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

Re: Need a test -- inverse and oneOf

From: Jim Hendler <hendler@cs.umd.edu>
Date: Wed, 11 Jun 2003 08:20:14 -0400
Message-Id: <p05200f12bb0cce76c2c6@[]>
To: Sean Bechhofer <seanb@cs.man.ac.uk>
Cc: webont <www-webont-wg@w3.org>

That would serve my needs.  Jeremy.could we turn this into a proposed 
incocnsistency test?

At 1:14 PM +0100 6/11/03, Sean Bechhofer wrote:
>On Wed, 11 Jun 2003, Jim Hendler wrote:
>>  To be able to close some of our LC comments, there must be a test in
>>  our test suite that includes both inverse and oneOf.  In an earlier
>>  message I outlined a simple one, but perhaps someone in the WG can
>>  come up with a better one.  It is my opinion that without such a test
>>  we will have trouble convincing some people that OWL DL is
>>  implementable (cf the comments from Jena and HP).  Can we please get
>>  such a test at least proposed??
>>    thanks
>>    JH
>How about the following?
>  Class(a:NiceCorporation partial
>   restriction(a:employs allValuesFrom a:NiceGuy)
>   restriction(a:employs someValuesFrom oneOf(a:tom a:dick)))
>  Class(a:NiceGuy)
>  ObjectProperty(a:employedBy
>   inverseOf(a:employs))
>  ObjectProperty(a:employs)
>  Individual(a:dick
>   type(complementOf(a:NiceGuy)))
>  Individual(a:niceCorp
>   type(a:NiceCorporation))
>  Individual(a:tom
>   type(restriction(a:employedBy allValuesFrom 
>The interaction of the oneof and the assertion that dick isn't a Nice Guy
>allows us to conclude that niceCorp must employ tom. But then we know that
>anything that employs tom cannot be a NiceCorporation (due to the
>inverse), so we get an inconsistency.
>It's pretty trivial, but I think you do need both one-of and inverse to be
>able to state it.
>	Sean
>Sean Bechhofer

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)
Received on Wednesday, 11 June 2003 08:20:24 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 23:04:46 UTC