W3C home > Mailing lists > Public > xmlschema-dev@w3.org > November 2000

Re: Unprefixed attributes (was: Re: <attribute ref="thisNamespace:foo"/> not working?)

From: Henry S. Thompson <ht@cogsci.ed.ac.uk>
Date: 30 Nov 2000 12:00:03 +0000
To: Robert Braddock <stormwarden@bigfoot.com>
Cc: xmlschema-dev@w3.org
Message-ID: <f5bzoih4qm4.fsf@cogsci.ed.ac.uk>
I would only add to Martin Gudgin's useful reply that the crucial
difference between prefixe==qualified attributes and
unprefixed==unqualified attributes is that you _cannot_ assume that
two attributes with the same unprefixed==unqualified name on two
different elements have _anything_ in common.  This is a direct
consequence of fundamental properties of SGML and XML.  Consider

<!ATTLIST door key (yale|mortice|skeleton) #IMPLIED>
<!ATTLIST bibEntry key ID #REQUIRED>

Perfectly OK to have both of these in the same DTD.

The _intent_ of the Namespace REC was to allow for the provision of
so-called 'global' attributes, where it would _not_ be possible for
such attributes to have differences depending on their parent.  XML
Schema has done its best to reconstruct this intention by providing
'top-level' attribute declarations as the only way to get
prefixed==qualified attributes schema-valid in instances.

Yes, local attributes are _indirectly_ part of their parent's
namespace, but they can't be _directly_ in it, or the possibility of
two distinct local 'key' attributes as illustrated above would be

  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
          W3C Fellow 1999--2001, 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/
Received on Thursday, 30 November 2000 07:00:12 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:55:50 UTC