- From: Lee Jonas <lee.jonas@cakehouse.co.uk>
- Date: Thu, 8 Mar 2001 17:09:42 -0000
- To: "'Stefan Kokkelink'" <skokkeli@mathematik.uni-osnabrueck.de>, Lee Jonas <lee.jonas@cakehouse.co.uk>
- Cc: Aaron Swartz <aswartz@swartzfam.com>, RDF interest group <www-rdf-interest@w3.org>
I see - effectively an ID at the XML syntax level (allowing serialisation of
graphs within a document) and not at the RDF model level (i.e. not an RDF
resource identifier). Ok.
Regards
Lee
-----Original Message-----
From: Stefan Kokkelink [mailto:skokkeli@mathematik.uni-osnabrueck.de]
Sent: 08 March 2001 16:05
To: Lee Jonas
Cc: Aaron Swartz; RDF interest group
Subject: Re: Again: Anonymous Resources
Lee Jonas wrote:
...
> What would the rdf:anonymous_ID actually accomplish over rdf:ID? If you
> give a name to something it is no longer nameless.
The interpretation is different.
An RDF parser should be allowed to *change* the name
by a one to one mapping in the first case. Of course it should not
be allowed to do this with URI's given by rdf:IDs.
For example the following 'RDF' would be
equivalent
<rdfs:Class rdf:ID="MyClass">
<dc:creator>
<rdf:Description rdf:anonymous_ID="1">
<name>Stefan</name>
</rdf:Description>
</dc:creator>
</rdfs:Class>
<rdfs:Class rdf:ID="MyClass">
<dc:creator>
<rdf:Description rdf:anonymous_ID="2">
<name>Stefan</name>
</rdf:Description>
</dc:creator>
</rdfs:Class>
since there is an obvious bijection ;-).
Greetings,
Stefan
Received on Thursday, 8 March 2001 12:10:06 UTC