17 May 2010

EPO Enlarged Board of Appeal found Referral on Computer-Implemented Inventions inadmissible (G 3/08)

As reported earlier, in case G3/08, being a referral of former EPO President Alison Brimelow to the Enlarged Board of Appeal (see full text of referral here), the Board had to decide on the following questions:
  1. Can a computer program only be excluded as a computer program as such if it is explicitly claimed as a computer program?
  2. Can the exclusion be bypassed by tying the claimed invention to a piece of hardware?
  3. Must a claimed feature cause a technical effect in the real world in order to contribute to the technical character of the claim?
  4. Does the programming of a computer necessarily involve technical considerations?
On May 12, 2010, the Enlarged Board of Appeal issued its opinion on referral G 3/08, in which the approach of the EPO regarding the patentability of computer programs has been generally confirmed (complete file here).

The EBoA's long awaited opinion on this controversally discussed case (see e.g. the more than 100 related Amicus Curiae Briefs) has, as expected, evoked a strong response in the blogoshere, where it has already been commented by leading IP blogs, such as K's Law, the IPKat, IP Jur, Patently O, the EPLaw blog, and also by some more idiosyncratic voices, such as the End Software Patents campainers. I am not overconfident enough to think that I can add some essential thoughts to the observations to those grandmasters of IP blogging, but still, I would like to offer also some interpretation and summary, just for the sake of completeness of this blog. 

As expected by many experts, e.g. by Emeritus Professor Straus of MIPLC (cf. here) and the German Patent Bar Association (cf. here), the referral has been found inadmissible since the point of law refered to the Enlarged Board of Appeal by Ms Brimelow has not been decided by two Boards of Appeal in different ways, as would have been required by Article 112(1)(b) EPC. This view of the Enlarged Board is expressed in the official headnotes, whereas the legal essence of the Enlarged Board's ruling is given under notes 4, 6 and 7:
  1. In exercising his or her right of referral a President of the EPO is entitled to make full use of the discretion granted by Article 112 (1) (b) EPC, even if his or her appreciation of the need for a referral has changed after a relatively short time.
  2. Different decisions by a single Technical Board of Appeal in differing compositions may be the basis of an admissible referral by the President of the EPO of a point of law to the Enlarged Board of Appeal pursuant to Article 112 (1) (b) EPC.
  3. As the wording of Article 112 (1) (b) EPC is not clear with respect to the meaning of “different/abweichende/divergent” decisions the provision has to be interpreted in the light of its object and purpose according to Article 31 of the Vienna Convention on the Law of Treaties (VCLT). The purpose of the referral right under Article 112 (1) (b) EPC is to establish uniformity of law within the European patent system. Having regard to this purpose of the presidential right to refer legal questions to the Enlarged Board of Appeal the notion “different decisions” has to be understood restrictively in the sense of “conflicting decisions”.
  4. The notion of legal development is an additional factor which must be carefully considered when interpreting the notion of “different decision” in Article 112 (1) (b) EPC. Development of the law is an essential aspect of its application, whatever method of interpretation is applied, and is therefore inherent in all judicial activity. Consequently, legal development as such cannot on its own form the basis for a referral, only because case law in new legal and/or technical fields does not always develop in linear fashion, and earlier approaches may be  abandoned or modified.
  5. Legal rulings are characterised not by their verdicts, but by their grounds. The Enlarged Board of Appeal may thus into account in examining whether two decisions satisfy the requirements of Article 112 (1) (b) EPC.
  6. T 424/03 does deviate from a view expressed in T 1173/97, concerning whether a claim to a program on a computer-readable medium necessarily avoids exclusion from patentability under Article 52(2) EPC. However this is a legitimate development of the case law and there is no divergence which would make the referral of this point to the Enlarged Board of Appeal by the President admissible.
  7. The Enlarged Board of Appeal cannot identify any other inconsistencies between the grounds of the decisions which the referral by the President alleges are divergent. The referral is therefore inadmissible under Article 112(1)(b) EPC.
This ruling has been considered a "farewell gift you'd rather not receive" by Oliver G. Randl of K's Law, since it has been speculated that Ms Brimelow wished to "dump the unresolved problem of the limitations of patentability in the field of computer-implemented inventions to the Enlarged Board of Appeal as a political body after the truly political bodies competent to resolve that matter, the EPC Diplomatic Conference as well as the law-making bodies of the EU, have failed so far to find a reasonable interpretation of the concept of technicality that has potential to last for many decades".

