RE: Guide: Proposed response to comments.


Not sure I understand.  rdf:list is in the table, just not
described in the Guide, since the Guide uses OWL constructors
like oneOf and intersection in the XML representation of 
things that require rdf:list to describe in triples.

- Mike

-----Original Message-----
From: Jim Hendler [] 
Sent: Monday, May 05, 2003 3:18 PM
To: Smith, Michael K; webont
Subject: Re: Guide: Proposed response to comments.

At 14:31 -0500 5/5/03, Smith, Michael K wrote:
>Proposed response to Lee Lacey's latest response.
>- Mike

Mike - generally I like this, I would ask whether:

># The cross-reference before references doesn't provide references to # 
>rdf:List, rdf:nil, rdf:type, and rdf:Property.  It would be helpful to 
># at least mention each of these at least briefly in the guide as an # 
>introduction before readers dive into the reference or semantics # 
>There is a link for rdf:type.  The Guide sticks pretty much to the DL 
>syntax, which does not require the list constructs and rdf:Property.

is as good an answer as just adding the properties he requests.  That 
would (i.) make it so we don't have to defend whether Guide is just 
DL or not, and (ii.) actually be technically correct as DL includes 
owl:oneOf, which builds an rdf:list (which includes rdf:nil) and also 
rdf:Property, which is, unless I am mistaken, still usable in owl, 
although redundant.
  I ask because there are several cases we have encountered in my 
group's work where, due to use of owl:imports, the following comes up 
we import an RDF document which has things like

(RDF document:)

ex1:foo a rdf:property;
   rdfs:range ex1:bar.


"" owl:imports "http://.../ex1.rdf".

ex1:foo a owl:objectProperty;

ex1:bar a owl:objectProperty.

This is a pattern by which I have assumed that RDF documents which 
would be in Full can be made into DL by having an OWL files that 
contains the typing information that RDF wouldn't include, and then 
can add new properties.

I thought this could be done in DL, but it is possible that I've 
missed a restriction.

Anyway, I digress a bit from my point - which is that a couple of the 
OWL DL constructs build triples that include some of those RDF 
constructors, so it wouldn't hurt to have them in the table...


Professor James 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	  240-731-3822 (Cell)

Received on Monday, 5 May 2003 17:32:57 UTC