Software Patents: Latest Case

March 8, 2007

In Capellini & Bloomberg v The Comptroller-General of Patents [2007] EWHC 476 (Pat), Mr Justice Pumfrey ruled that the Patent Office was right to reject claims to patent protection in two cases. The first involved a claim from Bloomberg in relation to a program that distributes financial data records to users working on different platforms and programmes by identifying the application that is operable by the intended recipient and delivering data in the appropriate format; the second related to a claim from Mr Capellini in respect of a novel algorithm for determining the routes to be taken by a carrier when delivering packages, the permitted routes being variable to allow two carriers to deviate, meet and transfer one or more packages.


In truth, there is no new law here but Mr Justice Pumfrey’s observations at [7] to [9] may be of interest:

  1. It is plain that the Court of Appeal in Aerotel thought that the “mental act” exclusion was a narrow one. The rival views are summarised in the appendix to the judgment at [94] to [98]. In summary, the rival contentions are that of Aldous LJ in Fujitsu (“Methods of performing mental acts, which means methods of the type performed mentally, are unpatentable, unless some concept of technical contribution is present”) and that of the Court of Appeal in Aerotel: “There is no particular reason to suppose that “mental act” was intended to exclude things wider than, for example, methods of doing mental arithmetic (every now and then someone comes up with a trick for this, for instance Trachtenberg’s system) or remembering things (e.g., in its day, Pelmanism).”

  1. As a matter of precedent, I am free to choose between these alternatives, while acknowledging that the second comes with a clearly expressed doubt as to the correctness of the first. As a practical matter, I doubt whether the difference will normally be of concern. The views of the Court of Appeal in Aerotel make it quite clear that, like most other problems that arise with a patent, the question of patentable subject matter is essentially a question of the scope of the claim. If the claim covers a method of arriving at a particular result by the exercise of rational processes alone, then it is, in my view, a claim to a “scheme, rule or method for performing a mental act”. I adhere to the view that I expressed in Halliburton v. Smith [2005] EWHC 1623 (PAT). That case was a case of a claim to a method of design, in which certain calculations were to be carried out recursively, modifying the results each time until a particular criterion was satisfied. Obviously, such a method was particularly susceptible to performance by a computer, but as a matter of principle the claim was not so limited. Nor was the claim limited to the employment of such a method in the production of a physical article. It would have been infringed had the person employing the method stopped at the end of the necessary calculations. Such a case, in my view, can be saved by limiting it to a method of manufacture of the resulting article. I do not think it can then be objectionable. But standing alone, I would respectfully suggest that an algorithm capable of being performed by a human being mentally, and complete (so far as the claim is concerned) when the algorithm terminates, is, as a matter of literal meaning, a scheme … for performing a mental act. If the physical article resulting from this design process becomes a feature of the claim, it cannot, in my judgment, be objectionable. But it is objectionable, in my view, unless “tethered” to that result. I do not, of course, say that every result must be a physical article before the claim is allowable. But if I revert for a moment to the four-stage test, there is no contribution lying outside excluded matter until the claim also covers the physical result of performing the claimed algorithm.

  1. Business methods are rather easier. It is only necessary to emphasise that I consider that Case T0931/95 Pension Benefits, while correct in the result, is incorrectly reasoned. The case is summarised in Aerotel at [100] to [106]. This was a case of a computer program to perform a particular business method, the business method itself being held to be excluded subject matter, but the computer so programmed held to be patentable subject matter but to be obvious. The basic reasoning appears to have been that the Technical Board of Appeal considered that contributions to inventive step lying in excluded matter should not be taken into account in considering the obviousness of the claim. I really cannot see how this is permissible reasoning, if only because a vast class of inventions depend for their non-obviousness on a new discovery of some property of nature – such a discovery being excluded subject matter. I prefer to approach this problem from the direction indicated by the Court of Appeal in Aerotel: what is the claimed invention as a matter of substance? A claim to a programmed computer as a matter of substance is just a claim to the program on a kind of carrier. A program on a kind of carrier which, if run, performs a business method adds nothing to the art that does not lie in excluded subject matter.