W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > April 2003

RE: nodeID on properties for refiication

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Tue, 8 Apr 2003 10:37:42 +0200
To: "Dave Beckett" <dave.beckett@bristol.ac.uk>, <w3c-rdfcore-wg@w3.org>
Message-ID: <BHEGLCKMOHGLGNOKPGHDAEHCCBAA.jjc@hpl.hp.com>


Strong oppose.

Specific problem: what does:

<rdf:Description>
  <eg:prop rdf:nodeID="a"/>
</rdf:Description>

generate.

It currently generates

_:xx eg:prop _:a .

where _:xx corresponds to rdf:Description.

Generic problem:
I agree this functionality would perhaps be useful, but RDF/XML suffers too
many 'useful' productions. We have got rid of enough, and clarified enough
to begin to make it possible to expect people to actually understand RDF/XML
(all of it); steps to add more abbreviated forms should be taken with great
care.

Jeremy

> -----Original Message-----
> From: w3c-rdfcore-wg-request@w3.org
> [mailto:w3c-rdfcore-wg-request@w3.org]On Behalf Of Dave Beckett
> Sent: 07 April 2003 16:29
> To: w3c-rdfcore-wg@w3.org
> Subject: rdf:nodeID on properties for refiication
>
>
>
>
> This is from the "giving yourself more work" category.
>
> It does not address a last call issue and is thus optional.
>
> Privately, I've been asked if rdf:nodeID could be allowed on
> properties so that statements could be reified with a blank node
> identifier rather than a URI being required.
>
> i.e. Presently allowed:
>
> ----------------------------------------------------------------------
> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>          xmlns:ex="http://example.org/vocab#">
>   <rdf:Description rdf:about="http://example.org/node">
>
>     <ex:foo rdf:ID="abc">foo</ex:foo>
>
>   </rdf:Description>
> </rdf:RDF>
> ----------------------------------------------------------------------
>
>
> making the 4 reification triples with reified node with URI <baseURI#abc>
>
> <baseURI#abc> rdf:type rdf:Statement .
> <baseURI#abc> rdf:subject <http://example.org/node> .
> <baseURI#abc> rdf:predicate <http://example.org/vocab#foo> .
> <baseURI#abc> rdf:object "foo" .
>
>
> Presently forbidden:
> ----------------------------------------------------------------------
> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>          xmlns:ex="http://example.org/vocab#">
>   <rdf:Description rdf:about="http://example.org/node">
>
>     <ex:foo rdf:nodeID="rei">foo</ex:foo>
>
>   </rdf:Description>
> </rdf:RDF>
> ----------------------------------------------------------------------
>
> making the 4 reification triples with blank node ID _:abc
>
> _:rei rdf:type rdf:Statement .
> _:rei rdf:subject <http://example.org/node> .
> _:rei rdf:predicate <http://example.org/vocab#foo> .
> _:rei rdf:object "foo" .
>
>
> Pros:
>   Flexibility
>   Consistency
>   No need to invent a URI
>
> Cons:
>   Another change
>   3 places to put rdf:nodeID - will this confuse?
>
>
> This would have low syntax doc impact, but would need some minor new
> words.
>
> In terms of implementing, I don't see it as too hard.
>
> Dave
>
>
Received on Tuesday, 8 April 2003 04:38:03 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:56:53 EDT