Re: The RDF Approach to Indicating Language-In-Use

Hmm, been thinking about this one a while - something tickling my brain


At 10:15 AM -0600 11/6/03, pat hayes wrote:
>>At 12:59 31/10/03 -0500, Peter F. Patel-Schneider wrote:
>>>   > Well, since you ask, I imagine that we could produce a three-part
>>>>   statement:
>>>>
>>>>   1/ The SW meaning of a set of SW documents in a SW language is completely
>>>>      determined from the normative specification of the SW language and the
>>>>      contents of these SW documents.
>>>>
>>>>   2/ The meaning of a set of SW documents does not necessarily 
>>>>include any of
>>>>      the meaning of any other document, except for those SW documents whose
>>>>      meaning is explicitly required to be a part of the meaning of the SW
>>>>      documents by the normative specification of the SW language and the
>>>>      contents of these SW documents.
>>>>
>>>>   3/ Applications are free to augment this meaning, perhaps by 
>>>>including the
>>>>      meaning of other SW documents, but are prohibited from indicating that
>>>>      this augmented meaning is part of the meaning that comes from the SW
>>>>      language.

imagine the above being rendered in the proper MUST, SHOULD, etc. terms

>>>>
>>>>   So, as far as RDF is concerned, the meaning of a set of SW documents in
>>>>   RDF/XML is determined solely from the RDF graph that results from the
>>>>   parsing of these documents and is not dependent on the contents of
>>>>   any other document.   OWL extends this to bring in the meaning of
>>>>   imported documents.

and being "normative"

>>
>>I think this is fine, and useful, as a description of 'SW meaning'.
>>
>>But I'm not sure that SW meaning has sufficient "meaning" to 
>>usefully relate SW application behaviour to user's expectations. 
>>Under what circumstances can a user regard the output of a SW 
>>application as being correct, and when so, to what question is it 
>>the correct answer?  I don't see 'SW meaning' telling us any of 
>>this.
>>
>>Yet application writers need to understand how programs interact 
>>with the world of their users.  Maybe you're right that it's not 
>>our role (as technologists) to define that, but I think that we (a) 
>>are reasonably involved in the debate, and (b) should try to 
>>clarify the boundary (and your description of SW meaning appears to 
>>help do that).


then imagine a number of informative use cases showing expansions 
where URIs are followed or etc. in ways that do not violate the above.

Then imagine some test cases which say something like
  All tools MUST return ...
  Some tools MAY return ...
(and this is not in any way implied to be a closed set - so tools 
which include the MUSTs but have some extra also would be considered 
conforming

>
>In passing, I would like to avoid this terminology where we talk 
>about *kinds* of meanings, if we can possibly manage to do so. It 
>suggests a kind of botanical classification of meaning-species, and 
>this tends to encourage a kind of Balkanization which we already 
>have a tendency to fall into, where each of us with various agendas 
>feels compelled to protect the rights of one kind of meaning over 
>the other upstarts. Maybe this is unavoidable, but we should try to 
>avoid it as far as possible. I am convinced that these different 
>'kinds' of meaning are all aspects of one notion, and that we should 
>be able to find a way to make that clear. Blind men and the 
>elephant, you know the story.

Then stop imagining that these are called "meaning" or kinds of 
meaning or anything like that -- rather they are given some more 
proper title ("inferences legitimized by URI use in the RDF and OWL 
frameworks"  is a bit long).

Then imagine this appears in a document that also addresses some of 
the MIME issues for RDF/RDFS/OWL (what do we expect to be served up - 
if it is not differentiated then how a reasoner is expected to 
determine the language-in-use)

Such a document would seem to me to be address the TAG issue, would 
be a heck of a contribution, and generating would give this Task 
Force a couple of clear questions to attack....



-- 
Professor James Hendler			  http://www.cs.umd.edu/users/hendler
Director, Semantic Web and Agent Technologies	  301-405-2696
Maryland Information and Network Dynamics Lab.	  301-405-6707 (Fax)
Univ of Maryland, College Park, MD 20742	  240-277-3388 (Cell)

Received on Friday, 21 November 2003 09:31:22 UTC