T 0273/02 (IC card/TOSHIBA) 27-04-2005

1. Admissibility of the appeal

The appeal meets the requirements referred to in Rule 65(1) EPC and is, therefore, admissible.

2. The identity/name of the opponent

2.1 The opposition was filed by Bull CP8. The opponent's name was subsequently changed to CP8 and accordingly registered by the EPO. On 8 April 2005 the Board was informed that in October 2002 CP8 had merged with Schlumberger Systèmes S.A. and, in March 2004, changed its name to Axalto S.A.

As evidence that the merger had taken place a copy was filed of a declaration to the Registre du Commerce et des Sociétés at the Tribunal de Commerce de Versailles dated 10 December 2002. The box "dissolution / disparition" due to "fusion" was crossed. The appellant explained that this meant that CP8 had ceased to exist as a consequence of the merger with Schlumberger Systèmes S.A.

The name change to Axalto S.A. was evidenced by an extract (with translation into English) from the minutes of a general meeting of the shareholders of Schlumberger Systèmes held on 11 March 2004. According to the "third resolution" taken by the assembly the company name was changed to Axalto S.A. There was no proof of a corresponding entry in an official register.

2.2 The respondent has objected to the transfer of the opposition and the change of names on two grounds: first, the evidence was filed so late that it was impossible for the respondent's representative to contact the respondent company (located in Japan) and also to clarify what was the legal consequence of a "fusion" under French law; second, the minutes of the general meeting were an unofficial document and not sufficient proof that the company name had changed. The lateness of the filing was in the respondent's view tactical.

2.3 The appellant replied that proof of a decision at a general meeting of shareholders to change the name of the company was sufficient since the subsequent registration was a mere formality. No tactics were involved. The changes had been reported as quickly as certain recent reorganisations within the appellant company had allowed.

2.4 The Board notes that the evidence concerning the merger and the name change enclosed with the appellant's letter dated 18 April 2005 was sent to the respondent on 20 April 2005, ie five working days before the oral proceedings. It is not known when the respondent actually received the letter. Clearly, the filing of documents shortly before oral proceedings always poses a problem. In the present case, however, the evidence in question consists of only a couple of pages which can be read and understood in a matter of minutes. The suggestion that the appellant had deliberately postponed filing the documents is not very convincing since it had apparently nothing at all to gain from producing the information late. On the contrary, it ran the risk of not being able to take part in the oral proceedings. The evidence, although filed at a late stage, should therefore be considered by the Board.

2.5 The respondent has observed that the exact legal consequences of a "fusion" according to French law might not be generally known outside France. The word "disparition" in the register excerpt seems however to indicate clearly that all activities of CP8 were transferred to Schlumberger Systèmes, in which case there can be no doubt about the latter company's right to continue the opposition proceedings. The transfer of the opposition from CP8 to Schlumberger Systèmes thus appears to have been adequately proved.

2.6 As to the subsequent change of name to Axalto, evidence has been furnished that the company actually decided this. It must therefore be assumed, and no indication to the contrary is known to the Board, that the change was subsequently entered into an official register. The mere fact that a register excerpt might have been a more direct piece of evidence than minutes of a general meeting does not invalidate the proof actually offered.

The Board thus finds that the appellant (opponent 02) is now Axalto S.A.

3. Public availability of D1

3.1 D1 is a proposal for a technical standard put forward by the European Telecommunications Standards Institute ETSI. It bears the date of 5 February 1993 (the priority date of the patent-in-suit being the 24 March 1993). On its first page it is stated expressly that it is an "unpublished work". In order to demonstrate that the contents of D1 were public in spite of this indication the appellant has referred to the "ETSI Directives", dated December 2004 (document D9). In Annex 6, "ETSI Intellectual Property Rights Policy", paragraph 10 "Confidentiality", the following is stated:

"The proceedings of a COMMITTEE shall be regarded as non-confidential except as expressly provided below and all information submitted to a COMMITTEE shall be treated as if non-confidential and shall be available for public inspection unless:

- the information is in written or other tangible form; and

- the information is identified in writing, when submitted, as confidential; and

- the information is first submitted to, and accepted by, the chairman of the COMMITTEE as confidential.

CONFIDENTIAL INFORMATION incorporated in a STANDARD or TECHNICAL SPECIFICATION shall be regarded as non- confidential by ETSI and its MEMBERS, from the date on which the STANDARD or TECHNICAL SPECIFICATION is published."

From this passage, and a similar one in the part "ETSI Guide on Intellectual Property Rights (IPRs)", version adopted by Board #48 on 1 September 2004, paragraph 2.3.6, the appellant draws the conclusion that the information in D1 was available since not identified as confidential in the text.

The respondent has denied that D1 was publicly available, observing that the ETSI Directives were published only in 2004 and that in any case the word "unpublished" on the first page of D1 showed that it was not intended for the public.

