View this document as a pdf (60kb)
In a letter dated 22 October 2008, the President of the EPO, Alison Brimelow, made a referral under Article 112(1)(b) EPC to the Enlarged Board of Appeal concerning the treatment of the computer program exclusion of Article 52(2) EPC. This referral is now pending under the reference G03/08 and can be accessed at the EPO website. The justification for the referral is that: “diverging decisions of the boards of appeal” regarding the computer program exclusion “have indeed created uncertainty”.
The referral defines a “computer” as “any programmable apparatus, such as a mobile phone or an embedded processor”. This is an extremely wide definition, since nearly every piece of electronic equipment nowadays, from DVD player to car management system, from medical monitoring systems to novelty greetings cards, include some form of processor.
The referral includes four specific questions, the first concerning whether claims can be directed specifically to a computer program if there is a corresponding allowable method and/or system claim. In particular, the referral draws attention to a difference between T1173/97 (which first allowed computer program claims at the EPO) and T424/03. The latter decision was part of the more recent case law that has downplayed the exclusion of Article 52(2) EPC, so that according to the referral, “overcoming the exclusion for programs for computers ... [has] become a formality”. However, the referral is perhaps slightly disingenuous here, since it completely ignores Article 56 EPC, yet T424/03 specifically requires technicality to be taken into consideration under Article 56 EPC, and hence in practice has not altered the overall bounds of patentability.
The second question asks whether the use of a computer in a claim specifically overcomes the computer program exclusion of Article 52(2) EPC, or whether a further technical effect is also needed. The referral cites two decisions, namely T1173/97, which first introduced the requirement for further technical effect, and T258/03, which interprets Article 52(2) EPC and Article 56 EPC in the same manner as T424/03.
The referral identifies a divergence between the treatment of (a) a claim to a method implemented by a computer program, and (b) a computer program for performing the method. It is suggested that for the former, under T258/03, the computer program exclusion is overcome through the use of a computer, whereas for the latter, a further technical effect is required to overcome the exclusion.
The referral appears to be a little confused here. Although both T1173/97 and T258/03 share the same difference in approach vis-à-vis Article 52(2) EPC and Article 56 EPC as discussed in above in relation to question 1, neither of them makes a significant distinction between claims to a method and claims to a computer program. Hence it is difficult to understand the argumentation underlying this question.
The third question asks whether a claimed feature must cause a technical effect on a physical entity in the real world in order to contribute to the technical character of the claim, and if so, can the physical entity be a computer. Alternatively, if no technical effect on a physical entity is required, the third question further asks whether features can contribute to the technical character if they are independent of the hardware used.
Whereas the first two questions are aimed primarily at claim format, the third question appears to be more broadly concerned with the boundaries of patentability. It cites two cases, T163/85 and T190/94, as supposedly requiring a technical effect on a real world entity, and two cases, T125/01 and T424/03, where this was apparently not required, thereby representing a divergence in case law. However, the existence of this divergence is far from clear-cut. Firstly, T163/85 and T190/94 may have identified the presence of a technical effect on a real world entity as a sufficient (rather than a necessary) condition. Hence the use of some other criteria to obtain patentability may not be inconsistent with these decisions. Secondly, it could be argued that there is a technical effect in T125/01 and T424/03. For example, the referral states that the invention in T125/01 simplifies the process of re-programming the module to adapt the program to work with modified hardware, while the invention in T424/03 simplifies data transfer between applications – both of which might well be considered as a technical effect (since they improve the general operation of a computer which is clearly technical apparatus).
The referral argues that “on the reasoning of the latter decisions [T125/01 and T424/03], it would appear than 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 ... It is therefore difficult to contemplate which aspects or effects of a computer program could fall within the exclusion”.
