T 1961/13 (Audio visual data signal description/COLUMBIA UNIVERSITY) 16-09-2014

Reasons for the Decision

1. The appeal complies with the provisions referred to in Rule 101 EPC and is therefore admissible.

2. The Board understands the appellants' main and first auxiliary requests as a single substantive request, namely that the decision under appeal be set aside and that a patent be granted on the basis of the claims filed with the statement of the grounds of appeal, in combination with various procedural requests relating to how the Board should deal with this substantive request.

The Board notes that it is not bound by the appellants' wishes in respect of the order in which it deals with the issues involved in the examination of a substantive claim request. However, it does consider it appropriate to deal first with the question of the public availability of document D2.

3. Arguments and evidence provided by the Examining Division and the appellants

3.1 The Examining Division found that the subject-matter of all claims lacked an inventive step in view of document D2. A prerequisite for this finding is that document D2 forms part of the state of the art pursuant to Article 54(2) EPC, i.e. that it was made available to the public before the priority date of the present application or, if the claimed right to priority is not valid, before the filing date. It is evident from the contested decision that the Examining Division was of the view that document D2 was published before the priority date, so that it was not necessary to investigate whether the claimed priority right was actually valid.

Although document D2 bears the date February 1999, it is apparent from the document itself that this date is meant to refer to an ISO/IEC JTC1/SC29/WG11 MPEG99 meeting that took place in February 1999 in Lancaster, UK, to which document D2 appears to have been submitted. Without further justification, this date can therefore not be taken as the date of public availability within the meaning of Article 54(2) EPC. In the contested decision the Examining Division indeed did not rely solely on the date printed on the document, but presented two lines of reasoning explaining why it considered that document D2 had been made available to the public before the priority date of the application.

3.2 According to the first line of reasoning, document D2 was made available to the public on or before 15 February 1999 through its publication on a website at the URL from which it had been retrieved. As evidence for this assertion, the Examining Division referred to document D7, which was a screenshot of search results returned by Google (in German) in response to a "quick search on Google cached pages" and which showed the URL to document D2 together with a date indication "15 Febr. 1999". According to the Examining Division, this showed that the URL to document D2 had been "snapshot" by Google on 15 February 1999.

3.3 In respect of this line of reasoning, the appellants' arguments can be summarised as follows:

- The URL provided by the Examining Division included the "CiteSeerX" domain, which had not been used until 2008, as shown by http://en.wikipedia.org/wiki/Citeseer#CiteSeerX. The URL had therefore not been available in 1999. There was no evidence of the existence of a URL to document D2 before the priority date.

- Even if a URL had existed, at the priority date it would not have been "discoverable" as required by decision T 1553/06 of 12 March 2012, as Google did not start indexing PDF documents until early 2001.

- The "15 Febr. 1999" date shown by Google was not necessarily the date of indexing. Document D10 showed an example of Google displaying search hits with dates as early as 15 September 1977. These could not represent the date that Google had indexed the documents because Google was not created until 1996. Document D11, which was the document dated by Google as "Sep 15, 1977", merely contained the phrase "September 1977". It seemed that Google had extracted this phrase, had added the specific date of the 15th, and had dated it as "Sep 15, 1977" for listing purposes.

- Inspection of the document properties of the PDF document downloaded from the URL provided by the examiner revealed that its creation date was 25 July 1999. It therefore seemed that no such PDF document existed in February 1999.

These arguments were already presented in the letter of 20 February 2013 filed during the first-instance proceedings in response to the summons to oral proceedings.

3.4 In response to the argument that the domain "CiteSeerX" had not been used until 2008, the Examining Division explained that the screenshot of document D7 showed that document D2 had been available in CiteSeer as of 15 February 1999, and that the link on the same screenshot now showed CiteSeerX as the domain, because all queries to CiteSeer had been redirected to CiteSeerX after its creation.

3.5 The Examining Division did not comment on the other arguments. Instead, it presented a second line of reasoning based on documents D8 and D9.

Document D8 is a paper that carries on its first page the indication "Published as a chapter in Advances in Multimedia: Systems, Standards, and Networks, A. Puri and T. Chen (eds). New York: Marcel Dekker, 1999". It contains a reference to document D2 (see page 28, reference [59]).

