Re: user-defined datatype (facet) support needed?

On 2002-04-30 21:02, "ext Dan Connolly" <connolly@w3.org> wrote:

> XML Schema has not only the 9 or so
> primitive types, but also bunch of facets
> for user-defined types ala "integers
> between 5 and 10".
> 
> I don't think any of the recent datatyping
> proposals allows me to express that datatype.
> 
> Is that a problem[ftf]?
> 
> I suggest, tentatively, that it is.
> 
> [crud; I was gonna write a bit more, but
> I've got another meeting. I think I better
> send this now, though, before I put it
> off for another three weeks...]
> 
> [ftf] http://www.w3.org/2001/sw/RDFCore/20020225-f2f/#d-2002-02-26-3

I don't see any problem here, since the semantics of the datatype
are fully opaque to RDF -- in the same manner that URI Scheme
semantics and syntax are opaque to RDF.

You are free to define a datatype in XML Schema using any of the
facets available, and then associate that datatype with nodes in
an RDF graph. You don't define the datatype in RDF, you simply
specify which datatype serves as the basis for interpretation.

Your extra-RDF application will need to understand the actual
datatype to make use of or determine the validity of RDF statements
referencing that datatype, and it may use XML Schema to do that,
by evaluating the definition of the datatype as expressed in
XML Schema to determine if a given lexical form is valid, etc.

Likewise, you can use any other language to define the semantics
and lexical space of a datatype, even English ;-) and still use
that datatype with RDF, so long as it is identified by a URI.

That's what it means for RDF to both support XML Schema datatypes
while also supporting other datatype formalisms.

Eh?

Cheers,

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com

Received on Thursday, 2 May 2002 03:16:27 UTC