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

Re: Tests

From: Ian Horrocks <horrocks@cs.man.ac.uk>
Date: Wed, 30 Apr 2003 00:25:33 +0100
Message-ID: <16047.2541.119792.981745@merlin.horrocks.net>
To: Jeremy Carroll <jjc@hpl.hp.com>
Cc: www-webont-wg@w3.org

On April 29, Jeremy Carroll writes:
> 
> Thanks Ian
> 
> one point is that you seem to be looking at some old data ....
> 
> the editors draft has all the syntactic fixes in, including the two you 
> correctly raise 
> 
> http://www.w3.org/2002/03owlt/editors-draft/draft/

Sorry - I was working from the working draft rather than the editor's draft.

> 
> The associated Manifest with that is up to date.
> I will regenerate the zip files, (approved.zip and proposed.zip in the 
> directory
>  http://www.w3.org/2002/03owlt/editors-draft/draft/
> 
> they are ever so slightly old (22nd Apr)
> 
> The files on the web are always up to date
> i.e. the raw data is the files retrievable from the URLs.
> 
> Concerning that a lot of the tests are OWL Full - I am aware that coverage is 
> needed - the goal is for each of the features to minimally have two tests for 
> Lite (if applicable), two for DL and two for Full. This is really only 
> achieved for a handful of features. Many of the full tests are fairly early 
> before it was even vaguely clear what being in Lite or DL meant.
> 
> Thanks a lot for the input on the cardinality tests, with that and with Jos's 
> message:
> 
> http://lists.w3.org/Archives/Public/www-webont-wg/2003Apr/0073.html
> 
> and the other discussion of cardinality-005 on rdf-logic I suggest we:
> 
> 1: unapprove cardinality test 005, and propose it as a non-entailment
> 
> 2: unapprove 
> http://www.w3.org/2002/03owlt/I5.24/Manifest004
> and at some point I will recast as a Lite/DL test rather than Full
> 
> Ian please can you verify
> http://www.w3.org/2002/03owlt/equivalentClass/Manifest004

Done. I can report success with this test and also with 
http://www.w3.org/2002/03owlt/equivalentClass/Manifest005

> 
> since Jos has withdrawn his endorsement.
> It would be good to have a list of more tests to approve before last call - 

I am working on it. As I mentioned in another email, the NI team also
have lots of results from Cerebra.

> Jos reports success on:
> http://www.w3.org/2002/03owlt/I5.8/Manifest006
> http://www.w3.org/2002/03owlt/I5.8/Manifest008
> http://www.w3.org/2002/03owlt/I5.8/Manifest009
> http://www.w3.org/2002/03owlt/I5.8/Manifest011
> 
> Ian reports success on
> 
> <description-logic/Manifest001#test>
> <description-logic/Manifest002#test>
> <description-logic/Manifest003#test>
> <description-logic/Manifest004#test>
> <description-logic/Manifest005#test>
> 
> <description-logic/Manifest105#test>
> <description-logic/Manifest106#test>
> 
> (The 900 ones are true but not what was intended, if I have understood 
> correctly, so need fixing).

902 and 904 actually fail because the entailment should not hold.

Ian



> 
> 
> Jeremy
> 
> 
> 
> 
> 
>  
> 
> 
Received on Tuesday, 29 April 2003 19:25:31 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:58 GMT