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

Re: Dave's modified tests

From: Ian Horrocks <horrocks@cs.man.ac.uk>
Date: Tue, 16 Sep 2003 12:15:49 +0100
Message-ID: <16230.61669.672549.560048@merlin.horrocks.net>
To: Jim Hendler <hendler@cs.umd.edu>
Cc: Jeremy Carroll <jjc@hpl.hp.com>, www-webont-wg@w3.org

On September 15, Jim Hendler writes:
> At 5:31 PM +0100 9/15/03, Ian Horrocks wrote:
> >I reiterate my strong opposition to the idea of modifying tests to
> >make them easier to pass - as I mentioned in other emails, this is
> >doing a disservice to implementors. I'm not convinced about the value
> >of adding easier versions of a given test, but could be persuadable on
> >that point.
> >
> >Ian
> Let me be clear - my feeling is that some of our tests (esp. the ones 
> being discussed in this thread) test too many things at the same 
> time.  My preference for two versions was that this way an 
> implementor could clearly see what was expected in the comprehension 
> axioms that might not be otherwise needed to pass the test.

I agree that small tests focusing on particular aspects of the
language are useful. In many cases, however, it is the interaction of
several features which can cause potential problems and which tests
need to highlight (OWL DL itself is a good example - the difficulty of
reasoning in this language is caused by the interaction of inverse,
cardinality and nominals). We should be careful not 

> As far as rewrites go, I think we're in agreement - however the email 
> from Jeremy/Dave said:
> >>  >Hence Dave automatically transforms Graph2 into a query which he can then
> >>  >execute againsts Graph1, and pass the test.
> which seems to me to be okay for a qualified pass -- if it said "Dave 
> works hard to recode the graph into a query using his human powers of 
> reasoning" then I would agree it wouldn't count -- but 
> "automatically" seems to me to be acceptable as long as note is made.

Agreed - no problem regarding automatic (or even systematic manual)
rewriting - this could easily be part of an implementation. There was,
however, some talk (from Sandro?) of proofs relying on human
intervention/guidance - this obviously can't count as a pass.


>   -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	  *** 240-277-3388 (Cell)
> http://www.cs.umd.edu/users/hendler      *** NOTE CHANGED CELL NUMBER ***
Received on Tuesday, 16 September 2003 07:17:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:56:55 UTC