Although this portion of the referral is far from clear, since the quoted passage starts by referring to inventive step and ends up discussing the computer program exclusion itself, the argument essentially appears to be, if we allow certain programming constructs, then what meaning is left for the computer program exclusion. This question of EPO practice is perhaps understandable, since it was proposed to delete the computer program exclusion altogether from Article 52(2) EPC for EPC 2000 on the basis that it would make no particular difference to their treatment of computer-implemented inventions (although this proposal was not adopted in the final version of EPC 2000). However, the treatment of computer programs has to be seen in the broader context of the entire set of exclusions set out in Article 52(2) EPC, which is interpreted by the EPO as a requirement for technicality. If the computer program exclusion were not present, then it could have been argued that the ordinary interactions between any computer program and a machine would necessarily bestow technicality. The impact of the computer program exclusion can therefore be seen in the requirement for a “further technical effect (emphasis added)”, as per T1173/97, over and above these ordinary interactions.
The fourth question asks whether programming a computer necessarily involves technical considerations, and if so, do all the programming features contribute to the technical character. Alternatively, if programming does not necessarily involve technical considerations, the fourth question further asks if features resulting from the programming contribute to the technical character only when they contribute to a further technical effect.
A particular focus of the fourth question is on the skills to be attributed to the skilled person. According to T833/91, T204/93 and T769/92, it is considered that a programmer’s activity is excluded by Article 52(2) EPC, actually as a mental act (another excluded item under Article 52(2) EPC). In contrast, T1177/97 and T172/03 regard a programmer as a person of skill in the art. This again seems to go to the heart of whether computer-implemented inventions are patentable under the EPC, since it is stated, “if a computer program is deemed to lack technical character ... it could follow that the activity used to produce the program has to be considered similarly non-technical in nature”.
An underlying assumption of the fourth question is that there is a clear separation between the roles of engineer and programmer – e.g. it asks: “would the design [of the technical system] be the task of an engineer who would then pass on his programming requirements to the programmer”. However, in reality there is often no such separation, since for many engineering projects, the programmers are also the engineers (and vice versa).
The four questions of the referral are supplemented by a description of the relevant “Legal Framework”. This discusses, inter alia, the TRIPS requirement that patents “shall be available for any inventions ... in all fields of technology”. This wording has now been carried over into Article 52(1) EPC. The referral argues “whether or not programs for computer are excluded from patentability under TRIPS depends on whether programs for computer are defined as inventions in a field of technology. This statement is clearly incorrect. The TRIPS wording is inclusive, and so does not exclude anything from patentability. Rather, TRIPS sets minimum requirements for patentability that must be matched or exceeded. The referral concludes that “no guidance can be found in the TRIPS agreement” regarding the question of “whether some programs can be considered to be inventions”. This appears to be a rather questionable statement. The general understanding of TRIPS is that it requires patents to be available for computer programs that are suitably technical.
As mentioned above, the legal basis for the referral under Article 112(1)(b) EPC requires there to be different decisions from two Boards of Appeal regarding the relevant issues. However, the referral does nothing to dispel the impression that current EPO law and practice regarding the computer program exclusion is actually reasonably settled. For example, the referral has to go back some 10 years to find a divergent decision (and even longer in the case of the fourth question), and certainly, recent decisions, the relevant Guidelines, and the practice of the Examining Divisions all appear to be in conformity with one another. Indeed, only last year, Alain Pompidou, the previous president of the EPO, refused to make a referral on the same issue to the Enlarged Board of Appeal, arguing that there were insufficient differences between the Boards of Appeal. He did indicate that the EPO would continue to monitor developments in case a situation arose where an application might be allowed by one Board but not allowed by a different Board. It is clear however that such a situation has not yet arisen, nor has there been any substantial change in EPO case law over the past year. In these circumstances, it is entirely possible that the Enlarged Board of Appeal might declare the referral inadmissible, and therefore refuse to give an opinion.
Overall there has been mixed reaction to the referral. In some quarters it has been welcomed as an opportunity to clarify the law, and certainly the UK Court of Appeal has indicated its desire for a referral. However, the referral goes significantly beyond what is necessary for reconciling UK and European positions. On the negative side, the referral casts a cloud over the current, settled EPO practice in this area. This will cause industry a significant period of uncertainty, at least until the Enlarged Board of Appeal issues its opinion, which is likely to take at least a year or two.