- From: Pierre-Antoine CHAMPIN <champin@cpe.fr>
- Date: Thu, 20 Jan 2000 09:15:08 +0100
- To: Sergey Melnik <melnik@DB.Stanford.EDU>, RDF Interest Group <www-rdf-interest@w3.org>
Sergey Melnik wrote:
> xml:lang is not represented in the RDF model according to RDF M&S 1.0,
> which is IMO a bug in the specs. Thus, the model generated for
>
> <rdf:Description about=".../x.html">
> <dc:Title xml:lang="fr">Bonjour</dc:Title>
> </rdf:Description>
>
> is just
>
> x.html --dc:Title--> "Bonjour"
>
> A way to keep the language information in the model would be:
>
> x.html --dc:Title--> 000123
> 000123 --rdf:value--> "Bonjour"
> 000123 --xml:lang--> "fr"
how funny :)
That's exactly the way I would treat ANY literal...
IMHO, the bug in the spec is deeper : it's about dichotomy between Literals and Resources.
Furthermore, they mention it as an issue to be resolved in future versions.
In my point of view, literals don't exist in RDF.
They are a SYNTACTIC shortcut, but should not appear in the MODEL.
Writing
<rdf:Description about=".../x.html">
<dc:Title>Bonjour</dc:Title>
</rdf:Description>
is the quick and easy way of writing
<rdf:Description about=".../x.html">
<dc:Title rdf:resource=".../title.txt"/>
</rdf:Description>
where title.txt contains the text "Bonjour".
Anyway, why should there be a modelling difference between those two statements ?
Since RDF litterals are part of a resource (the RDF document),
the parser could easily (?) compute their URI like x.rdf#the_literal
(in XLM syntax of RDF, I guess XPointer could be used),
so there is no need for the (resource,resource,literal) triple kind,
only (resource,resource,resource) would stand.
The fact that we want to express RDF statement about literals,
which is not possible in the curent models,
reinforces my opinion that they should be treated as URI instead !
Pierre-Antoine
--- Quid quid Latine dictum sit, altum viditur
Whatever is said in Latin sounds important.
Received on Thursday, 20 January 2000 03:18:16 UTC