The EPO's Reductio ad Absurdum
Published 13:55, 27 October 08
I've written a lot about the danger that software patents pose to open source. The problem is that intellectual monopolies like patents are fundamentally incompatible with the idea of unfettered exchange of ideas, and the possibility that software patents might be strengthened within the European Union is a serious concern.
Court decisions on whether software patents are permissible within Europe have see-sawed wildly, with some decisions in favour being counterbalanced by others that confirm that software cannot be patented “as such”. Unfortunately, those meaningless weasel words “as such” have provided a tiny opening for proponents of software patents – typically large companies that want to use intellectual monopolies to stifle competition, and software patent lawyers who want more lucrative business – that the latter are constantly trying to widen.
Here's the latest attempt, which takes the form of a referral of a “point of law” concerning software patents by the President of the European Patent Office (EPO) to the EPO “Enlarged Board of Appeal”, something that seems to happen quite rarely. Now, you do not have to be a genius to see the problem with this; essentially, the EPO is asking itself whether it wants to widen its own jurisdiction, increase its power and boost its income by allowing software patents. Unless the Enlarged Board of Appeal consists entirely of self-denying, altruistic masochists, I think we can all guess what the answer will be.
Even if the outcome is pretty much a foregone conclusion (modulo the legalistic justification for it), the referral is an extremely important document because it lays bare with laudable clarity the fundamental problems with the idea of software patents, and the insane linguistic and logical contortions that those in favour of them resort to in an attempt to get them accepted.
The main stumbling block for software patents in Europe is Article 52 of the European Patent Convention, which states:
(1) European patents shall be granted for any inventions, in all fields of technology, provided that they are new, involve an inventive step and are susceptible of industrial application.
(2) The following in particular shall not be regarded as inventions within the meaning of paragraph 1:
(a) discoveries, scientific theories and mathematical methods;
(b) aesthetic creations;
(c) schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers;
(d) presentations of information.
(3)Paragraph 2 shall exclude the patentability of the subject-matter or activities referred to therein only to the extent to which a European patent application or European patent relates to such subject-matter or activities as such.
The exclusion of “programs for computers” is pretty plain – except for that “as such” tucked on at the end of paragraph 3, which has caused all the problems. The central issue is whether a patent might still be given for software, not for itself - “as such” - but for what has come to be known as its “technical effect”. Roughly speaking, this means that it does something in the outside world, although pinning down what exactly a technical effect is has proved as difficult as deciding what is or is not patentable in software.
One trick that has been tried by software patent lawyers is to define something called a “computer-implemented invention”, which is basically a repackaging of software to include a technical effect in order to be patentable. But as current referral points out in one of the core issues it wants “clarified” by the Enlarged Board of Appeal:
In the field of computer technology, innovation frequently lies in the particular method performed by a computer program while executed by conventional hardware. Consequently, the exclusion of computer programs as such under Article 52(2) and (3) EPC should be of key importance in this field. However, if one were to follow the reasoning of [EPO decision] T 424/03, overcoming the exclusion of programs for computers would become a formality, merely requiring formulation of the claim as a computer implemented method or as a computer program product.
In other words, if the idea of a computer-implemented invention is allowed, it threatens to eviscerate the basic ban on patenting software – the whole point, of course.
Attempts have even been made to extend this idea of a chimerical “technical effect” on the outside world *inside* the computer. But again, this causes big problems, as the referral notes:
In the case of features related to computer programs whose effects are confined to the internal working of the computer, there is uncertainty about where the line is to be drawn between technical effects and effects lying solely in the field of programs for computers, in particular if the aspects relating to programming are claimed in any detail.
That's not really surprising, since no line can be drawn once you admit this ill-defined idea of a “technical effect”, and then allow it to be used in a purely digital realm. As the referral concedes:
it would appear that an inventive step could be based on a programmer’s choice of elementary programming constructs (tables, loops, subroutines, objects) which solely serve the efficient execution of the program or indeed simplify the programmer’s work (e.g. using a subroutine rather than repeating lines of code). It is therefore difficult to contemplate which aspects or effects of a computer program could fall within the exclusion.
If you are going to allow a programmer's choice of “elementary constructs” as an inventive step, why not allow them for the constructs themselves, since there is no clear line between one that is elementary and one that isn't? Once again, admitting this kind of sophistry – allowing technical effects inside the computer - undermines the entire system.
There's another issue that the referral raises: how do you regard things like improving performance?
The effects caused by a computer program (which may or may not contribute to its technical character) may occur when the program is executed (for instance how much memory it occupies, how quickly it carries out the tasks for which it was programmed, etc.). On the other hand, there may be effects relating to software development which affect the programmer in his work (ease of maintenance of the program, flexibility, portability, reusability etc.).
In other words, there are no clear lines between the inherent nature of what a program does – like sorting a list faster – which is clearly just software, and not patentable, and the “technical effect” on the outside world of making it easier for a programmer to produce code by sorting a list faster, say, which might be according to some interpretations.
This referral is important because it exposes – officially and definitively - the inherent contradictions in the position that software can be patented. Whenever software patent lawyers try to pin things down with an even more subtle interpretation of some of the words or ideas in this domain, there are yet more conflicting court cases that try to follow those equivocal interpretations.
The referral is an attempt – a pretty desperate one, actually – to obtain a definitive resolution of those contradictions. But the appeal is being made in vain: there is no resolution, because the idea of a software patent is fundamentally flawed, an attempt to patent pure knowledge. I trained as a mathematician, and therefore understand – as do many others rather better qualified than I am in this area – that software consists of mathematical algorithms; and if they were patentable, all of mathematics, economics and business would simply grind to a halt, hopelessly entangled in impenetrable patent thickets.
Mathematics also provides a handy explanation why every attempt to come up with a coherent framework for software patents results in the conceptual dissonance that so troubles the EPO: it's called a Proof by Contradiction. If, when you assume some statement is true, you can end up proving through a series of logical steps something that is contradictory or untrue, you know your initial assumption was false. The assumption that in some way software is patentable has led to nothing but contradictions at every turn. Ergo, the initial assumption is false. Software is not patentable.