Search This Blog

Labels

T 0117/10 - Tempus fugit

Calendar app

After two appeals and ten years of prosecution the applicant in this software case has now been awarded a grant (well, nearly..). 

The application concerns a calendar application with a novel cache function (a 'moving cache'). 

The application was refused for the first time in 2006. According to the (then) Examining division the implementation of a cache to 'represent one out of several straightforward possibilities from which a person skilled in the art would select, without the exercise of inventive skill, (...)'. In previous letters the cache management scheme was found 'to relate to pure business issues regarding the management of events, and the conflict notification functionality.' Though the latter consideration was not part of the decision.

With an appeal (T 1265/06) the applicant won a remittal on 15.04.2009. The board had found that 
'the examining division did not cite a single prior art document which was related to the use of a
cache in a handheld device' and that  the 'main request can thus be allowed on the condition that the
description is adapted accordingly.' The appeal decision did not include a order to grant though.


The Examining division cited new documents and in a decision on the file, rejected the application on 1.12.2009. In the new refusal, the Examining division argued that 'the application variables (located in the memory) which are loaded through a database fetch already are the cache. It is noted that such a database fetch is implicit to any database access.'


After the new appeal, part of which is reproduced below, the applicant has won a new remittal, this time with an order to grant.


Reasons for the Decision
1. The appeal is admissible and allowable since in the Board's judgement the decision under appeal does not prove to the necessary standard the lack of inventive step and there are no other objections against the application as it stands apparent from the file. This holds for the claims of the main request, the description as amended with the statement of grounds and the drawing as originally filed.

2. The Board agrees with the appellant that the new documents cited do not improve the relevance of the available prior art. In particular, the decision under appeal fails to substantiate the leading argument that use of a "moving cache" in a handheld computing device is prior art. In the preceding decision T 1265/06 (see point 5 of the reasons), the Board stated that "the examining division did not cite a single prior art document which was related to the use of a cache in a handheld device, let alone something like a 'moving cache' for implementing the calendar application on a handheld device". This factual situation has not substantially changed in spite of the efforts of the examining division to produce more relevant prior art.


3. Firstly, the Board does not agree that D3 implicitly discloses a calendar application cache (see point IV above). Even if it is acknowledged that an application-level variable could be used as a (high-level application) cache, it does not mean that such a variable is actually, quasi automatically, used as a cache. Document D11 cited by the examining division itself is very clear about that.

The present application explains that the term "cache" means "a section of a computing device's memory which retains certain data in order to speed up repeated access to the same data" (see the A-publication, paragraph 0020). This is essentially the ordinary meaning which the term has in the field of computer technology. The term implies specific functions, whether they are hardware or software implemented, which distinguishes it for example from a direct memory access technique. A cache replicates original data and holds them in storage for multiple access, which requires the implementation of suitable algorithms for fetching, placing, and replacing data. None of those functions are addressed in document D3, either explicitly or implicitly, nor is there any indication in document D3 that application variables should be used for caching purposes or that the content should be prefetched.

4. Moreover, the specific update process that makes up the "moving cache" is not a feature of some abstract scheme of calendar administration or of any other non-technical consideration, but serves the purpose of reducing the number of time-consuming database accesses and is thus a technical aspect of the invention relevant to inventive step. The circumstance that a specific ordering and grouping of calendar entries which have their rationale in non-technical considerations, are essential to this update process does not override its overall technical character. A per se non-technical feature which interacts with a technical feature in such a manner that a technical effect is achieved has, in this specific combination, to be taken into account in assessing inventive step.

5. The Board also reaches a different conclusion regarding how the skilled person would assess the disclosures about caches in the other documents.

The common general knowledge and the prior art documents cited in the European search report have already been considered as not relevant in the first appeal decision (see T 1265/06, point 5 f. of the Reasons).

New documents D4, D5, D8, and D11 are post-published and can thus not directly be used as prior art. D6, D7, D9, and D10, forming part of the prior art, do indeed address caching techniques to some extent.

Document D6 focuses on strategies for ensuring data availability such as caching or prefetching in mobile computing (see D6, paragraph 5 Conclusions and perspectives at page 6). The document mentions handheld devices in the opening paragraphs, but then goes on to deal only with "powerful laptop computers". It addresses calendar applications in a very cursory manner. Caching techniques are presented in a kind of summary of general strategies for addressing problems like off-line browsing or managing concurrent updates, without any specific reference to calendar applications. It explains "prefetching", but does not give any details about the data to be fetched. Even taking into consideration the fact that "[P]refetching algorithms attempt to guess what information will soon be needed" as disclosed in D9 on page 477, top of right-hand column, does not immediately suggest the idea of a moving cache. To then invoke some application of the general principle of "locality of reference" as mentioned, for example, in the second paragraph of (post-published) D8, to suggest that the skilled person might consider prefetching data either side of the retrieved date, is, in the Board's view, stretching the limits of obviousness too far. Thus, in the Board's judgement, it would not be obvious to provide a concrete caching strategy for calendar applications like the "moving cache" of the present invention.

The remaining documents do not provide any more relevant disclosure of caching techniques and also lack any calendarial context.

Hence, these documents do not provide the necessary incentive to combine the calendar application disclosed in document D3 with a caching method to arrive at the "moving cache" of the present invention.

(...)

Order
For these reasons it is decided that:
The decision under the appeal is set aside and the case is remitted to the first instance department for further prosecution with the order to proceed with the grant of a patent on the basis of the main request.

This decision has European Case Law Identifier: ECLI:EP:BA:2015:T011710.20150305. The whole decision can be found here (pdf). The file wrapper can be found here. Photo "Prague Astronomical Clock" by Juan Pedro Asencio Flores obtained via 500px under CC BY 3.0 license (no changes made).
 

Comments