Re: First public WD of "Architectural Principles of the World Wide Web"

** Reply to message from Paul Prescod <paul@prescod.net> on Mon, 02 Sep 2002
21:47:00 -0700

> "Anthony B. Coates" wrote:
> > http://www.w3.org/2001/XMLSchema
> > refers to a compound resource (if any) that identifies the whole of W3C XML
> > Schema, while
> > http://www.w3.org/2001/XMLSchema#int
> > identifies the integer datatype within W3C XML Schema.
> You say that that identifiers the "integer datatype".

Well, it is really the W3C XML Schema Datatypes document that says this, not
just l'il ol' me.

> If we were forced to state that in a formal way we might say that
> "http://www.w3.org/2001/XMLSchema#int has rdf:type dataType"
> But the XPointer specification disagrees with you. The XPointer
> specification says that *if* the XML Schema document is XML then 
> http://www.w3.org/2001/XMLSchema#int has rdf:type XMLElement
> I think that this is a pretty fundamental disagreement.

It would come as no surprise to me if RDF & XPointer disagree on this, but
there may be a valid reason for that.  XPointer isn't about concepts, it is
about locations.  So naturally XPointer would identify the location as an XML
element.  However, this isn't inconsistent with that XML element being used to
represent a concept (in this case, a data type).  It may be that RDF doesn't
cope with this well (and I'm not the one to say), but perhaps RDF needs to
support the fact that the "type" that users assign to something can depend on
the context in which they refer to it.

	Cheers,
		Tony.
====
Anthony B. Coates, Information & Software Architect
mailto:abcoates@TheOffice.net
MDDL Editor (Market Data Definition Language)
http://www.mddl.org/

Received on Tuesday, 3 September 2002 11:17:37 UTC