will be more difficult following Bilski. According to the Court, it is not enough to limit a process claim to a particular field by tacking on token limitations (e.g. use of a computer in the implementation of a software routine). Instead, the use of a specific machine or transformation must impose “meaningful limits on the claim’s scope”. Likewise, the limitation must involve more than “insignificant extra-solution activity” (e.g., recording data for a software method). Unfortunately, the Court gave little guidance for determining what is “meaningful” or “insignificant”.
There is a possibility that the Canadian Intellectual Property Office (CIPO) may apply a similar line of reasoning. Earlier this year, following the Court of Appeals for the Federal Circuit decision in the matter of In re Nuijten (No. 2006-1371, September 20, 2007, Re-hearing denied February 11, 2008), CIPO changed its policy on electronic “signal claims” to adopt essentially the same position as the U.S. Court. In addition, CIPO recently initiated a review of the subject matter chapter in its patent practice guidelines and there are other indications that significant revisions are coming. While it’s far from certain that CIPO will wholly apply U.S. reasoning to eligible subject matter, as it did with signal claims, it is nevertheless probable that Bilski will have an effect.
In Canadian patent practice, software and business methods are frequently claimed as “arts” falling under section 2 of the Patent Act. The current CIPO guidelines consider an “art” to be patentable subject matter if: (1) it is an act or series of acts performed by some physical agent upon some physical object and produces in that object some change of either character or condition, and (2) it produces an essentially economic result relating to trade, industry or commerce (See Manual of Patent Office Practice, section 12.02.01). The similarities between the first part of this test and the Benson/Bilski transformation criteria are readily apparent. For their part, Canadian innovators would welcome a broadened, Bilski definition of physical object that includes virtual representations. At the same time, a Bilski approach to “meaningful” and “significant” limitations would make it more difficult to claim software and business method innovations simply by adding the physical element of computer implementation.
For innovators in the United States, In re Bilski clearly indicates a shift away from the “anything under the sun” patentability policy of the last decade. During this time, Canadian policy regarding the patentability of software and business method subject matter has been unsettled, even confused. It remains to be seen if Bilski will help to shed light on the patent eligibility of software and business method innovations here in Canada.