Instead, the comparably clear language under § 7.2.7 of the Enlarged Board's opinion may be interpreted in a way that the Board "felt a bit abused by the President's attempt to use it as an instrument to clarify a political issue which had already caused the legislator to fail", cumulating in the articulated demand that "it's time for the legislator to take over":
[7.2.7] Given its object and purpose, the right of referral does not extend to allowing the President, for whatever reason, to use an Enlarged Board referral as a means of replacing Board of Appeal rulings on CII patentability with the decision of a putatively higher instance. For example, a presidential referral is not admissible merely because the European Parliament and Council have failed to adopt a directive on CII patenting or because consistent Board rulings are called into question by a vocal lobby. [...] When judiciary-driven legal development meets its limits, it is time for the legislator to take over.
Even if it really would be time for the legislator to take over, which is in fact frequently demanded by the open source/anti-patent community, there remain serious doubts among patent professionals and legal experts about whether at all some sort of agreement among the member states of either the EU or the EPC would be possible on this issue - not to mention the even more serious doubts about whether such an agreement could possibly solve this issue in a practically feasable while legally correct way, or rather represent some sort of rotten compromise found after heavy political horse trading. Besides that, the European Union might have some even more serious problems for the next couple of years than patentability of software-implemented inventions.

Considering the Enlarged Board's attitude as expressed in § 2.7.2 of its opinion, it is more than  comprehensible that not only the current EPO practise with regard to patentability of computer-implemented inventions has been confirmed but also some help is offered for correctly interpreting established case law and understanding the consistency of the EPO's approach (cf. sections 8 to 13, under "The questions of the referreal", p. 22 ff.). In the following please find some passages that I personnally found relevant for the EPO's approach to computer-implemented inventions with respect to both patentability and inventive step.

With respect to question 1 of the referral:
[10.2] It is argued in the referral that according to T 424/03 only a claim of the form "computer program for method x" could possibly be excluded from patentability as a computer program as such, whereas claims of the form "computer implemented method x" or "computer program product storing executable code for method x" would not be excluded (irrespective of the nature of the method x). T 1173/97 is said however to place the emphasis on the function of the computer program rather than on the manner in which it is claimed, for example as a computer program product or a computer-implemented method.

[10.3] T 1173/97 also referred to the "technical effect which goes beyond the 'normal' physical interactions between the program (software) and the computer (hardware) on which it is run" as a "further technical effect", and this is the expression which is generally used. As may be seen from the quotations above, the general condition for a claimed invention not to be excluded from patentability by Articles 52(2) and (3) EPC was considered to be that the claimed subject-matter has a "technical character". Thus in the particular case of a claim to a computer program it has a "technical character" if and only if the program causes a "further technical effect" when run.

[10.4] It is notable that the definition of further technical effect in T 1173/97 makes no reference to the state of the art. Thus according to this decision it may be determined whether a claim to a computer program is excluded from patentability by Articles 52(2) and (3) EPC independent of the prior art. That is, the identified further technical effect need not be new. By taking this position the Board consciously abandoned the so-called "contribution approach" which had been applied, somewhat inconsistently, in the earlier case law. [...] To our knowledge no decision of the Boards of Appeal has since challenged this choice by the Board in T 1173/97. It therefore stands as the established case law [...].

[10.5] T 1173/97 also drew the consequence from its abandonment of the "contribution approach" that, "Determining the technical contribution an invention achieves with respect to the prior art is therefore more appropriate for the purpose of examining novelty and inventive step than for deciding on possible exclusion under Article 52(2) and (3)".

