- From: Andrew Layman <andrewl@microsoft.com>
- Date: Fri, 4 Jun 1999 13:30:16 -0700
- To: "'Paul Rabin'" <prabin@odi.com>
- Cc: Tim Bray <tbray@textuality.com>, xml-names-editor@w3.org
Unqualified attributes are intended to act exactly as they do in XML 1.0 without namespaces. They are specific to their element type. Two attributes with the same name on different element types are distinct attributes. Attribute names are not in any global namespace, most certainly they are not immediately in the same namespace as contains their element. Thanks for encouraging us to make our wording more clear. I cannot speak to what wisdom or foolishness might be in the XSLT or DOM drafts. -----Original Message----- From: Paul Rabin [mailto:prabin@odi.com] Sent: Friday, June 04, 1999 1:05 PM To: Andrew Layman Cc: Tim Bray; xml-names-editor@w3.org Subject: RE: namespace for unqualified attributes Andrew, Thanks. I did note the model described in the appendices, but am concerned that it is not reflected in the normative text. Is the intent that unqualified attributes have no namespace URI of their own, but that for purposed of matching two such attributes (for instance, in XSLT), the expanded attribute name should be used? I note that the XSLT draft uses the more familiar <URI, local name> model of expanded name for purposes of matching, and that this is also the model in the DOM level 2 draft. Paul
Received on Friday, 4 June 1999 16:30:35 UTC