W3C home > Mailing lists > Public > public-swd-wg@w3.org > April 2008

[SKOS] Comments on SKOS-XL proposal

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Tue, 08 Apr 2008 12:16:58 +0200
Message-ID: <47FB461A.3050104@few.vu.nl>
To: SWD WG <public-swd-wg@w3.org>, Alistair Miles <alistair.miles@zoo.ox.ac.uk>

Dear Alistair,

Aside to the core discussion on SKOS-XL [1], I'd have a small number of 
comments that are maybe less important. Maybe they can be useful for the 
next version...

Editorial:
- use skosxl as namespace?
- not using lll as variable name? (it messes a bit with other symbols)

There is also a bit trickeir issue, which I'm not sure about, with your 
"The 3 examples below are each inconsistent with the XL data model, 
because xl:literalForm is a functional property."
Actually the incoherence comes from the fact that functionality says 
that "foo" should be the same as "bar" which would raise an error. But 
does RDFS semantics for literals really includes that kind of statement?
At our level we can sure say that the example you give are incoherent, 
but maybe we should mention that this is because *we* say it to be so, 
independently of whatever would be in RDF specs.

Finally, I think I do approve of the story that explains what the 
'identity' of an instance of xl:Label is (combination of literal plus 
provenance infromation, e.g. one concept scheme). I would feel however 
delighted if there was a formal definition for it. But of course it's 
not possible, because the current XL solution does not introduce 
something like inScheme for instances of xl:Label.
Couldn't we introduce one such property, if this is so important for the 
instances of xl:Label?

Cheers,

Antoine

[1] 
http://www.w3.org/2006/07/SWD/wiki/SkosDesign/SKOS-XL?action=recall&rev=2
Received on Tuesday, 8 April 2008 10:17:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 April 2008 10:17:26 GMT