[10.7] [...] T 258/03 came to the conclusion that any claim involving technical means was not excluded from patentability by Article 52(2) EPC, and since a claim directed to a method of operating a computer involved a computer it could not be excluded from patentability by Article 52(2) EPC. [...] T 424/03, Microsoft, finally extended the reasoning applied in T 258/03 to come to the conclusion that a claim to a program ("computer executable instructions" in the claim in question) on a computer-readable medium also necessarily avoids exclusion from patentability under Article 52(2) EPC: "The subjectmatter of claim 5 has technical character since it relates to a computer-readable medium, i.e. a technical product involving a carrier.This statement is quite unequivocal and stands alone as a reason for the claim not to be excluded under Article 52(2) EPC.

[10.7.1] The decision in T 424/03 did go on to note that the particular program involved had the potential of achieving a further technical effect when run and thus also contributed to the technical character of the claimed subject-matter. This fact however was not necessary to the conclusion that the claimed subject-matter avoided exclusion, since according to the reasoning of T 258/03 any technical means claimed was sufficient to overcome the exclusion of Article 52(2) EPC. The question whether the program itself caused a "further technical effect" when run, and would therefore also qualify as technical means, only assumed importance for the question of inventive step - in parallel to these decisions the Board had been developing an approach to the appraisal of inventive step taking into account the fact that some of the features of a claim might, considered alone, fall under the exclusions of Article 52(2) EPC (see T 154/04, Duns). For this approach it is important which features contribute to the technical character of the claimed subject-matter, since only such features are taken into account for the assessment of inventive step. In the particular case of T 424/03, both the computer-readable medium and the program itself were features which gave the subject-matter of the particular claim as a whole a technical character, and were both therefore to be taken into account for the assessment of its inventive step.

[10.13] The present position of the case law is thus that a claim in the area of computer programs can avoid exclusion under Articles 52(2)(c) and (3) EPC merely by explicitly mentioning the use of a computer or a computer-readable storage medium. But no exposition of this position would be complete without the remark that it is also quite clear from the case law of the Boards of Appeal since T 1173/97 that if a claim to program X falls under the exclusion of Articles 52(2) and (3) EPC, a claim which specifies no more than "Program X on a computer-readable storage medium," or "A method of operating a computer according to program X," will always still fail to be patentable for lack of an inventive step under Articles 52(1) and 56 EPC. Merely the EPC article applied is different. [...]

[13.1] [...] Indeed if the Boards continue to follow the precepts of T 1173/97 it follows that a claim to a computer implemented method or a computer program on a computer-readable storage medium will never fallwithin the exclusion of claimed subject-matter under Articles 52(2) and (3) EPC. [...]

[13.2] [...] It would appear that the case law, as summarised in T 154/04, has created a practicable system for delimiting the innovations for which a patent may be granted.
With respect to question 2 of the referral:
[11.2] The referral argues that "claims for a computer program and a computer implemented method can be seen as having an identical scope," and also that "the scope of a method claim would encompass a computer program for carrying out that method,", which would not appear to be quite the same thing, since the latter suggests that the scope of the method claim could be greater than that of a claim to a computer program. [...]

[11.2.2] There seem to be two logical weaknesses in the argumentation. The first is the assertion that a method can be divorced from the device that it is (to be) carried out on, which is made as a general statement, not limited to computer implemented methods. [...]

[11.2.3] The second logical weakness seems to be a confusion between a set of instructions for carrying out steps and the steps themselves. [...] There is a general distinction in logic to be made between an object and a name or description of the object. [...] For a famous [...] example of this kind of wordplay see Lewis Carroll's "Through the Looking Glass (and What Alice Found There)", search expression, "The name of the song is called". In such a complicated situation it is easy to confuse names or descriptions and the things they refer to.

[11.2.5] [...] the Enlarged Board considers that the skilled person understands the word "program" to refer to the sequence of instructions specifying a method rather than the method itself.

[11.2.7] While on a correct interpretation there is a distinction between a computer program and the corresponding computer-implemented method, it is conceivable that there is nonetheless a divergence in the case law of the Boards of Appeal, resulting from a false usage by Boards of the word "program". [...] It is nonetheless necessary to consider whether Boards have in fact used the expression "computer program" to mean a method in the cases where such use is asserted in the referral.

[11.2.8] [T 1173/97 said], "It is self-evident that a claim to such a computer program product must comprise all the features which assure the patentability of the method it is intended to carry out when being run on a computer." [and] "When this computer program product is loaded into a computer, the programmed computer constitutes an apparatus which in turn is able to carry out the said method." This would appear to indicate that the Board in that case saw computer programs in the same way as the Enlarged Board does.

