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

Re: Syntactic checking of weak v. fast graphs

From: Jim Hendler <hendler@cs.umd.edu>
Date: Thu, 26 Sep 2002 23:12:36 -0400
Message-Id: <p0511170fb9b97cb6a49a@[129.2.177.222]>
To: phayes@ai.uwf.edu
Cc: webont <www-webont-wg@w3.org>

At 1:35 PM -0400 9/26/02, Jim Hendler wrote:
>Pat-
>Let me start with an example:
>
>  In D+O we have a separation of datatypes and object types.
>If a user says
>  x a daml:DataTypeProperty;
>then it is "undefined" if the person also says
>  x a daml:ObjectTypeProperty;
>this "undefined" works as I understand it by saying in the reference 
>document that you must pick one, and in the MT by assuming they are 
>separate.
>
>Suppose we now decide that we are going to allow things to be of both types.
>In this case I see two ways we can do it -- one is we simply drop 
>the comment in the reference manual.  Another is that we say in the 
>reference manual that
>"You can make something be both of those by saying
>  x a daml:bothDataAndObject.
>
>In both cases we get the same effect semantically, but there is a 
>big difference - in one case we can know that if we do not see 
>"daml:bothDataAndObject" then we have separation.
>
>So for the issue of OWL vs. OWLinRDF (Fast Owl and Weak OWL)
>  If we say OWLinRDF is defined as XXX, but if you don't use features 
>YYY,ZZZ, and AAA then you are sure to be in FAST OWL.   Suddenly, we 
>have a nice property - both Weak and Fast are WELL FORMED, but you 
>can tell!
>
>  So, if Fast OWL is actually a subset of OWLinRDF, and there was 
>some way we could tell, we could be closer to having our cake and 
>eating it too.
>
>  that is my question
>  JH
>

Pat - got more time to look things over, I understand we cannot do 
this just by using syntactic features like the above.  It does seem 
like we could say "if you use the following discipline (i.e. avoid 
this, don't do that, etc.) or gaurantee compliance with the abstract 
syntax" then you get the Fast Owl advantages...
  -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 Thursday, 26 September 2002 23:12:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:52 GMT