T 2239/15 (MPEG INPUT DOCUMENTS / Fraunhofer - Dolby) 19-06-2019

Priority right (Article 87(1) EPC)

1. The present application claims a right of priority from earlier US application 61/312,127 of 9 March 2010.

The appellants did not challenge the findings of the Examining Division, which the Board endorses, with regard to the invalidity of the claimed priority.

Anything made available to the public before the filing date of 4 March 2011 may thus be taken into account when deciding on the issues of novelty and inventive step.

Relevance of documents D1 and D2

2. The contents of both documents are highly relevant to the patentability of the claimed subject-matter. Reference is made in this respect to Figures 1 and 2 and to sections 2 and 3 in D1; and to Figures 1 and 2 and section 2 in D2. All in all, the Board concurs with the view of the Examining Division, as developed in section 3 of the impugned decision, that the subject-matter of claims 1, 18 and 19 is disclosed in both documents.

3. The appellants did not challenge the findings of the Examining Division with regard to the disclosures of D1 and D2, but only referred to their submissions in first instance, to the effect that they were not publicly available.

About MPEG

4. The question of the public availability of documents D1 and D2 is directly linked to the procedures of MPEG (Moving Picture Coding Experts Group ISO/IEC JTC1/SC 29/WG 11) when elaborating new standards for the encoded representation of moving pictures and audio signals. In addition to the facts established by the Examining Division, the Board's reasoning is based upon the following factual background.

5. MPEG operates in the framework of the joint ISO/IEC Technical Committee (JTC 1) on Information Technology. It is, formally, Working Group 11 (WG 11) of Subcommittee 29 (SC 29) of JTC 1. It is composed of experts, i.e. delegates accredited by national standards bodies, who have been selected to participate in the regular meetings organised by MPEG (cf. A35, pages 7, 10). The conditions for becoming member of a national delegation depend on the national bodies.

6. In the course of a meeting, which normally lasts for a full working week, the participants come together to discuss the development of new standards, their contents, and possible improvements or corrections. It may take several meetings, before a proposal is considered ripe for publication as a new International Standard (IS). A meeting may, in particular, involve discussion of the contents of draft documents ("input documents", also referred to as "m" documents) which are made available to the participants on a password-protected MPEG-dedicated server, some time before the meeting. Discussions may require the consultation of other experts, who have more specific knowledge of the topics addressed than the delegates themselves have, but who are not actually present at the meetings. Each meeting ends with a closing, plenary session, at which the committee reviews and reports on the week's progress and approves resolutions and documents ("output documents", also referred to as "w" documents) (cf. A23, slides 30, 34, 38).

7. Development of an MPEG standard follows a series of steps. A work plan is agreed, with a set of functional and performance objectives for the new standard. In order to decide upon basic technology of a standard, a competitive trial may be arranged, in which interested parties are invited to submit their proposed solutions for evaluation. These proposals are made by companies or organisations with an interest in the outcome of the standardisation process. At each meeting, proposed solutions are discussed and may be adopted or discarded (A19, chapter 4.2.5). This is thus a collaborative process of drafting and improving. The collaboration includes the definition and improvement of a "working model", which embodies early versions of the standard. The working model evolves by having alternative proposals challenge those already in it. A proposal may, for example, involve adding some new tool to the working model, or replacing one that is already there. The merits of a proposal are assessed by means of "Core Experiments", carried out under predefined conditions by multiple, independent parties (cf. A23, slides 21-24). The results of the core experiments are the bases for technological choices; improvements and additions to the working model are based on the results of the core experiments.

Public availability of documents D1 and D2 - the appellants' view

8. In the appellants' view, the requirements that apply to experts attending MPEG meetings, and the MPEG rules themselves, establish that participants are bound to secrecy.

9. The appellants put much emphasis on the fact that attendance at MPEG meetings required accreditation by a national standards body or standards committee in liaison (cf. A1a; A36, sections 1.7, 1.17). As a result, participation was strictly limited to people who, as members of a national delegation, were under the responsibility of a head of delegation appointed by the national body. The participants were bound to secrecy due to their national accreditation. This was corroborated by various documents reflecting the rules set out by various national standards bodies according to which members were bound by an explicit secrecy clause (cf. A5, section 7; , A6, section 7.9).

