W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > August 2002

Re: rdf:Alt's relationship to individual statements (fwd)

From: Frank Manola <fmanola@mitre.org>
Date: Mon, 26 Aug 2002 09:30:13 -0400
Message-ID: <3D6A2D65.6020201@mitre.org>
To: pat hayes <phayes@ai.uwf.edu>
CC: Dan Brickley <danbri@w3.org>, w3c-rdfcore-wg@w3.org

Comments below.

pat hayes wrote:

> 
>> Dan Brickley wrote:
>>
>>>
>>>  RSS and Dublin Core folk are trying to use rdf:Alt and it's painful to
>>>  watch :(
>>
> 
> Indeed.
> 
>>  >
>>
>>>  Which spec owns the problem of helping these folk? MT? Primer? Schema?
>>>
>>
>> Well, certainly the Primer is a candidate.  You might look at what the
>> Primer version at
>> http://www.w3.org/2001/09/rdfprimer/rdf-primer-20020725.html says about
>> Alt and see what you think.
>>
>>>
>> snip
>>
>>>
>>>  Considering the following example from
>>>  http://dublincore.org/documents/2002/04/14/dcq-rdf-xml/:
>>>
>>>  <rdf:Description rdf:ID="17">
>>>     <dc:identifier>
>>>      <rdf:Alt>
>>
>>  >      <rdf:li rdf:resource="http://dublincore.org/"/>
>>  >      <rdf:li rdf:resource="http://purl.org/dc/"/>
>>
>>>      </rdf:Alt>
>>>     </dc:identifier>
>>>  </rdf:Description>
>>>
>>>  Does this mean that one can say that the identifier of online:#17 is
>>>  http://dublincore.org/, or that its identifier is a collection of
>>>  alternative values of which http://dublincore.org/ is the first 
>>> prefernce?
>>
>>
>> It seems to me that it says the latter.  Anyone disagree?
> 
> 
> I don't think we can say either of these. It certainly isn't the first; 
> but if we say the second, then the identifier is a collection, so its 
> definitely not either of http://dublincore.org/ or http://purl.org/dc/  
> (neither of which are collections.)


Huh?  Neither of the URIs is a collection, the Alt is a collection, right?


> 
> I think the best way to say it is that its means that the identifier is 
> EITHER http://dublincore.org/ OR http://purl.org/dc/, but we don't know 
> which it is.
> 


It seems to me we need to be clear here about whether we're going to use 
the special "Alt" semantics associated with the Alt collection or not. 
If we take the model theory intepretation, what Pat suggests here seems 
to be the correct interpretation, but that's not the "intent" (at least 
as suggested by M&S).  The Primer characterizes Alt has a kind of "hint" 
as to how the contents of the collection are supposed to be interpreted, 
but notes that RDF doesn't (because it can't) actually enforce this 
interpretation.  But if we're going to ignore this type of thing 
entirely in telling people what things like Alt mean (or, at least, how 
they might use it), it seems to me we'd be better off eliminating it 
entirely.

--Frank




-- 
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 Monday, 26 August 2002 09:17:34 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:50:28 EDT