If it embeds the authority, there are two 
pieces of information there and how to 
reconcile which is in effect or both is 
still a processing issue.  You are saying 
that there is no RDF System specification 
and that interoperability of such systems 
is left to the vendors to imagine.

Once again, given the rules of the game, 
the biggest vendor wins.  If the vendor 
chooses not to support it, the Semantic 
Web based on RDF processing collapses or 
remains a niche.   Ok. 

Being silent about such things (is outside 
the scope of RDF) is proving as it did in XML 
Namespaces to be a naive approach to 
specifications.   I understand what you 
are saying, but as we learned in SGML, 
it just postpones work to be done 
elsewhere and engenders non interoperable 
systems.

Data is portable.  Systems interoperate. 

len


From: Bill de hOra [mailto:dehora@[...].net]

That's to misunderstand RDF (or give it too much credit, take your
pick). URIs are handy because we can create new unique ones with
minimal centralization and fuss. In RDF they're just globals, are
opaque, and in that respect there is no choice. Provenance is an
important use case for RDF. 

Dereferencing URI refs for more information is outside the scope of
RDF. There isn't an 'official' processing model for RDF, its
standardization to date has been careful to avoid talking about
processing, beyond what entailments can be made via the model
theory and the minimal amount of things you say or imply when
specifying serialization syntaxes like the XML.

-----------------------------------------------------------------
The xml-dev list is sponsored by XML.org <http://www.xml.org> , an
initiative of OASIS <http://www.oasis-open.org> 

The list archives are at http://lists.xml.org/archives/xml-dev/

To subscribe or unsubscribe from this list use the subscription
manager: <http://lists.xml.org/ob/adm.pl>