Re: Dublin Core, the Primer and the Model Theory

Jeremy Carroll wrote:

> 
> Summary:
>   - the current model theory
>     misarticulates the meaning of the triple:
>  <eg:doc1> <dc:creator> "John Smith" .
>   - many such triples occur both in the primer
>     and on the web.
>   - the model theory should be overhauled along the 
>     lines of Pat's simpledatatype2
>

snip

> 
> I believe that we will have great difficulty persuading the 
> Dublin Core community to stop using string literals to 
> represent real world entities. 


Lest anyone fail to make the obvious generalization here, it is not just 
the Dublin Core community that uses string literals to represent 
[actually, not "represent", but "denote" or "name"] real world entities. 
  Anyone who uses a relational database to store information about real 
world entities does the same thing.  In the Primer, there are examples 
in which an employee number is used to construct a URI that identifies 
the person in question.  But in reality that employee number would have 
been taken from a relational database as a string literal, and might 
well be used in literal form as the value of some RDF property.  If used 
as a dc:creator value, it wouldn't cause the specific mis-entailment 
cited by Jeremy (at least, not if kept within an appropropriate 
context!), but only because an "extra-model-theory" mechanism enforces 
(some) uniqueness constraint on the association of values to entities. 
But, as noted, RDF can't tell the difference between those kinds of 
literals and others.

--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 Thursday, 16 May 2002 07:30:06 UTC