[11.2.9] [T 38/86 said] "Although a computer program is not expressly recited in Claim 1, it is clear to a reader skilled in the art that the claim covers the case in which a computer program is used and, indeed, in the only embodiment disclosed in the application the text processing system is controlled by a set of programs and data stored in memory." Claim 1 was a method claim. However this does not mean that a computer program is a method, merely that, as it says, the claim could be satisfied by the use of a computer program. [...]
With respect to question 3 of the referral:
[12.2.1] [...] The case law of the Boards of Appeal as a whole is consistent in considering all the features that are claimed. As mentioned above the Boards have always avoided approaches which involve weighting of features or a decision which features define the "essence" of the invention. It is true that the COMVIK/Hitachi approach to deciding whether there is an inventive step may involve ignoring some features, but the method starts with a consideration of all the features together to determine whether the claimed subject-matter has a technical character. Only once this determination has been made can the Board turn to the question of which claimed features contribute to that technical character and therefore should be taken into account for the assessment of whether there is an inventive step.

[12.2.2] It is in fact a well-established principle that features which would, taken in isolation, belong to the matters excluded from patentability by Article 52(2) EPC may nonetheless contribute to the technical character of a claimed invention, and therefore cannot be discarded in the consideration of the inventive step. This principle was already laid down, albeit in the context of the so-called "contribution approach", in one of the earliest decisions of the Boards of Appeal to deal with Article 52(2) EPC, namely T 208/84, VICOM.
With respect to question 4 of the referral:
[13.2] The referral asserts, correctly in our view, that T 1177/97, SYSTRAN, considers that programming always involves technical considerations, at least implicitly, and that T 172/03, Ricoh, assumes the same in that it considers the skilled person, who, it is emphasised, is a technical expert, to be a software project team, consisting of programmers. On the other hand, T 833/91, IBM, T 204/93, AT&T, and T 769/92, Sohei, are said to consider the programmer's activity, programming, to be a mental act, falling within the exclusions of Article 52(2) EPC.

[13.5] While the referral has not actually identified a divergence in the case law, there is at least the potential for confusion, arising from the assumption that any technical considerations are sufficient to confer technical character on claimed subject-matter, a position which was apparently adopted in some cases (e.g. T 769/92, Sohei, Headnote 1). However T 1173/97, IBM sets the barrier higher in the case of computer programs. It argues that all computer programs have technical effects, since for example when different programs are executed they cause different electrical currents to circulate in the computer they run on. However such technical effects are not sufficient to confer "technical character" on the programs; they must cause further technical effects. In the same way, it seems to this Board, although it may be said that all computer programming involves technical considerations since it is concerned with defining a method which can be carried out by a machine, that in itself is not enough to demonstrate that the program which results from the programming has technical character; the programmer must have had technical considerations beyond "merely" finding a computer algorithm to carry out some procedure.

[13.5.1] Defining a computer algorithm can be seen in two different lights. On the one hand it may be seen as a pure mathematical-logical exercise; on the other it may be seen as defining a procedure to make a machine carry out a certain task. [...] Depending on which of these views is favoured the question whether computer programming always involves "technical considerations" may be answered negatively or positively. Either view may apparently be genuinely held, as may be seen from the lack of consensus in the amicus curiae submissions; which one is held depends on one's intuitive notion of the term "technical". It was apparently the intention of the writers of the EPC to take the negative view, i.e. to consider the abstract formulation of algorithms as not belonging to a technical field [...] In T 1173/97 the Board concentrated on the effect of carrying out an algorithm on a computer, noting that there were always technical effects, which led the Board, since it recognised the position held by the framers of the Convention, to formulate its requirement for a "further" technical effect. Only if a computer program, when run, produced further technical effects, was the program to be considered to have a technical character. In the same way, it would appear that the fact that fundamentally the formulation of every computer program requires technical considerations in the sense that the programmer has to construct a procedure that a machine can carry out, is not enough to guarantee that the program has a technical character [...] By analogy one would say that this is only guaranteed if writing the program requires "further technical considerations".