3.2 From the indication in D9 that confidential information contained in a standard shall be regarded as non- confidential from the date on which the standard is published, the conclusion appears permitted that an unpublished proposal for a standard might contain confidential information. The appellant's observation that no information in D1 is marked as confidential carries little weight as long as it has not been demonstrated that confidential information would in fact be indicated as such. Moreover, as the respondent has observed, D9 does not describe the situation in 1993 but in 2004. Therefore, even when ignoring the literal meaning of "unpublished work", the Board finds that the appellant has not shown conclusively that the information in D1 was available to the public at the date of priority.

3.3 The Board will therefore not consider D1 further.

4. Public availability of D2

4.1 D2 is a user's guide concerning the chip card system TB100 manufactured by Bull CP8 (the original opponent 02). It bears the date of 2 January 1991. The appellant has argued that it was publicly available as of that date, something which the opposition division accepted on the balance of probabilities and without asking for proof. The Board did require proof, referring in its communication to the "up-to-the-hilt" level of proof (T 472/92, OJ EPO 1998, 161) usually applied in those prior public use cases where practically all the evidence lies within the power and knowledge of the opponent. In reply, the appellant filed, by letter dated 8 April 2005, documents D10-D16.

4.2 Although the respondent has objected that this evidence was filed too late, the Board is prepared to consider it. It is true that the appellant could have filed the evidence earlier since the Board's communication was issued already in November 2004. Still, with the exception of D12 (which is an earlier version of D2) the documents are very short, and merely intended to confirm the appellant's previous argumentation. There was thus nothing surprising about the information they convey. Document D12, however, is not admitted since it is very long and, above all, was possibly never received by the respondent.

4.3 Document D14 is an invoice dated 7 January 1993 for 2500 pieces of an article referred to as "COMP. SPOM21 ROM4 TB100". The appellant has explained that this item refers to the chip used for the TB100 card, and the Board sees no reason to doubt this. The respondent's suggestion that it might merely refer to a ROM appears far-fetched considering that IC cards use on-chip memory. Moreover, "SPOM21" is defined in D2 (page 1.7) to mean the monolithic microcalculator ("Self Programmable One chip Microcalculator") having a programmable memory. The Board thus accepts that the TB100 chip or chip card was sold before the relevant date.

4.4 D10 shows that the "TB 100 User's Guide", ie D2, was used for training courses at about March 1992. Since there is a mention of "customers" in the text it is not likely that internal training is intended, as the respondent has suggested. D10 is thus in itself a strong indication that D2 was available to those who participated or could have participated at these training courses. Furthermore, since it is difficult to imagine that cards or chips are shipped without proper documentation, D14 together with D10 seem to prove that D2 was also available to purchasers of the TB100 system.

4.5 The respondent has argued that even if D2 were available to customers or trainees there was an implicit secrecy agreement due to the use of the chip cards in bank applications requiring confidentiality. The Board finds this unlikely. Customers would have to program their cards according to their needs and the final users must be informed how to handle them. Thus many people would have to know at least the general principles of the system. It is well known that the security of a chip card is mainly ensured by the use of keys. Obviously these are secret, but D2 is not concerned with keys other than at a general level.

4.6 The Board therefore holds D2 to be prior art.

5. The invention

Claim 1 concerns an IC card, and in particular its memory structure. The memory contains "files" which in their turn contain "areas" for the storage of data. The access to a file can be "inhibited". This means that a file and any areas associated with it cannot be read (and/or cannot be written or erased, as mentioned in the description, eg paragraphs [0026] and [0061]). The read access to the areas depending on a specific file may be inhibited separately so that the data they contain cannot be read.

6. The prior art

The parties agree that D2 is the closest document. D2 discloses an IC card to be used with a terminal device. The IC memory is divided into Data Files which may contain Elementary Files (page 2.11), such as Working Files.

The Working Files can be read (and written and erased) under parameterizable access controls (page 2.12). The parameters in question include WRP (Write Read Protection, 4 bits) and RP (Read Protection, 4 bits) (page 8.51). The bit combinations defining the access conditions for different users, such as "without protection" or "forbidden", are defined on page 9.60.

A Data File can be "invalidated" when the card has reached its last phase, called "end of life" (fig. 1.4). At the same time all its Elementary Files are invalidated (page 5.37). A subset of the commands can be performed on an invalidated Data File, including the read command (whereas writing and erasing are not allowed).

7. Novelty

7.1 It is common ground that D2 concerns an IC card with files and areas and means for storing access inhibit information for both the files and the areas. It is however in dispute whether D2 discloses "second inhibiting means ... for inhibiting a read access operation to the areas depending to the specific file by the inhibit information stored in said second storage means when the access operation to the specific file is not inhibited by the inhibit information stored in said first storage means".

