Key points
- "The board agrees with the appellant [opponent] that document D9, and in particular its example 6, is such a suitable starting point for the discussion of inventive step. The respondent's [proprietor's] arguments that the opposition division had correctly considered the disclosure in document D8 to represent the closest prior art and that the disclosure in document D9 was not suitable as the closest prior art are not convincing for the following reasons. "
- " Although assessing inventive step from only one "closest" prior-art disclosure may constitute a suitable approach for addressing the situation in which multiple similar disclosures are presented as starting points for the assessment of inventive step, it is not an appropriate approach if the alternative starting points likewise represent (alternative and different) routes to the invention, as document D8 and D9 do in the case in hand. In such a situation, each starting point needs consideration because, under Article 56 EPC, in order for an inventive step to be acknowledged, the claimed invention must not be obvious to a skilled person having regard to the (complete) state of the art, i.e. any prior-art disclosure excluding documents under Article 54(3) EPC (see e.g. decisions T 694/15, points 13 to 15 of the Reasons, T 816/16, point 3.7.1 of the Reasons and T 261/19, point 2.5 of the Reasons)."
- "example 6 of document D9 differs from the claimed subject-matter solely on account of the requirement in claim 1 that the classification step (e) and the estimation step (f) "are performed on one or more processors running under program instructions for the determining, classifying and estimating"."
- "the appellant [opponent] is correct that, considering the number of data points and the workflow disclosed in document D9, the skilled person, on the relevant date, would have taken computer implementation as an obvious route for improvement.
EPO
The link to the decision can be found after the jump.
Peter, the more time goes on, the more I dislike the expression "closest" prior art. Of course, the obviousness enquiry needs only one route that succeeds in showing the claimed subject matter to be obvious and judicial economy dictates that one should therefore select as starting point that single prior art reference that can provide basis for such a route. But labelling that document the "closest" prior art, while convenient for those who understand EPO-PSA, is not helpful in that it encourages irrelevant debate about the choice of prior art starting point. Furthermore, it is misleading for all those innumerable attorneys and jurists who think they understand EPO-PSA but don't. It didn't matter in the past as much as it matters now, with UPC and its non-EPO patent litigation community increasingly dominating the obviousness jurisprudence in Europe.
ReplyDeleteI wonder, do you agree with me on this point?
I tend to agree. One could consider phrases such as 'starting point" or "starting point prior art" - further suggestions are welcome. On the other hand, I think that when starting from the 'traditional' CPA, one can take a few assumptions in the next steps of the traditional PSA that are not justified when starting from 'any' prior art. But a complete analysis thereof is more suited for a journal article.
Delete