- From: Dan Brickley <Daniel.Brickley@bristol.ac.uk>
- Date: Fri, 11 Aug 2000 10:33:53 +0100 (BST)
- To: Pierre-Antoine CHAMPIN <champin@bat710.univ-lyon1.fr>
- cc: Dan Connolly <connolly@w3.org>, www-rdf-interest@w3.org
On Fri, 11 Aug 2000, Pierre-Antoine CHAMPIN wrote: > Dan Connolly wrote: > > Huh? Which text from XML-NS says anything about how the namespace > > name is used, let alone that it can't be combined with > > some other piece of information via a function that's not > > invertable? > > Although non normative, the annex A suggests to use a URI/name pair for extended qnames. > Anyway, XML namespaces are designed for XML vocabularies (element names and attribute names) rather than RDF vocabularies (URI refs), and their use in RDF (primarily the non-invertible cancatenation function) seems kind of twisted to me... That is like saying that since XML documents are intrinsically order preserving (which they are), the RDF strategy of ignoring this information and tunnelling unordered data-oriented structures is somehow wrong. If so, RDF is in good company. It's XML's job to facilitate all kinds of interesting applications that sit on top of it; as such the XML infoset preserves a lot of structure which higher level apps might sometimes throw away (eg. whitespace, document ordering, Qname internal structure). That doesn't make either wrong, they're just different information models. --dan
Received on Friday, 11 August 2000 05:34:10 UTC