W3C home > Mailing lists > Public > public-swd-wg@w3.org > January 2007

Re: [SKOS] structuring f2f discussion

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Tue, 16 Jan 2007 20:11:28 +0100
Message-ID: <45AD2360.2060707@few.vu.nl>
To: Alistair Miles <a.j.miles@rl.ac.uk>
CC: SWD WG <public-swd-wg@w3.org>

Hi Alistair,

Thanks for the contribution. I think indeed we can do like that. I have 
remarks inline...

> Hi,
> Some rough thoughts on how to structure SKOS discussion at the f2f ...
> At a higher level, we could start developing an overview of the 
> different *types* of application described in the use cases we have. 
> This could help us to define some application "patterns" i.e. general 
> scenarios that illustrate our expected application context for SKOS - 
> these general scenarios could even be used as a preface to the 
> requirements in the UCR doc?
> This could also help us to highlight patterns we expected to find but 
> that aren't yet represented by the use cases we have.
> E.g. an application pattern might be something like: retrieval of 
> multimedia objects, manually indexed using a controlled structured 
> vocabulary.

The problem is that it seems at a first thought unlikely that we find 
unexpected patterns, since nobody explicitly proposed patterns in 
addition to the ones you sent around in [1]. But this shall however not 
be a motication for not applying the first part of your proposal, which 
is definitely relevant.

> At the lower level, we could start working towards a set of 
> representational test cases for SKOS. I imagined that the extracts of 
> vocabularies taken from the accepted use cases could provide the basis 
> for a set of test cases - i.e. if it is possible to generate the 
> extract from a SKOS representation then the test is passed.
> This means that we would need vocabulary extracts to cover all of the 
> major vocabulary features and styles we expect SKOS to be able to 
> represent. This discussion might help us highlight where we need to 
> return to use case submitters and ask for more detail, and/or where we 
> need to approach specific vocabulary owners how haven't submitted a 
> use case, for permission to use an extract from their vocabulary.

OK for this, which I think is in line with what Daniel thought about 
when mentioning metadata requirements (and being given an action on 
that) in the teleconf. I would also mention that this subject is roughly 
covered by R1 to R6 in my draft list [2]
Concerning vocabulary extracts, I think we have tried to keep as much 
was possible from the orginal contributions in the summed up use cases. 
The detailed ones have everything, so we'll know quickly if we have to 
go back to the community.

I would add that we will have to sort out the other types of 
requirements, that is the ones that do not concern only the 
representation constructs offered by SKOS. They are quite diverse, some 
of them only vague design goals that we might not keep in the end...



[1] http://lists.w3.org/Archives/Public/public-swd-wg/2006Oct/0022.html
[2] http://www.w3.org/2006/07/SWD/wiki/CandidateReqList
Received on Tuesday, 16 January 2007 19:11:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:07:48 UTC