W3C home > Mailing lists > Public > www-webont-wg@w3.org > October 2002

RE: LANG: Proposal to close issue 5.17 - XML syntax

From: Smith, Michael K <michael.smith@eds.com>
Date: Wed, 30 Oct 2002 07:25:07 -0600
Message-ID: <B8E84F4D9F65D411803500508BE322141187C2D5@USPLM207>
To: Jim Hendler <hendler@cs.umd.edu>, "Smith, Michael K" <michael.smith@eds.com>, Dan Connolly <connolly@w3.org>
Cc: webont <www-webont-wg@w3.org>

Good example.  I think it gets to several issues.  One is the
media type question. I believe that a little redundancy can 
be a good thing.  A declaration of intention makes sense. 
I don't fully understand all the media type issues, but it
seemed to me that it might be necessary to mark something to
be interpreted as OWL.

Given your example there is no reason for me to know how to 
interpret it, at least as regards RDF vs OWL/Fast 
vs OWL/RDF.  There is no requirement that I track down the 
document tied to the jim: namespace.  

Regarding syntax.  I was not clear enough in my message.   The
real problem is that RDF/XML admits sentences that OWL does not.  
E.g.

<owl:Class id="A">
 <owl:intersectionOf>
   <rdf:description rdf:resource="B"/>
 </owl:intersectioOf>
</owl:Class>

The syntax of a language should limit me to sentences that
are defined in the language.  

- Mike

-----Original Message-----
From: Jim Hendler [mailto:hendler@cs.umd.edu]
Sent: Tuesday, October 29, 2002 9:53 PM
To: Smith, Michael K; Dan Connolly
Cc: webont
Subject: RE: LANG: Proposal to close issue 5.17 - XML syntax


At 7:25 PM -0600 10/29/02, Smith, Michael K wrote:
>Dan,
>
>I think this is the same discussion we had before. And you
>probably gave me pointers to the email message on the RDF
>discussion thread that contains the latest version of the
>RDF/XML syntax that would be sufficient to account
>for OWL.  Should I provide a reference to that? 
>
>You are asserting that if I write some XML without a single
>OWL tag, but that satisfies the RDF/XML standard, then I have
>written OWL.  Perhaps, given the semantic support for OWL/RDF,
>I have.  I still find this odd. 
>
>Once I get past the examples, the best way I currently know
>of to determine what's legal OWL is to read the Reference,
>which gives a pretty good verbal description of the composition
>of OWL components. 
>
>- Mike


Mike-
  Suppose I have a file that says

<rdf:RDF>

<jim:elephant rdf:id="Clyde">
   <jim:owner :BarnumBailey>
   <jim:color color:grey>
</jim:elephant>

and it turns out that jim: is a file with an ontology statement in it 
that defines elephant to be a class with the various properties.

This is XML with no OWL in it (and also RDF with no Owl in it).
Using application/rdf+xml  seems to me we would handle it right, and 
if the namespace reference to jim: was followed, then it would be 
recognized to be OWL (or at least instances of OWL).   So the issue 
of "what is OWL" might be more complicated to answer than it seems...
  -JH

-- 
Professor James Hendler				  hendler@cs.umd.edu
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)
http://www.cs.umd.edu/users/hendler
Received on Wednesday, 30 October 2002 08:25:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:53 GMT