10. Independently of the secrecy obligations resulting from the existence of a national accreditation, explicit secrecy obligations also derived, in the appellants' view, from the relevant MPEG rules, as emphasised by reference to the "Directives" issued by ISO/IEC and "Procedures for the Technical Work" issued by JTC 1 (cf. A21, 3rd paragraph).

11. It was further stressed that the guidelines for the delegates and experts contained explicit warnings and instructions with regard to confidentiality issues (cf. A35, page 13). Concretely, it was emphasised that:

ISO working documents are not publicly available. [...] Experts are expected to respect the confidentiality of this information and to restrict the sharing of internal discussions and working documents, except for that which is necessary to the development of the publication concerned and to obtaining consensus on the content (cf A35, page 13).

12. Although referring primarily to ISO standards, A35 contained a warning, on page 1, to the effect that ISO/IEC Directives part 1 and ISO/IEC Directives, Supplement, prevailed.

13. In further confirmation for this restrictive approach, the appellants referred to A36 (Annex L, Appendix 2, § 7.5.3), according to which documented information must be adequately protected (e.g. from loss of confidentiality). As underlined by the appellants, the information was thus, from the very beginning of the standardisation procedure, confidential. This status could later be changed to non-confidential under the strict condition that the group explicitly decided in favour of a publication.

14. These general principles were further corroborated by document A23 which specified that

Input documents are all considered private to MPEG and may only be distributed outside of MPEG by the author or permission from the author in separate communication.

In the context of A23, the term "private" had to be attributed the meaning of "confidential".

15. As to documents D1 and D2, they not only belonged to the category of such "input documents", but were also documents that had been drafted at the very beginning of the standardisation process. It was a principle within MPEG that the earlier a document was drafted, the less likely it was to be published. Depending on the circumstances, several years might elapse between the production of an input document and the first publicly available, complete draft. It might also happen that an input document did not develop into a standard, if MPEG's quality and complexity requirements were not met. Input documents belonged to the category of restricted documents or "other information" referred to in section H7.1 in A31, which "must be kept private to the defined recipients". This was particularly true for documents produced at the beginning of the standardisation process, when it was still not clear whether a proposal might obtain the support of the group, whether the proposal was viable, and what lines of developments might be adopted.

16. Document A31 corresponded to the fifth Edition of the ISO/IEC JTC 1 Directives. It specified which measures had to be in force, to meet the privacy policies of ISO, IEC and JTC 1. A clear distinction was made, in A31, between open and restricted documents. Technical committee and subcommittee plenary meetings, as well as working group meetings, were open only to MPEG members (cf. A36, section SF 9.6).

17. As a result of the confidentiality obligations with respect to documents submitted for discussion at MPEG meetings, the appellants conclude that D1 and D2 were not publicly available before the filing date of the application.

Public availability of documents D1 and D2 - the Board's view

18. The Board does not agree with the appellants' conclusion.

19. A19 ('The MPEG-4 and H.264 Standards') explains the composition of MPEG:

The 'Experts' in MPEG are drawn from a diverse, world-wide range of companies, research organisations and institutes. Membership of MPEG and participation in its meetings is restricted to delegates of National Standards Bodies. A company or institution wishing to participate in MPEG may apply to join its National Standards Body [...]. Joining the standards body, attending national meetings and attending the international MPEG meetings can be an expensive and time consuming exercise. However, the benefits include access to the private documents of MPEG (including access to the standards in draft form before their official publication, providing a potential market lead over competitors) and the opportunity to shape the development of the standards.

20. While certain national standards bodies may impose strict conditions on membership, as submitted by the appellants (cf. e.g. A4, A5), no evidence was provided that this applied to all the bodies represented in MPEG. Furthermore, MPEG has no control over the conditions on membership set by national bodies and, consequently, also has no direct control over who may join and the obligations of confidentiality required, which may change over time. In effect, ISO/IEC broadly encourages national experts in the field to join, these experts normally being involved, in one way or another, with the relevant industry and research, so as to benefit from their knowledge and expertise. In return, membership of the group is recognised as a key to accessing technical information that might be used to get a market lead over competitors (see A19, 4.2.1).

