What is an "XML Namespace"?

In recent posts to public-qt-comments@w3.org 
(http://lists.w3.org/Archives/Public/public-qt-comments/2003Aug/0004.html) and <A HREF="mailto:www-xml-schema-comments@w3.org">www-xml-schema-comments@w3.org</A> 
(http://lists.w3.org/Archives/Public/www-xml-schema-comments/2003JulSep/0031.ht
ml) I drew attention to a seeming discrepancy between the definition of an 
XML namespace in the Namespaces in XML 1.0 Recommendation (where it is limited 
to element type names and attribute names) and the use of "XML namespaces" in 
W3C XML Schema, XQuery, XPath 2.0 and XSLT 2.0 to contain datatype and function 
names.

I appreciate that the seeming discrepancy may be "put right" by altering the 
definition of an XML namespace in the Namespaces in XML 1.0 Rec (and the 1.1 
CR).

However, since the question of what an XML namespace might or might not 
legitimately contain may impact on many XML-related technologies as well as 
potentially impacting on current TAG namespace-related considerations I thought it 
appropriate that TAG consider whether they should define what are the limits of 
what an "XML namespace" may legitimately contain.

Andrew Watt
"XHTML 2.0 - the W3C leading the Web to its full potential ... to implement 
yesterday's technology tomorrow"