W3C home > Mailing lists > Public > www-webont-wg@w3.org > February 2003

Re: WOWG: Agenda Feb 6 telecon

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Wed, 05 Feb 2003 17:49:06 +0000
Message-ID: <3E414E92.1030709@hpl.hp.com>
To: Guus Schreiber <schreiber@swi.psy.uva.nl>
CC: WebOnt WG <www-webont-wg@w3.org>

> 4.1 Concrete syntax of OWL Lite/DL
> 
> Concerns Typing of nodes and properties in the concrete syntax.
> Carroll will post message to the list as input for discussion.
> 

I am doing an on-going piece of work which will result in a proposal for 
many minor changes to the syntax, in order to get a cleaner expression as 
rdf triples.

Part of that which is baked, and ready for WG discussion is the following 
suggestion:

[[
For every node of the graph and every URI reference used as a property or 
datatype at least one of the following holds:
+ It is one of the built-in URI references from RDF, RDFS, OWL or XML 
Schema datatypes.
+ It is a literal.
+ It is the subject of at least one triple with predicate rdf:type; with an 
object other than rdfs:Class, rdf:Property, owl:DeprecatedClass, 
owl:DeprecatedProperty, owl:FunctionalProperty
]]

(I think the list of exclusions is complete, but I might have missed one or 
two - the intent is that the required explicit rdf:type triple permits the 
assignment of the uriref to one of the abstract syntax categories of uriref).


Benefits:

- easy to articulate
- easy to check

Drawbacks:

- more restrictive than needed
- some minor changes to mapping rules
- requires a class for owl:DataRange
- requires a class for annotations (either owl:AnnotationProperty or 
owl:DatatypeProperty or owl:ObjectProperty depending on orthogonal 
discussion about annotation semantics)

Example changes:

Current:
   <owl:Ontology rdf:about="">
     <dc:creator>Jeremy</dc:creator>
     <owl:priorVersion rdf:resource="old.rdf"/>
   </owl:Ontology>

New:
   <owl:Ontology rdf:about="">
     <dc:creator>Jeremy</dc:creator>
     <owl:priorVersion>
         <owl:Ontology rdf:about="old.rdf"/>
     </owl:priorVersion>
   </owl:Ontology>
<!-- either -->
   <owl:DatatypeProperty rdf:about="&dc;creator"/>
<!-- or -->
   <owl:AnnotationProperty  rdf:about="&dc;creator"/>


Current:
    <owl:Class rdf:ID="union">
      <owl:sameClassAs>
        <rdf:Description>
           <owl:unionOf rdf:parseType="Collection">
              <owl:Class rdf:about="#a"/>
              <owl:Class rdf:about="#b"/>
           </owl:unionOf>
         </rdf:Description>
      </owl:sameClassAs>
    </owl:Class>

New (also legal Current):
    <owl:Class rdf:ID="union">
      <owl:sameClassAs>
        <owl:Class>
           <owl:unionOf rdf:parseType="Collection">
              <owl:Class rdf:about="#a"/>
              <owl:Class rdf:about="#b"/>
           </owl:unionOf>
         </owl:Class>
      </owl:sameClassAs>
    </owl:Class>



Suggested change to test cases style guide:
  - for OWL Lite and OWL DL test cases
       rdf:resource
and
       rdf:Description
    are avoided.


Jeremy

For reference:
Two pointers to my on-going efforts on syntax:

http://lists.w3.org/Archives/Public/www-archive/2003Jan/att-0107/01-t
http://lists.w3.org/Archives/Public/www-webont-wg/2003Jan/att-0356/01-jjc

I hope to post a fully baked proposal in the next week.
Received on Wednesday, 5 February 2003 12:49:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:57 GMT