Re: QNames?

Dear Joe,

The Query/XPath Data Model document uses the term "expanded-QName"
to refer to a namespace/localname pair.

In contrast we use "QName" to refer to a prefix:localname pair
to be consistent with the Namespaces Recommendation, and 
use the simple type xs:QName of XML Schema Part 2 to be
consistent with its definition.

I think this usage is actually inline with your suggestion.

Best regards,
Marton Nagy
XML Query WG


> -----Original Message-----
> From: Joe Kesselman [mailto:keshlam@alum.mit.edu]
> Sent: Thursday, March 14, 2002 12:54 PM
> To: www-xml-query-comments@w3.org
> Subject: QNames?
> 
> The Query/XPath Data Model document uses the term "QName" to
> refer to a namespace/localname pair. I've been told that
> this usage is pervasive throughout the XQuery/XPath2/XSLT2
> specifications.
> 
> Unfortunately, that's in conflict with estaliblished usage.
> The Namespaces Recommendation defines a QName specifically
> as a prefix:localname pair (see
> http://www.w3.org/TR/REC-xml-names/#ns-qualnames). The DOM
> Level 2 spec also uses that nomenclature. Since prefixes may
> be locally rebound (yielding more than one namespace per
> prefix) or bound to the same namespace URI (yielding the
> reverse), the two concepts are not interchangable, and
> attempting to overload the term is likely to cause
> confusion.
> 
> The Namespace spec does suggest a name for the ns/local
> pair: in appendix A3 they call this the Expanded name --
> more specifically the Expanded Element Type (ExpEType) and
> Expanded Attribute Name (ExpAName).
> (http://www.w3.org/TR/REC-xml-names/#ns-expnames).
> 
> Admittedly, this nomenclature is only introduced "for
> convenience", so it's a bit less official. But I strongly
> suggest that you either adopt ExpETye and ExpAName, or
> invent some similar term which subsumes both (ExpName,
> perhaps?), and use this new term whenever you intend
> ns/local -- leaving QName for its original purpose of
> prefix:local.

Received on Friday, 22 March 2002 15:18:50 UTC