RE: What is an RDF datatype?

> -----Original Message-----
> From: ext Brian McBride [mailto:bwm@hplb.hpl.hp.com]
> Sent: 30 January, 2003 19:32
> To: Stickler Patrick (NMP/Tampere); pfps@research.bell-labs.com;
> www-rdf-comments@w3.org
> Cc: graham Klyne; Jeremy Carroll
> Subject: RE: What is an RDF datatype?
> 
> 
> At 17:54 30/01/2003 +0200, Patrick.Stickler@nokia.com wrote:
> 
> [...]
> 
> >Facets are not part of RDF Datatypes.
> 
> My memory may be failing me, but I don't recall the WG 
> deciding this.  

Yes. Your memory is failing you ;-)

RDF Datatyping has no concept of facets as they are defined
by XML Schema. It only includes abstractions of a value space,
a lexical space and an N:1 mapping from lexical to value space
where N>=1. That's it. And that's compatible with all XML
Schema datatypes (though not identical).

Facets are mechanisms for defining the nature of the value space
of an XML Schema datatype and RDF Datatyping is agnostic regarding
how datatypes are defined. Only that they possess the above 
characteristics. Whether facets or any other mechanism is used
to define a datatype is irrelevant to RDF Datatyping. Facets are
XML Schema specific and have no definition in RDF Datatyping.

> I'd 
> like to hear from the editors of the concepts doc before we 
> regard this as 
> a statement of the WG position.

Fair enough, but with all due respect, if the editors were to 
disagree with the above, they would be mistaken. The recent
wording claiming that RDF Datatyping *is* XML Schema datatyping
is an editorial error, and one that I have pointed out several
times before and has had no presence in any of the previous
draft specs defining RDF Datatyping nor in any of the definitions
of RDF Datatyping voted on and adopted by the WG. One should not base the 
presence of facets in RDF Datatyping on that erroneous statement
(though it would be reasonable to do so, hence the seriousness of
the error) and the statement must be fixed and the issue clarified 
further as needed.

Regards,

Patrick

Received on Friday, 31 January 2003 01:41:17 UTC