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

Re: editorial tweak to Semantics

From: Frank Manola <fmanola@mitre.org>
Date: Sun, 17 Aug 2003 16:18:44 -0400
Message-ID: <3F3FE324.8000301@mitre.org>
To: pat hayes <phayes@ihmc.us>
CC: w3c-rdfcore-wg@w3.org

pat hayes wrote:

> Frank Manola wrote:

>
>> Argh!  I hope the Primer isn't misleading on this question, because I 
>> don't think your last sentence is accurate.  What I try to say is that 
>> rdf:ID provides a way of indicating that, in addition to generating 
>> the usual triple, a reification of that triple with a specific URIref 
>> (generated from the rdf:ID value) is to be generated as well.  It 
>> doesn't (I don't think) "attach a URI reference to a triple in an 
>> RDF/XML document".
> 
> 
> Well, isn't that attachment done by the fact of the URI appearing in the 
> rdf:ID in the document in question? But OK, I could rephrase it thus:
> 
> (In the RDF/XML syntax described in RDF/XML Syntax Specification 
> (Revised) [RDF-SYNTAX] the use of rdf:Id with a URI reference as value 
> creates a reification of the enclosed triple, with that URI reference as 
> the reified triple.)
> 
> Would that be better? Its more awkward but maybe more accurate.
> 
> Pat
> 


Pat--

What I'm getting at is what I say later in the Primer reification section:

"Using rdf:ID as shown in Example 20  generates the reification 
automatically, and provides a convenient way of indicating the URIref to 
be used as the subject of the statements in the reification. Moreover, 
it provides a partial "hook" relating the triples in the reification 
with the piece of RDF/XML syntax that caused them to be created, since 
the value triple12345 of the rdf:ID attribute is used to generate the 
URIref of the subject of the reification triples. However, this 
relationship is once again outside RDF, since there is nothing in the 
resulting triples that explicitly says that the original triple had the 
URIref exproducts:triple12345 (RDF does not assume there is any 
relationship between a URIref and any RDF/XML that it might have been 
used or abbreviated in)."

That is, if you look at the graph (Figure 17), there's nothing that 
connects the triple described by the RDF/XML ("the enclosed triple") to 
the URIref used in the reification triples (generated from the rdf:ID 
attribute value).  So there's sort of an "attachment" between the 
generated graph and some RDF/XML, but the interpretation of that 
attachment is still strictly outside of RDF.  I just wouldn't want 
anyone to interpret this parenthetical remark (whatever text it 
contains) as an exception to or qualification of the original statement 
that there isn't really a connection defined by RDF.  This would be an 
application-dependent convention that related (in a specific way) the 
resulting graph to the RDF/XML used to generate it;  and this 
information wouldn't necessarily be available to a second party 
receiving the graph.

NB:  "rdf:Id with a URI reference as value" isn't strictly accurate 
either.  rdf:ID ("ID" in caps) doesn't take a URIref as a value;  its 
value is used (as a fragment id) to *construct* a URIref.

--Frank




>> --Frank
>>
>> pat hayes wrote:
>>
>>>
>>> After reading Frank's Primer text,  I have changed some text in the 
>>> reification section 3.2.1 of Semantics from this:
>>>
>>> "Although RDF applications may use reification to refer to triple 
>>> tokens in RDF documents, the connection between the document and its 
>>> reification must be maintained by some means external to RDF. RDF 
>>> syntax provides no means to 'connect' an RDF triple to its 
>>> reification. "
>>>
>>> to this:
>>>
>>> "Although RDF applications may use reification to refer to triple 
>>> tokens in RDF documents, the connection between the document and its 
>>> reification must be maintained by some means external to the RDF 
>>> graph syntax. (The RDF/XML syntax described in RDF/XML Syntax 
>>> Specification (Revised) [RDF-SYNTAX] provides rdf:Id to attach a URI 
>>> reference to a triple in an RDF/XML document, and create the 
>>> corresponding reification.) "
>>>
>>> I hope the WG does not object to this; I think the rather sharp 
>>> negative was misleading.
>>>
>>> Pat
>>>
>>
>>
>> -- 
>> Frank Manola                   The MITRE Corporation
>> 202 Burlington Road, MS A345   Bedford, MA 01730-1420
>> mailto:fmanola@mitre.org       voice: 781-271-8147   FAX: 781-271-875
> 
> 
> 


-- 
Frank Manola                   The MITRE Corporation
202 Burlington Road, MS A345   Bedford, MA 01730-1420
mailto:fmanola@mitre.org       voice: 781-271-8147   FAX: 781-271-875
Received on Sunday, 17 August 2003 15:54:01 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:59:39 EDT