Wednesday 4 January 2012

T 2339/09 – Balance Act


The applicant filed an appeal after the Examining Division had rejected the application under consideration. The Board discussed the question whether Internet document D4 was publicly available.

*** Translation of the German original ***

[2] The way in which the EPO deals with citations of Internet documents is explained in a communication of the EPO (OJ EPO, 8-9 2009, p. 456-462) and in decision T 1134/06.

Disclosures made on the Internet are deemed to have been accessible as from the date on which they are published online. As a rule, it is necessary to assess whether the documents form part of the state of the art by considering the balance of probabilities (nach Abwägen der Wahrscheinlichkeit). Product catalogues or product lists of companies such as D4 are, in general, directed at potential clients and, therefore, intended for publication.

According to the search report, D4 is an Internet article dated May 22, 2006, which was found on March 21 in the Internet archive www.archive.org and which concerns a product catalogue of the HBE company. The date of the online publication of this product catalogue, therefore, is May 22, 2006, which is before the filing date of the present application, i.e. November 17, 2006. Moreover, the catalogue is marked “11.10.04”, which even refers to an earlier date of publication of this catalogue. In any case the relevant dates are before the date of filing, which makes it necessary to consider that D4 forms part of the state of the art.

It is the [applicant] who bears the burden of proof for the assertion to the contrary, i.e. that D4 has not been published before the date of filing. In this context, he has argued that when the search engine www.archiv.org is used, D4 was found with a publication date of January 7, 2007, i.e. after the filing date. However, no evidence was provided. Moreover, the name of the search engine is not identical to that of the search report and the time at which the request was made is unknown, but it may be expected to be well after the date of the search report. Therefore, the submissions of the appellant are neither sufficient nor suitable for invalidating the assumption that D4 forms part of the state of the art.

I am not sure whether this decision is really in line with T 1134/06 (approved in T 1875/06 [9]) where it was stated that “ … the fact that an Internet disclosure is state of the art under A 54(2) should be proved “beyond any reasonable doubt”.” If my understanding (and translation) is correct, this decision rather applies the standard of “balance of probabilities”. Patent attorneys should note both decisions, so that they can use whatever suits the situation they have to face.

To download the whole decision (in German), click here.

The file wrapper can be found here.

6 comments:

pat-agoni-a said...

Balance of probabilities vs. beyond any reasonable doubt. Can anybody give me an example of a situation in which one "standard" is satisfied but the other not? I see nice little words, but fail to recognize the meaning behind them.

oliver said...

A famous (admittedly non-patent) example I have heard of is the O. J. Simpson case. Simpson was tried for the murder of his ex-wife and her boyfriend. He was acquitted in the criminal trial (where the “beyond reasonable doubt” criterion is applied) because his attorney was able to persuade the jurors that there was reasonable doubt about some DNA evidence. It was not absolutely certain that Simpson was the killer. He then had to face a civil trial (where something like the “balance of probabilites” applies) ; the jury unanimously found there was a preponderance of evidence to hold Simpson liable for damages. Simpson very likely did it. So at least for Simpson, the difference between the two criteria was quite sensible.

pat-agoni-a said...

So as long as patent cases do not become criminally relevant, the balance of probabilities is the relevant criteria? Is that the conclusion of your example?

Myshkin said...

To my understanding T 1134/06 certainly did not follow the practice that was (later) laid down in OJ 2009, p. 456-462. T 1134/06 quite explicitly called for a higher standard of proof to be applied to internet documents than to documents from a "reliable source" such as patent documents.

Personally I don't see much logic in the reasoning of T 1134/06. Why should the reliability of the source have anything to do with the standard of proof to be applied? Instead, the reliability of the source should play a role in the application of the standard of proof. If a document is from a reliable source, then it will more easily pass the "balance of probabilities" test than when it came from an unreliable source.

pat-agoni-a said...

@ Myshkin: I completely agree with your reasoning.

Manolis said...

T 1134/06 was (one of) the first decision(s) regarding internet citations. In reasons 4.1 it calls for the standard of proof to be "beyond any reasonable doubt". Later it was considered that this was excessive and the standard was a bit relaxed to "the balance of probabilities" (as explained in OJ 2009, pp 457, 458).
In reality, as with any other piece of prior art, the face value of the internet document is assessed and it is to be decided whether the presented facts (eg. a publication date) are to be taken as correct or not and whether further investigation/evidence is needed.
The reliability of the source plays a role in assessing the presented facts and deciding whether further investigation/evidence is needed, as Myshkin said.