During the "normal life" of a card (D2, fig. 1.4) files are typically not invalidated so that, in the words of claim 1, "the access operation to /a/ specific file is not inhibited". At the same time read accesses to Working Files (corresponding to the areas in claim 1) may be inhibited due to the parameters WRP and RP. Thus, the "second inhibiting means" as defined in claim 1 is known from D2.

7.2 On the other hand, the "first inhibiting means" in claim 1 is not known from D2, something which the appellant acknowledges. In other words, the prior art does not offer the possibility of inhibiting read operations on an entire file including the areas associated with it. Other operations, such as writing, may be inhibited, but reading is expressly permitted in D2. This was also the difference found by the opposition division in the decision under appeal (paragraph 13.1, second sentence).

7.3 Thus, the subject-matter of claim 1 is new.

8. Inventive step

8.1 As mentioned above, the only difference between the invention and D2 is that an inhibited file cannot be read. This is a matter of programming: in order to go from D2 to the invention it was necessary to remove the operation "read data" from the list of permitted operations on page 5.37. That this was easily done is self-evident, and indeed the respondent has acknowledged that the skilled person could have arrived at the invention from D2 (see point 8.7 below).

8.2 In the present case the Board finds the use of the problem-solution approach essential. The opposition division did not formulate a technical problem, but held that D2 taught away from the invention because of the expressly mentioned possibility of reading invalidated files. The Board notes that although this kind of argument can serve to strengthen a problem- solution analysis, it cannot replace it. The mere fact that there is a difference between the invention and the prior art only means that the invention is new. The examination for inventive step, however, must take into account the technical effect obtained. This is conventionally done by formulating a suitable technical problem which the invention solves.

8.3 D2 states on page 5.37 that invalidation is a procedure that permits to "end the life" of a Data File. It is not said why an invalidated file must be readable. The appellant has argued that this is something which merely depends on the circumstances: it is the issuer of a card (typically a bank) rather than the manufacturer who decides whether or not a file is to contain data which permanently require read protection. Thus, only the nature of the data matters. But according to long-standing jurisprudence data representing "information per se" cannot form the basis of an invention (cf. T 163/85, OJ EPO 1990, 379).

8.4 This point is crucial for determining the technical problem. The person skilled in the art in the sense of Article 56 EPC is in this case a microprocessor programmer. He acts on instructions from the card issuer. If the issuer finds it unacceptable that an invalidated file can be read because, due to the nature of the card application and the information stored, all the data in the file always need protection, then it is up to the programmer to change the program so that reading is not permitted.

8.5 The technical problem was therefore to modify the card known from D2 so that an invalidated file cannot be read. The solution consisted in re-programming the processor correspondingly, which was straight forward.

8.6 The respondent has argued that the technical problem instead consisted in replacing the "enormous black lists" created by banks to prevent some card-holders from carrying out transactions (see the specification, col. 1). But this problem also involves the non- technical aspects mentioned above: it is not a technical consideration whether cards need to be inhibited or not or whether the data stored on inhibited cards should be readable or not. Only how this is done might be technical.

8.7 The respondent has further referred to the "could/would approach" and argued that D2 contains no information which would urge the skilled person to modify the known card in accordance with the invention.

The Board agrees that nothing in D2 suggests that the card should be modified. It only shows how it may be modified if this were necessary. But, as explained above, no technical considerations were necessary to decide that the card needed modification.

This conclusion is not in contradiction with the "could/would approach". In decision T 2/83 (OJ EPO 1984, 265), in which this jurisprudence was developed, the invention concerned the addition of a barrier to a layered tablet. The deciding Board stated (see point 7 of the reasons; italics added):

"The question regarding the inventive step, in relation to the modification of the layered tablet of the state of the art as suggested by the present applicants, is not whether the skilled man could have inserted a barrier between the layers but whether he would have done so in expectation of some improvement or advantage. Since the Yen tablet was, on the face of it and from what was assumed in view of its commercialisation, a satisfactory answer to the problem of undesirable migration, the addition of a barrier would have appeared superfluous, wasteful and devoid of any technical effect. In view of the recognition that a barrier has, after all, a substantial effect, the outcome was not predictable and the claimed modification involves an inventive step on this basis."

It can be seen that the decision that a barrier should be added was based on observations of the properties of the known tablet, ie technical considerations, which naturally had to be taken into account when assessing the inventive activity. This is fundamentally different from the present case where the decision not to permit stored data to be read out is arbitrary from a technical point of view. Thus, the "could/would approach" only applies if the "would" part involves technical considerations. If it does not, the fact that the invention could be arrived at is sufficient to render it obvious.

8.8 The Board is therefore of the view that the subject- matter of claim 1 does not involve an inventive step.