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

more about rdfms-difference-between-ID-and-about

From: Ron Daniel <rdaniel@interwoven.com>
Date: Fri, 29 Jun 2001 09:12:24 -0700
To: "RDF Core" <w3c-rdfcore-wg@w3.org>
Message-ID: <EMEKICCGFEKJFGKMFLEPKEDLCPAA.rdaniel@interwoven.com>
Hi all,

I had an action to bring up another point about
rdf:ID and rdf:about.

The idea of an isDefinedBy predicate has been mentioned.
While that would be an extension of the 1.0 M&S, and
thus out of scope for the clarification document, if we
combine rdf:ID and rdf:about now, we won't be able to add
such a thing later.

Is it important to add such a thing? Don't know, but I
do know of at least one situation where I've made use
of the distinction the M&S 1.0 spec makes between the
two. PRISM is a metadata specification
that makes extensive use of terms in controlled
vocabularies. For example, to indicate that a story
is about Greece, the PRISM description of that story
would contain:

  <rdf:Description rdf:about="story.txt">
      <dc:coverage rdf:resource=
                   "http://prismstandard.org/vocabs/ISO3166/GR" />
  [...]

Unless the system receiving such a description already knows
about the ISO3166 vocabulary, it can't do anything that is
human meaningful with that. Since PRISM expects there to be
many controlled vocabularies used, we can't assume that the
receiver will know all of them. So, PRISM shows a particular
RDF idiom:

  <dc:coverage>
    <pcv:Descriptor rdf:about=
                "http://prismstandard.org/vocabs/ISO3166/GR">
        <pcv:label xml:lang="en">Greece</pcv:label>
        <pcv:label xml:lang="fr">Grece</pcv:label>
    </pcv:Descriptor>
  </dc:coverage>

which allows some information (in this case, the human
readable label) about the controlled vocabulary
entry to be provided in-line. Sort of like pre-filling
a cache. Systems just have to know how to handle the
pcv:label predicate, i.e. 'show this to the user', a much
smaller task than having to know all possible vocabularies.

So note that the rdf:about attribute was used above. This was
to explicitly indicate that the in-line description was NOT
the authoritative description of the GR resource. The PRISM
spec says that when you do create a document that IS the
authoritative definition of that resource, the rdf:ID
attribute must be used instead.

So, this is one data point where the difference between the
two was used, based on what the M&S 1.0 spec says. Presumably,
in the future the implementations could be updated to insert
an isDefinedBy predicate if such a thing were created.

However, having provided that data point, I want to point
out that the question we need to address is whether such a
difference is valuable enough to be maintained. PRISM's world
would not collapse if that difference were erased. Also, it
would avoid the ugly situation where multiple graphs claim to
be the authoritative definition for a resource.

Regards,
Ron Daniel Jr.
Standards Architect
Tel: +1 415 778 3113
Fax: +1 415 778 3131
Email: rdaniel@interwoven.com 

Visit www.interwoven.com
Moving Business to the Web 
Received on Friday, 29 June 2001 12:14:11 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:37:19 EDT