21. This set-up delivers the expertise that MPEG, in turn, can draw upon. During their meetings, delegates may also consult with further experts "at home" (other members of the national standards body they belong to); cf. also the requirement to provide teleconferencing facilities at the meetings (see A36, Annex SF.5). Thereby, the delegates can provide further input to the discussions in the meetings from these other experts, who may have more specific knowledge on a subject than the delegates themselves. Intermediate results from the meetings may also be discussed with them (cf. the section "About MPEG" above). In fact, the British Standards Insitution (BSI) requires delegates and experts attending any international standardization meeting to provide a comprehensive report of the outcomes of the meeting (see A6, chapter 7.5.4).

22. Such discussions outside the strict limits of the MPEG committee are not prohibited by the system of standardisation. On the contrary, the system is based upon them. Indeed, several passages from the cited documents refer to the fundamental ISO principle that

The ISO standardisation process encourages the widest possible dissemination of the working documents needed for preparing standards free of charge within the ISO system to ensure that all interested parties have the opportunity to contribute to the development of a standard (see A31, page 140).

A32, lines 209-211 repeats this principle verbatim. A35, under the heading "Confidentiality" on page 13, notes that ISO working documents

receive a broad, controlled circulation. e.g. amongst appointed experts, members of national mirror committees and potentially among the organizations represented in such mirror committees.

23. In line with the documentary evidence on file, D1 and D2 are proposals submitted, in the framework of MPEG, by individuals affiliated with the appellants, for consideration by experts. They constitute "input contributions" or "input documents" as referred to on slides 34 and 36 of A23.

24. Page 13 of document A35 also deals with how MPEG committee members (as members of a working group within the ISO system) should treat documents. It corroborates document A23 to the extent that ISO working documents are not publicly available, but receive a wide distribution within the ISO system, notably amongst experts including members of "national mirror committees" and potentially among the organisations they represent. A35 goes on to say that experts are expected to keep the information contained in the working documents confidential and

to restrict the sharing of internal discussions and working documents, except for that which is necessary to the development of the publication concerned and to obtaining consensus on the content.

25. Documents D1 and D2 have almost the same authors; all in all, 13 people were involved in drafting these documents submitted in preparation of the meetings in Geneva (26-30 July 2010, concerning D1) and Guangzhou (11-15 October 2010, concerning D2). The exact number of members of WG11 could not be established with any certainty, but this is without relevance (the appellants estimated the number of working group members present at the meetings as 20 - 30).

26. The Board considers that this group would have been small enough to make the explicit signing of confidentiality agreements possible, if 'absolute' confidentiality (to keep it strictly limited to this group of members present in the meetings) had been intended. MPEG did not take this route.

27. On the basis of A23 and A35, with extracts from documents from various national standards organisations (cf. A2 to A7b), the appellants asserted that a tacit understanding of confidentiality existed between the members of WG 11. They further submitted that, in contrast to many standardisation organisations, it was, in effect, the intention within MPEG that members keep the contents of the discussions confidential; this approach was justified by issues regarding intellectual property rights, but also by the intention of keeping the sharing of information to a restricted group.

28. However, although A23 and A35 both state that document distribution should, in principle, be limited to those who have a legitimate interest in the development of the standard, neither A23 nor A35 refers to an absolute obligation of confidentiality. According to A23, slide 36, the authors may give permission for documents to be circulated outside MPEG. In A35, page 13, there is talk of a "broad, controlled circulation ... amongst appointed experts, members of national mirror committees and potentially among the organizations represented in such mirror committees ..." and of sharing internal discussions and working documents if "... necessary to the development of the publication concerned and to obtaining consensus on the content." So, although it might be the declared intention of MPEG to control access to the standardisation process by, for example, limiting attendance of the meetings to accredited delegates and password-protecting document archives, in practice, consultation outside the restricted circle of delegates is foreseen and expected. In effect, there is a rather broadly defined group which is party to the process which takes place behind closed doors. In this context, A35 recognises that ISO cannot guarantee the confidentiality of information. This is corroborated by the note on page 35 (point A.40) of A6 which concerns the circulation of material in the British Standards Institution. Here, it is stated that the principle of restricting the dissemination of material to those with a legitimate interest

does not preclude due consultation on such material with organizations represented on the committee. For this reason it has to be recognized that absolute confidentiality cannot be guaranteed and BSI can offer no assurances in this respect.