Document D9 represents software documentation and shows a date of 15 June 1999 on its cover page. It contains a reference "Scheirer, E.D., Lee, Y. & Yang, J.-W. (in press, a). Synthetic audio and SNHC audio in MPEG-4. In A. Puri & T. Chen (eds.), Advances in Multimedia: Systems, Standards, and Networks. New York: Marcel Dekker" (see page 37, reference 8).

The Examining Division essentially argued that document D9, being "external documentation to a software" and hence "destined to the public", had clearly been available to the public before the priority date of the application. Since document D9 cited the book Advances in Multimedia: Systems, Standards and Networks (hereinafter referred to as AMS) and document D2 had been cited in document D8, which was a chapter of the same book, it followed that document D2 had been available to the public before the priority date.

3.6 The appellants' arguments in respect of this second line of reasoning are as follows:

- The reference to the book AMS in document D9 acknowledged that the book was "in press", hence not yet published at the time document D9 was written.

- Document D9 bore a date of 15 June 1999 on its front page, but every other page bore the date 1 June 1999. It was clear that neither of these data indicated its publication date. There was no evidence that it had been directly and unambiguously accessible on the Internet before the priority date via "known means and methods" as required by decision T 1553/06.

- Two of the authors of paper D8 were co-authors of document D2. It would have been perfectly possible for them to include a reference to document D2, even though document D2 had not yet been published.

3.7 These arguments were presented to the Examining Division in the appellants' letter of 14 March 2013. The decision under appeal, although mentioning in its summary of facts and submissions that the applicants in said letter had given arguments concerning the availability of document D2, did not comment on those arguments.

4. Procedural violation

4.1 The right to be heard under Article 113(1) EPC encompasses the right of a party to have its comments considered in the written decision (see decision T 763/04 of 22 June 2007, reasons 4.3 and 4.4). Although a decision does not have to address each and every argument of a party in detail, it must comment on the crucial points of dispute in order to give the losing party a fair idea of why its submissions were not considered convincing (cf. decision T 1557/07 of 9 July 2008, reasons 2.6).

4.2 In the present case, the Examining Division did not comment on most of the appellants' arguments in respect of the public availability of document D2.

In particular, since the Examining Division in its first line of reasoning relied on a date indication shown by the Google search engine, it should have addressed the appellants' argument that these date indications were generated by simply extracting a date from the text of a document and were therefore of no evidentiary value in respect of the date on which documents had been available on the Internet.

Furthermore, since the Examining Division in its second line of reasoning relied on the book AMS having been published before the priority date of the present application, and for evidence of the publication date of this book relied on the reference to it in document D9, it should have addressed the appellants' argument based on the inclusion in this reference of the words "in press".

4.3 In the Board's view, the failure to comment on most of the appellants' arguments in respect of what was evidently a crucial point of dispute constitutes an infringement of the right to be heard and hence a substantial procedural violation.

5. Evaluation of the evidence presented by the Examining Division

5.1 The Examining Division's first line of reasoning based on the screenshot shown in document D7 (see point 3.23.2 above) has been convincingly refuted.

5.1.1 The Examining Division refers to document D7 as displaying the results of a search on "Google cached pages". This terminology is at least confusing in that it suggests that the search was performed among web pages stored in Google's web cache. This is not the case. Document D7 shows the result, filtered by a data range, of a regular Google search for Internet documents with part of the title of document D2 used as search criterion. The single result returned appears to point to document D2 at the same URL from which it was retrieved, presumably by the Examining Division, on 5 May 2011. This does not mean that document D2 was cached (or "snapshot") by Google, but that Google at some point in time has found and indexed document D2 at this URL.

5.1.2 The screenshot shown in document D7 displays this search result in connection with the date indication "15 Febr. 1999". Although the terminology used by the Examining Division is rather vague and imprecise, it appears to the Board that the Examining Division made the assumption that 15 February 1999 was the date that Google found (and/or indexed) document D2.

If this assumption were correct, then document D7 would indeed give a strong indication that document D2 was published before the priority date of the present application.

