Thursday, October 11, 2012

Alice and the Abstract Idea

Looking Glass When is a computer-implemented invention a mere "abstract idea”? Does the presence of a computer in a claim make an unpatentable "abstract idea" a patentable invention? Should it matter that a computer-implemented invention is claimed as a method, system, or storage medium? Is there any real difference between these types of claims?

These are the questions that the United States Court of Appeals for the Federal Circuit is going to be considering over the next 12 months.

CLS meets Alice

On 9 October 2012 the Federal Circuit granted en banc review of CLS Bank International v. Alice Corp., Fed. Cir., No. 11-1301, 10/9/2012.

The patents under consideration involve a computerised trading platform for exchanging obligations in which a trusted third party settles obligations between a first and second party so as to eliminate “settlement risk.” The district court thought the claims were not patentable because they are directed to an abstract idea. On appeal the Federal Circuit panel reversed, concluding that the system, method, and media claims are directed to practical applications of patent eligible subject matter.

The Federal Circuit has asked the parties to file new briefs addressing my opening questions. The court has asked for the United States Patent and Trademark Office (USPTO) to file a brief as amicus curiae. Other briefs of amici curiae may be filed without consent of the parties or leave of the court. All the court asks is that the briefs comply with Federal Rules of Appellate Procedure and Federal Circuit Rules.

I guess it's time to study Alice Through the Looking-Glass. I wonder what we will find there?

Legislative intervention

Back in New Zealand we await the next chapter of our own fairy tale. Parliament resumes its Sitting Programme next week after a two week recess. The Patents Bill has passed its Second Reading on 12 September amidst much parliamentary debate. I cover the Second Reading in an earlier post Patent party in the House.

The Bill contains an exclusion for computer programs, added in response to submissions received from the open source movement. There doesn't appear to be a problem in New Zealand with the current Patents Act and the jurisprudence around it. The exclusion seems to be a preemptive strike against a harm that might eventuate. It is possible that there may be a patent troll under every bridge unless our lawmakers intervene. Just ask Alice. She found that anything is possible in Wonderland.

The exclusion, as set out in a Government Supplementary Order Paper, reads as follows:
10A Computer programs
(1) A computer program is not an invention for the purposes of this Act.
(2) Subsection (1) prevents anything from being an invention for the purposes of this Act only to the extent that a patent or an application relates to a computer program as such.
There is still the Committee of the whole House stage and the Third Reading ahead of us. We are getting there slowly. But at least we are getting there.

Photo courtesy of author Elena Kalis under Creative Commons licence.


  1. Don't be conned. You may be getting "there", but "there" is not where you think it is. The supposed "exclusion" in the Supplementary Order Paper is not what it seems. The precedent from other jurisdictions is that Section 10A(2) almost completely negates the restriction on patenting software in clause 10A(1).

    Very similar wording is used in Article 52 of the European Patent Convention. In those earlier, naive times, the drafters of the EPC might have honestly thought they were ruling out software patents, but in reality the European Patent Office (EPO) interprets this as allowing software patents provided the patents play the appropriate linguistic games in referring to the effect of running a program rather than the text of a program itself. And this interpretation, and the appeals that have upheld it, are then cited as precedent in other patent jurisdictions that supposedly exclude software patents.

    This NZ case is perhaps worse, in the drafters of the supplementary order are probably aware that this form of words is used in the EPC, and the fact that the limitation on "programs as such" has turned out to be a green flag to the granting of the worst kinds of software patents.

    Either the drafters of 10A(2) are incompetent, or they are two-faced: setting up a framework that allows software patents while being able to con people into believing they aim to prevent them.

    1. Chris,

      Thanks for your comments. I don't think anyone is getting conned here, although I accept it may look that way to those encountering this issue for the first time.

      Back in March 2002 New Zealand officials prepared a discussion paper ( They said that "[t]he Board of Appeal of the European Patent Office has, however, held that the exclusion from patentability of computer programs as such in Article 52(2) does not prevent the patenting of computer-related inventions. This means that computer software is now patentable in the European Community, provided the software is directed to carrying out a technical process”.

      Many of the submissions to the Parliamentary Select Committee in 2009 that asked for some sort of restriction on computer software favoured harmonisation with Europe. All submitters were aware, or should have been aware,of the 2002 discussion paper.

      In 2010 New Zealand officials prepared a January Report ( lawmakers. They recommended an exclusion from patent protection for computer programs, "the wording of the exclusion being similar to, or the same as, the exclusion for computer programs contained in the EPC and the UK Patents Act." The MED report stated that "IPONZ has indicated that if such an exclusion was provided, it would use EPO and UK case law and practice to interpret the exclusion".

      The report went on to state that "despite what appears to be a 'blanket' exclusion, the EPO and UK courts have ruled that some types of computer programs can be patented. In general, if an invention involving a computer program possesses a 'technical character' it will be patentable."

      I think we are all reasonably clear on where we are heading.

  2. The report says "despite what appears to be a 'blanket' exclusion, the EPO and UK courts have ruled that some types of computer programs can be patented."

    In reality, any type of computer program is patentable subject matter, provided it is described in the right way. The supposed exclusion is a mere illusion in this Looking Glass world. One source of the problem is alluded to in the next sentence "In general, if an invention involving a computer program possesses a 'technical character' it will be patentable."

    The idea of 'technical character' and 'technical effect' (from the UK patent system) are known to be deeply problematic when it comes to software. The critical problem is that *any* program can be characterised as having a technical character, even if it is internal to the computer on which it is running.

    While the text of a program that adds two numbers is itself not patentable subject matter, as it is "a program as such", I can present this software in terms that make it patentable subject matter. All I have to do is describe of a system that accepts the representation of two integers, performs a calculation, and stores a representation of the sum of the two integers. I have transformed my "program as such" into the description of a physical system that, under the EPO's interpretation of Article 52, is no longer "a program as such", and is thus patentable subject matter. (Under the UK system, I might have to go on to describe a specific application of this amazing ability to perform addition, but the problem is essentially the same.)

    Under the EPO's interpretation, and as upheld in appeals, the supposed restriction can always be circumvented.

    If you want to allow all and any software to be patentable subject matter, 10A(2) will give you that. If you think at least some kinds of software should not be patentable subject matter, then another more robust form of words is required. And it needs to be tested critically against a range of "bad" patents, and linguistic sophistry.

    1. Thanks Chris.

      That January Report I mentioned before deals with a lot of what you say. The Report mentions that a EPO/UK exclusion 'will probably be considered too liberal by those opposed to software patents, and too restrictive by those who support them'. The EPO/UK exclusion is effectively a compromise.

      The main point here is that we are not getting conned. We all know where we are heading. We've been looking at this particular issue (if you can call it that) for ten years now


Related Posts Plugin for WordPress, Blogger...