29. This is compatible with the general idea behind a standardisation body of trying to build a consensus around a standard by wide consultation with interested parties. So, while the whole set-up of the system puts particular emphasis on the need to keep input documents and the content of discussions rather confidential, it also acknowledges the necessity of some consultation with affiliated organisations for developing a technically sound standard that can be accepted by the relevant community. Consequently, the MPEG set-up did not guarantee or even envisage absolute confidentiality within the relatively small group present at meetings, but did, in fact, envisage a wider discussion among experts in the elaboration of standards fit for purpose.

30. Based on the evidence submitted, and contrary to the assertions of the appellants, it cannot be concluded that absolute confidentiality existed among the members of the MPEG working group responsible for drawing up and discussing D1 and D2.

31. A disclosure is regarded as made available to the public if, at the relevant date, it was possible for a member of the public to gain knowledge of the content of the disclosure with no bar of confidentiality restricting its use or dissemination (T 877/90, "T-cell growth factor/HOOPER, point 2.1.5). In the absence of any explicit confidentiality agreement, a bar of confidentiality cannot be seen to have been in place. Moreover, as already indicated with reference to A6 and A35, in view of the nature of the collaborative development process and the consensus-building procedure inherent to MPEG, confidentiality could not be guaranteed.

32. The authors of D1 and D2, and by extension the members of WG 11, came from different companies (often competitors) and were spread out across the world.

Given the heterogeneous nature of WG 11, the lack of any explicit, signed confidentiality undertaking and the consensus-building nature of its work, including the actual involvement of an indefinite number of experts from National Standards Bodies, it can be concluded that the authors, by submitting documents D1 and D2 to the members of WG 11, made their contents available to the public, at the latest by the end of their respective MPEG meetings.

33. This conclusion is not, in any way, contradicted by the affidavit of Mr Schuyler Quackenbush (A30). This affidavit affirms that input documents are not public and that MPEG delegates are not allowed to distribute non-public MPEG documents to any person that is not an authorized MPEG delegate. This does not mean, however, that knowledge of the contents of such documents is restricted to the delegates. As explained above, discussion of the contents outside the group of delegates was rather expected.

34. The appellants were not able to demonstrate that the MPEG system could guarantee, or even expect, confidentiality. On the contrary, all evidence points in the direction of no such strict limitation, but to a system designed to guarantee a certain "privacy" of its data by controlling access and transmission while at the same time being sufficiently pragmatic and flexible in order to allow such transmission with other parties in order for it to satisfactorily fulfil its mission.

Allegedly conflicting decisions

35. In T 1155/12, an MPEG document was cited as evidence of the common general knowledge at the priority date. The Board considered that the evidence available was insufficient to prove that the document was indeed available to the public.

36. The evidence available in the present case allows a deeper understanding of the structure and working procedures within MPEG. The impugned decision, in the present case, also contains a comprehensive explanation of why the Examining Division considered that documents D1 and D2 had been made publicly available. The Board, in the present case, is thus in a different situation.

37. In the present circumstances, the preponderance of evidence favours the analysis of the Examining Division.

38. Although reaching different conclusions, no contradiction can, therefore, be identified between the treatment of case T 1155/12 and the present case.

Novelty of claim 1

39. As a result of the above considerations, D1 and D2 were publicly available at the filing date of the application. Therefore, for the reasons given in the contested decision, the subject-matter of claim 1 lacks novelty.

Correspondence with DQMS - Principle of good faith and legitimate expectations

40. Even before documents D1 and D2 were made available to the MPEG delegates, the appellant Fraunhofer-Gesellschaft had complained (in connection with a different application) to the President of the EPO about the citation of MPEG input documents as prior art.