5.1.3 In order to challenge this assumption, the appellants submitted documents D10 and D11. Document D10 is a print-out of the results of a Google search on the phrase "information superhighway" filtered by a date range ending at 1 December 1992. One of the search results points to document D11. The associated date indication reads "Sep 15, 1977". Since 1977 is long before the Web, let alone Google's search engine, came into existence, this obviously cannot be the date on which Google found document D11.

As the appellant has pointed out, the significance of "Sep 15, 1977" becomes clear from reading the first full sentence of document D11:

"It all started 30 years ago, in September 1977, one year after Viktor Belenko's defection from the Soviet Union to Japan in his MiG-25 Foxbat jet fighter."

Apparently, Google dated document D11 by scanning its text for a date. When it found "September 1977", it completed the date by adding the 15th to it.

5.1.4 The appellants' challenge is evidently successful. The date shown in document D10 is clearly not indicative of the date on which document D11 was found or indexed by Google. The same applies to 15 February 1999 as shown in document D7, which in all likelihood was derived from document D2 in the same simple way as the date in the example of documents D10 and D11.

5.1.5 What this means is that a date reported by Google is inherently unsuitable to serve as evidence of the publication date of a document. Such a date, extracted from the text of the document, adds nothing to what is already shown by the document itself.

5.1.6 The Board notes that it should not have been necessary for the appellants to investigate the relevance of Google's date indications. It is the task of the examiner to make an objective assessment of what a particular date indication is intended to represent and how reliable it is, and to make further investigations if necessary. If it is not understood how a particular date reported by a search engine was generated, it cannot be used as evidence of a publication date.

5.1.7 The appellants' observations that the "CiteSeerX" domain from which document D2 was retrieved did not exist until 2008 and that the document properties of the corresponding PDF document revealed a creation date of 25 July 1999 are further indications of the incorrectness of the Examining Division's assumption and need no further discussion.

5.1.8 The Board can leave aside the question whether a URL to document D2 would have had to be "discoverable" at the priority date in order for document D2 to have been made available to the public through publication on the Internet.

5.2 Regarding the second line of reasoning (see point 3.53.5 above), the Board agrees with the appellants that it has not been proved that the book AMS was published before the priority date.

Therefore, even if it were assumed that document D9 including its reference to the book was written in June 1999, i.e. shortly before the priority date, it could not be concluded that document D8, which is a different chapter of the book than that referred to in document D9, had already been finalised at that time. Even if the chapter corresponding to document D8 did exist, it might not yet have included the reference to document D2.

In addition it must be noted that document D8, on the face of it, was not actually taken from the book AMS, but retrieved by the Examining Division as a separate document from another (unmentioned) source. The reference to document D2 may therefore well have been inserted later and not yet have been present in the chapter as published in the book.

Finally, it appears to be entirely possible that the authors of this chapter, among whom were two co-authors of document D2, included the reference at a time when document D2 was not yet publicly available.

5.3 The Board therefore considers that the evidence presented by the Examining Division is insufficient to conclude that document D2 was made publicly available before the priority date.

6. Remittal to the department of first instance

According to Article 11 RPBA, a board shall remit a case to the department of first instance if fundamental deficiencies are apparent in the first-instance proceedings, unless special reasons present themselves for doing otherwise.

In view of the filing date of the application and the large delays that occurred in the first-instance proceedings (in particular a period of six years between entry into the regional phase and issuance of the European supplementary search report), the Board was in principle willing to carry out a further examination of the appeal. In the communication accompanying the summons, it indicated that document D2 might still be prior art under Article 54(2) EPC by virtue of its submission to an MPEG working group in February 1999. In addition, a number of other concerns were raised in connection with the allowability of the sole claim request.

The appellants have not responded to the Board's communication. They must therefore be held to maintain their request that the case be either granted on the basis of the claims as currently on file or remitted to the department of first instance for further prosecution.

Under these circumstances, although the Board could have decided on the objections it has raised ex officio, it deems it more appropriate that the further examination of the case be carried out by the Examining Division on an expeditious basis.

7. Since lack of inventive step was the only ground for refusal, the substantial procedural violation was causal for the appeal. Reimbursement of the appeal fee under Rule 103(1)(a) EPC is therefore equitable.