On May 12, 2010, the Enlarged Board of Appeal ("EBA") handed down an Opinion on the patentability of computer-implemented inventions following a referral by the President of the European Patent Office. It ruled the referral inadmissible. Nevertheless, the Opinion, running to 55 pages, is still of interest, because it sets out why the EBA considers the existing body of case law to be consistent, and in so doing inevitably sets out how the examination of computer-implemented inventions is and should be conducted. For those representing the more extreme positions in the public debate on this subject—about a hundred amicus curiae briefs were filed in relation to this referral—it is now clear that they will have to direct their efforts to convincing the legislature that a change of law is required. It is also likely that national interpretations of the law in at least some contracting states to the European Patent Convention ("EPC"), notably England, will now develop further.
Under the EPC, the President of the EPO may refer a point of law to the EBA in order to ensure uniform application of the law, or if a point of law of fundamental importance arises, where two Boards of Appeal have given different decisions on that question. The Boards of Appeal referred to here are Technical Boards of Appeal, the bodies that normally decide appeals against decisions in application or opposition proceedings. The EBA ruled the referral inadmissible because it did not consider the decisions identified in the referral to be different or, insofar as they were different, that these differences reflected a legitimate development of the case law.
The EBA's Interpretation of the Case Law
There are basically four aspects of patentability: There must be an invention in a field of technology, it must be new, it must involve an inventive step, and it must be susceptible to industrial application. According to the EPC, programs for computers as such are not to be regarded as inventions, and thus do not meet the first of the four requirements. Note the significance of the phrase "as such," because it is these two words that have long formed the basis for allowing patents in Europe to all manner of inventions that are or can be implemented purely in computer code. Because the claims of a patent application or patent define the invention, the questions posed in the referral at issue relate to both the relevance of the way in which a computer-implemented invention is claimed to its patentability and to the kinds of computer-implemented inventions that are patentable.
Of interest mainly to practitioners is the question of whether it makes a difference if a claim is to a "computer program" or to a "computer program product storing a computer program" (so-called "Beauregard claim"). It is only on this question that the EBA finds that the case law has developed. Whereas one of the earliest decisions (T 1173/97 IBM) stated that it made no difference, the case law has developed (T 424/03 Microsoft). Now, any claim including at least one technical feature, which may be a computer-readable medium, is considered not to be excluded from patentability. Such a claim is to be assessed for inventive step according to the established practice of the EPO, according to which only technical features can contribute to an inventive step.
The methodology for assessing inventive step was not the subject of the referral, at which the EBA expresses its surprise, but it remarks that this is probably because the case law, as summarized in T 154/04 Duns Licensing, is settled in this regard.
Summarizing the present state of the law as the EBA finds it, a claim to a computer program that does not, when run, provide any further technical effect beyond the normal physical interactions between the program and the computer does not define a patent-eligible invention. The further effect is assessed without regard to the prior art, and thus need not be new. As soon as a computer program is claimed in combination with hardware, including in particular a data carrier, the discussion shifts to one of inventive step. In the assessment of inventive step, nontechnical and technical features together determine the context of the invention, but only the technical features can contribute to a finding that the invention involves an inventive step. Those who had hoped that the EBA would give guidance on how to determine whether a feature is "technical" will be disappointed that it has explicitly declined to do so.
Likely Consequences of the Opinion
The Opinion of the EBA is also likely to be a disappointment to the free and open source software movement, which had seized on the referral as a further chance to lobby for a more restrictive approach to patenting computer-implemented inventions. Their aims will now surely only be achievable by a change in the law. Given that previous attempts at legislation, in the form of an EU directive, failed in the course of the complicated co-decision procedure used to arrive at new EU legislation, it is unlikely that any politician would be brave enough to try again.
One possible result of the Opinion may be a shift in the manner in which the patentability of computer-implemented inventions is assessed by the courts and patent offices of EPC contracting states, notably in the United Kingdom. Indeed, one can speculate that one reason for the referral by the then president of the EPO, herself a former head of the UK Patent Office, is that the approach to assessing the patentability of computer-implemented inventions taken by the courts and the Patent Office there has differed from that taken by the EPO.
By law, the provisions of the UK Patents Act relating to patentable subject matter are so framed as to have the same effect as the corresponding provisions in the EPC. The Court of Appeal has earlier ruled (Actavis UK Ltd v. Merck & Co. Inc.  EWCA Civ 444) that in principle, the Court of Appeal is bound by its own previous decisions, but that it may depart from these on a point in the field of patent law if it is satisfied that the Boards of Appeal of the EPO have formed a settled view on that point, which differs from that arrived at in that previous decision.
When it comes to computer-implemented inventions, the main Technical Board of Appeal of the EPO in this field and the Court of Appeal disagreed quite openly on the question of whether there was a settled view. Arguing that there was none, the Court of Appeal developed its existing case law into a four-step test (Aerotel Limited v. Telco Limited; Macrossan's Application  RPC 7). This four-step test focuses on the contribution an invention makes, asking whether it falls solely within the excluded subject matter and is actually technical in nature. It has tended to lead to more restrictive outcomes than the EPO's approach to assessing computer-implemented inventions. Following on this decision, the main Technical Board of Appeal responsible for computer-implemented inventions was asked to make a reference to the EBA, but it refused, ruling that the case law of the boards of appeal was perfectly consistent, and that the Court of Appeal had not adopted a good-faith interpretation of the EPC. In a subsequent decision (Symbian v. Comptroller General of Patents  EWCA Civ 1066), the Court of Appeal nevertheless maintained its earlier position. One of the five reasons it gave for this was that there was no decision by the EBA yet, and another was that subsequent decisions of the Technical Boards of Appeal of the EPO were still not consistent. Now that there is a ruling from the EBA, stating quite clearly that there is a consistent line in the case law of the boards of appeal, and what that line is, it seems inevitable that the English courts will have to revisit the issue again, and it would be no surprise if they decided to depart from their present stance.
With the judiciary having had its final say in the matter, and with legislative initiatives looking extremely unlikely, it would appear that applicants can look forward to a period of predictability of outcomes. This is not to say that it has in any way become easier to obtain a patent in this field. Initiatives like "raising the bar" and other quality improvement measures have shifted the focus onto the issue of inventive step, with only those inventions characterized by technical features representing a non-obvious contribution over the prior art making it through.
The eyes of practitioners in this field will now be turned on the U.S., where the Supreme Court is expected to rule on patentable subject matter any day now (Bilski v. Kappos).