W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2001

Re: draft partitioning of the issues

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Wed, 27 Jun 2001 14:45:38 +0100
Message-ID: <3B39E382.BBDD42BA@hplb.hpl.hp.com>
To: Aaron Swartz <me@aaronsw.com>, rdf core <w3c-rdfcore-wg@w3.org>


Brian McBride wrote:


> So on the procedural question, I'm persuaded - we should consider this
> one.  Is anyone willing to write up a proposal?  It will need to cover
> questions like, given a literal - what exactly is the uri a parser should
> produce, questions of equality and equivalence (is <data:text/xml;A> the
> same as <data:text/plain;A>, effects on implementations (do they have
> to allow for larger URI's?)

I'm not sure that data uri's are in fact isomorphic with literals as
defined in m&s.  Consider:

<rdf:Description>
  <foo:bar xml:lang="en">weekend</foo:bar>
  <foo:bar xml:lang="fr">weekend</foo:bar>
</rdf:Description>

With literals this would be:

  _:a <foo:bar> "weekend"-"en" .   (I've invented this syntax for lang attrs)
  _:a <foo:bar> "weekend"-"fr" .

with data uri's it would be:

  _:a <foo:bar> <data:text/plain;weekend> .
  <data:text/plain;weekend> <xml:lang> <data:text/plain;en> .
  <data:text/plain;weekend> <xml:lang> <data:text/plain;fr> .

The later is not isomorphic with the former.

Brian
Received on Wednesday, 27 June 2001 09:48:03 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:37:18 EDT