W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > September 2002

RE: Datatyping: moving away from "literal as 3-part thing" to "literal as dt+opaque bit"

From: <Patrick.Stickler@nokia.com>
Date: Wed, 4 Sep 2002 13:09:11 +0300
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B5FBABF@trebe006.europe.nokia.com>
To: <bwm@hplb.hpl.hp.com>, <jjc@hpl.hp.com>, <w3c-rdfcore-wg@w3.org>



> -----Original Message-----
> From: ext Brian McBride [mailto:bwm@hplb.hpl.hp.com]
> Sent: 04 September, 2002 12:48
> To: Stickler Patrick (NMP/Tampere); jjc@hpl.hp.com;
> w3c-rdfcore-wg@w3.org
> Subject: RE: Datatyping: moving away from "literal as 3-part thing" to
> "literal as dt+opaque bit"
> 
> 
> 
> At 09:24 04/09/2002 +0300, Patrick.Stickler@nokia.com wrote:
> 
> [...]
> 
> >I am asserting that the xml:lang code is part of the *RDF* 
> representation
> >of the literal
> 
> I am not clear on what basis this assertion is being made.  
> Have I missed 
> the point, or is this not the very issue that is being "discussed".
> 
> Perhaps being clear what the issue is and discussion of pro's 
> and cons 
> might be more productive than dogmatic assertion of 
> individual viewpoints.
> 
> I think the question might be:
> 
>    Should the abstract syntax of a datatyped literal have an xml:lang 
> component?

I did not know this was an open issue. Has there been some
change to the decisions made during the Bristol f2f?

> So far, I have seen:
> 
>    1) If its not then one can't usefully say:
> 
>       <rdfs:label rdf:lang="en" rdf:datatype="...string">foo</...
>       <rdfs:label rdf:lang="fr" rdf:datatype="...string">bar</...

Well, there would be other means to say that, and I'd myself
prefer to see that information as explicit triples in the
graph rather than encapsulated in the literal itself, but
I'm trying to write the datatyping spec according to my
understanding of what the WG has decided thus far.

If that has changed, please someone let me know (with 
pointers to any relevant minutes or other documents).

>    2) the xml:lang component has no effect on the mapping 
> from the lexical 
> form to the value.

True. Only the lexical form is meaningful to the L2V mapping,
and only the unicode string component of a literal alone
constitutes a lexical form.

> On this last point, do we expect that always to be the case?

I see no reason to expect it would change.

> What would we 
> do if a new version of schema datatypes did take xml:lang 
> into account?  

Slap them ;-)  And then decide whether to follow suit or
maintain a divergent view.

> Do 
> we know they are unlikely to do this?

Can't say. But the current XML Schema specs seem pretty clear
on the issue, so any change would likely be motivated by a 
very good reason, in which case, we'd do well to consider
such motivations ourselves (or some future RDF WG would).

Patrick


> Brian
> 
> 
Received on Wednesday, 4 September 2002 06:09:13 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:50:56 EDT