41. The appellants now claim to have relied on the letter of reply, dated 19 August 2009, by Mr Hubert Plugge, the director of DQMS (A37), which stated:

In view of legal advice and discussions with the standards bodies concerned, the EPO agrees with your view that MPEG input documents should not be consulted in searches or cited as prior art. New instructions to this effect have been issued to examiners working in the Audio Video Media field.

This letter was sent 11 and 14 months, respectively, before the meetings in which D1 and D2 were discussed.

42. In the appellants' view, by citing these documents in the International Search Report issued for the present application, and in using them as prior art against the claimed inventions, the EPO had behaved in contradiction to its own statements and thus in violation of the principle of protection of legitimate expectations.

43. This, however, overlooks the letter dated 14 September 2009, also from DQMS to Fraunhofer-Gesellschaft, in connection with a similar complaint, filed on 11 August 2009 with regard to yet another pending application (EP-08 018 793). In this letter it was stated:

By letter dated 19 August 2009 [...], you were informed that the EPO had accepted the position not to consult MPEG input documents in searches or to cite them as prior art; examiners had been instructed accordingly.

In the meantime, however, the EPO has become aware of new facts with respect to the availability of document XP030014821 to the public. The European search report should mention those documents, available to the European Patent Office at the time of drawing up the European search report [...].

44. In an email of 17 February 2012, the appellant Dolby International AB had also complained about the citation of MPEG input documents as prior art. This email referred to the letter, dated 19 August 2009, addressed by DQMS to Mr Franz Zinkler, as representative of Fraunhofer Gesellschaft, who happened to be the "external patent counsel" of Dolby International AB. The position of DQMS, expressed in its second letter to Franhofer-Gesellschaft (14 September 2009, see paragraph above), was reiterated in the reply of DQMS to Dolby International AB, dated 8 February 2012.

45. The principle of the protection of legitimate expectations of users of the European patent system, as invoked by the appellants, provides that users must not suffer any disadvantage as a result of having relied on erroneous information received from the EPO (cf. G 2/97, "Good faith", OJ 1999, 123, point 4.1; Case Law of the Boards of Appeal 9th ed. III.A.).

46. However, this principle protects parties only from disadvantageous procedural consequences of the omission of procedural steps, in relying on erroneous information from the EPO. It has no bearing on substantive law (cf. Case Law 9th ed. III.A.1.2.2), and cannot render patentable what otherwise would not be. This applies, in particular, when the information - even if it were wrong - was issued by a department of the EPO that was not competent to examine patentability. The competent Examining Division cannot ignore aspects (in this case documents) that are relevant to its decision on patentability.

47. Thus, there is no principle of protection of legitimate expectations to which the appellants could have recourse. However, even if there had been, there were no legitimate expectations to protect in the present case. Both had been informed (paragraphs 44. and 45. above) that the information in the letter dated 19 August 2009 was incorrect.

48. During oral proceedings before the Board, the appellants submitted that the Board had reached its conclusion regarding the public availability of documents D1 and D2 without hearing the appellants on the significance of the correspondence with the EPO concerning the citation of MPEG input documents in EPO international search reports.

49. The Board takes the view that it was not necessary to hear the appellant on the significance of this correspondence, because:

- the Boards conslusion that D1 and D2 were publicly available is based on the nature of MPEG procedures, on which the EPO's letters have no bearing;

- as just explained, the letters cannot affect substantive issues.

50. The Board notes, however, that the decision to mention a document in a search report is not a final decision as to whether that document was publicly available. In the present case, the appellants have had, and have availed themselves of, ample opportunity of contesting what amounts to the Search Division's preliminary view.

51. Likewise, it had no bearing on the case that the Board's attempt to obtain a copy of the agreement between the EPO and IEC, referred to in the press release cited above, was to no avail.

Request for referral to the Enlarged Board

52. For the reasons developed above (section "Allegedly conflicting decisions") the appellants' request for a referral to the Enlarged Board of Appeal must fail. There is no contradictory case law concerning the public availability of documents produced during the drafting of standards documents in the framework of the MPEG standardisation processes on which the appellants can rely.