- From: Dan Connolly <connolly@w3.org>
- Date: 17 Oct 2002 15:43:45 -0500
- To: Patrick Stickler <patrick.stickler@nokia.com>
- Cc: ext Jeremy Carroll <jjc@hpl.hp.com>, w3c-rdfcore-wg@w3.org
On Thu, 2002-10-17 at 01:07, Patrick Stickler wrote: > ----- Original Message ----- > From: "ext Jeremy Carroll" <jjc@hpl.hp.com> > To: <w3c-rdfcore-wg@w3.org> > Sent: 16 October, 2002 15:36 [...] > > XSD says: > > http://www.w3.org/TR/xmlschema-2/#string > > I consider the XML Schema position to be far too extreme. Yes, some > writing systems may not be expressable as a plain Unicode string, > but many, of not most, are, and as you point out, there is no reason > why there could not exist a subtype of xsd:string which includes > a specialized lexical grammar to provide for such writing systems. > In fact, I would argue that an XML literal, in RDF at least, is > really a subtype of xsd:string, with a lexical grammer that reflects > lexical forms of XML infosets. Hm... that argument suggests that integer and date are subtypes of xsd:string too, with a lexical grammar that reflects numerals and date-strings. XML infosets are not strings any more than integers are. > > IMO, XML Schema got it wrong. I disagree. Moreover, XML Schema followed the advice[1] of the I18N WG on this; I recommend we do likewise. If we do otherwise, I'll be obliged to get the I18N WG to confirm that we had a good reason for doing so. [1] I can't seem to find records of their advice. But I'm pretty sure it's there, and I will have to check with the I18N WG, as well as the XML Schema WG, if we do otherwise. -- Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Thursday, 17 October 2002 16:43:10 UTC