W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > October to December 2002

Re: xs:QName in Schema 1.1

From: Henry S. Thompson <ht@cogsci.ed.ac.uk>
Date: 11 Dec 2002 08:55:01 +0000
To: "Kay, Michael" <Michael.Kay@softwareag.com>
Cc: www-xml-schema-comments@w3.org
Message-ID: <f5bd6o9hr4q.fsf@erasmus.inf.ed.ac.uk>

"Kay, Michael" <Michael.Kay@softwareag.com> writes:

> > 
> > > Here are some suggestions:
> > > 
> > > 1. Clarify whether the namespace URI of a QName that is 
> > written with 
> > > no prefix is the null namespace or the default namespace, 
> > or whether 
> > > it's entirely up to the application to decide. Possibly, 
> > allow this to be controlled via a facet.
> > 
> > Where is clarification required -- I wasn't aware there was 
> > any area where is was not clear that unprefixed QNames are 
> > default-namespace qualified.
> The only thing that Schema Part 2 (section 3.2.18) says about the mapping of
> the lexical space to the value space for QNames is the Note: "The mapping
> between literals in the ·lexical space· and values in the ·value space· of
> QName requires a namespace declaration to be in scope for the context in
> which QName is used."

This will be clarified in a forthcoming erratum, sorry.

> It says nothing about whether the mapping is done using the conventions for
> element names (no prefix => default namespace) or the conventions for
> attribute names (no prefix => no namespace).
> One of the first applications to make extensive use of QNames in attribute
> values was XSLT 1.0, and it uses the (no prefix => no namespace) convention,
> so I am surprised to hear you say that Schema only supports the (no prefix
> => default namespace) convention. This certainly confirms my belief that it
> needs to support both: and above all, to be clear as to what it does
> support. 

This is, as I'm sure you know, a bit of a religious war issue.

The way in which QName is used in Part 1 makes clear that it's the
"element convention" which applies as far as XML Schema is concerned.

The fact that XPath uses the attribute convention for the names of
elements in paths is what's broken, in my opinion, and I would oppose
any move to change QName.

A request to consider adding a PName type, like QName but ignoring the
default declaration, would certainly be in order, however.

  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
          W3C Fellow 1999--2002, part-time member of W3C Team
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht@cogsci.ed.ac.uk
		     URL: http://www.ltg.ed.ac.uk/~ht/
 [mail really from me _always_ has this .sig -- mail without it is forged spam]
Received on Wednesday, 11 December 2002 03:54:59 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:50:00 UTC