RE: timbl-03 collection clutter

Regarding my action...

ACTION 20030425#5 danc figure out the cost to WebOnt of accepting
timbl-03

where timbl-03 is short for
http://www.w3.org/2001/sw/RDFCore/20030123-issues/#timbl-03

The relevant bit of the OWL spec is:

"the ontologies in O, taken together, provide a type for every
individual ID;"
 -- Web Ontology Language (OWL) Abstract Syntax and Semantics
Section 4. Mapping to RDF Graphs
 http://www.w3.org/TR/owl-semantics/mapping.html#4.1
http://www.w3.org/TR/2003/WD-owl-semantics-20030331/mapping.html#4.1

And without the rdf:List triples, the collection syntax
doesn't meet that constraint.

Jeremy and Peter discussed the cost of relaxing that constraint,
and it seems, to me, to be considerable: the details of the
mapping of owl-dl-abstract-syntax to RDF graphs are somewhat
subtle and impact many parts of the S&AS spec, especially
the semantic layering bits.
http://www.w3.org/2001/sw/WebOnt/webont-issues.html#I5.3-Semantic-Layering

Meanwhile, note that the exact details of the mapping
to RDF graphs were not decided by consensus; there is
an outstanding objection (from JJC/HP).
http://www.w3.org/2001/sw/WebOnt/webont-issues.html#I5.26-OWLDLSyntax


In sum, the rdf:List triples do serve a purpose in the graph
syntax of OWL; a purpose which rdfs:domain/rdfs:range won't
help with.


On Tue, 2003-05-06 at 03:24, Jeremy Carroll wrote:
> > I personally agree
> > with Tim that these (type List) triples are redundant, and I would
> > bet that the OWL stuff, even if it does need them, could be trivially
> > rewritten so that it did not. I was hoping that we could get back to
> > Peter and find out whether he thought the problem was serious.
>
> Peter punts the problem back to me - since I am the most recent person on
> record as arguing for type triples in general (in OWL DL).
> 
> Here are my thoughts:
> 
> 1: this is merely aesthetic either way
>    People who don't/do want the type triples can have what they want simply
> by not using the rdf:parseType="Collection" syntax (whichever way we
> decide).
> 
> 2: the triples *are* redundant
>    It is easy to automatically add them; or for that matter, automatically
> remove them.
> 
> 3: the construct was added for OWL, and so the aesthetic judgement as to
> whether they should be there or not in the convenience
> rdf:parseType="Collection" syntax, should be made by WebOnt.
> 
> 4: Given that in OWL DL and OWL Lite many type triples are needed, it is, in
> my opinion, more aesthetic, to have these type triples also for Lists. The
> need for type triples is principally to distinguish the various types of
> user defined in Property in OWL Lite and OWL DL.
> 
> Jeremy
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Tuesday, 6 May 2003 10:13:09 UTC