- From: Jonathan Borden <jborden@mediaone.net>
- Date: Thu, 15 Feb 2001 11:37:45 -0500
- To: "Graham Klyne" <GK@ninebynine.org>
- Cc: <www-rdf-interest@w3.org>
Graham Klyne wrote: > At 06:13 AM 2/15/01 -0500, Jonathan Borden wrote: > >Without getting too too complicated, also remember that a URI reference is > >defined as: > > > >URI-reference := [absoluteURI | relativeURI] ["#" fragment-id] > > > >The idea of explicitly using a "#" to delimit the local name from the > >namespace URI in a typedNode qname is not half baked (IHMO) but grounded in > >RFC 2396 itself. > > But there's no requirement (in XML namespaces or in RDF) that the local > part be a fragment identifier. right. local names are defined by XML namespaces, fragment identifiers by RFC 2396. I'm simply suggesting a mapping of QName as defined in XML namespaces into URI reference as defined in RFC 2396. > > (Though, to confuse matters, there is a frequent presumption that they > often will be fragment identifiers, reinforced by the definition of ID= in > <description>s, and the way RDF schema encourages such things be defined.) > Not to confuse. This is exactly why the mapping is useful. -Jonathan
Received on Thursday, 15 February 2001 11:51:56 UTC