Design Issue (2) - To concept or not to concept

I've added this issue to the discussion on the RDF Thesaurus wiki page
<http://esw.w3.org/topic/RdfThesaurus>

Here is a summary:-

2.2 Issue 2 - To concept or not to concept


A thesaurus is a collection of concepts. So for the core-vocab we need to
model abstract concepts in RDF. 

Option 1 - We define an rdfs:Class called soks:Concept. We use this to type
resources that are intend ed to refer to abstract concepts. 

Option 2 - We define no such class. We use some other way to determine
whether a resource is a concept or not, if at all we need to. 

=== Comments on Issue 2 === 

AJM>> 

At the recent SWAD meeting at HP, Chaals said (correct me if I'm wrong) in
RDF every resource with a URI that has a fragment identifier necessarily is
an abstract concept. Therefore we don't need a type for concepts. 

I say: 

1.	My reading of the debate & TimBL's writeups is that resources with a
<http://> uri and a frag id MAY (but NOT necessarily) refer to an abstract
concept. Resources with a <http://> uri and without a frag ID may NOT be an
abstract concept (must necessarily be a document). 
2.	If we have a soks:Concept class, we can type b-nodes as concepts. So
we can use reference by description to make statements about abstract
concepts without URIs. 
3.	It makes the format look nicer if it starts with 'Concept' rather
than 'rdf:Description' all the time. This may be a serious point for KOS &
DL people. 



CCLRC - Rutherford Appleton Laboratory
Building R1 Room 1.60
Fermi Avenue
Chilton
Didcot
Oxfordshire OX11 0QX
United Kingdom

Email:        a.j.miles@rl.ac.uk
Telephone: +44 (0)1235 445440

Received on Friday, 31 October 2003 11:22:24 UTC