- From: Jos De_Roo <jos.deroo.jd@belgium.agfa.com>
- Date: Thu, 7 Feb 2002 09:33:44 +0100
- To: "Pat Hayes <phayes" <phayes@ai.uwf.edu>
- Cc: "Brian McBride <bwm" <bwm@hplb.hpl.hp.com>, "w3c-rdfcore-wg" <w3c-rdfcore-wg@w3.org>
[...]
>>>Or, better, why not just trash it, since apparently nobody uses it
anyway.
>>
>>We decided at last week's telecon to move forward with clarifying
>>what it means. We've had some excellent discussion this week, with
>>the issues becoming clearer - thanks Pat for your excellent
>>questions earlier.
>>
>>To my simple mind it boils down to a choice. Does a reified
>>statement represent a statement or a stating (an occurrence of a
>>statement in a graph).
>>
>>The formal model part of M&S is clear that its a statement.
>>However, the intended application was provenance, for which a
>>stating is required. The original WG were not aware, and did not
>>consider the difference. We have a simple choice:
>>
>> o change the formal definition to suit the intended
>> application of the original WG
>>
>> o stick to the formal model and let someone invent a
>> new vocabulary for stating.
>>
>>Please lets stay out of the rat holes, choose and move on.
>
>OK. But let me ask: suppose there were two groups, and one said it
>was a statement and the other said it was a stating. Are there any
>entailment tests (or some other kind of behavioral test??) where they
>would disagree about what the right answer was?
yes, Brian's etc (entailment test case) we discussed this week
we, the 'statement' guys, would say
( <http://www.agfa.com/w3c/n3/b1.nt> )
log:entails
<http://www.agfa.com/w3c/n3/b2.nt> .
whereas the 'stating' guys would say
( <http://www.agfa.com/w3c/n3/b1.nt> )
log:notEntails
<http://www.agfa.com/w3c/n3/b2.nt> .
--
Jos
Received on Thursday, 7 February 2002 03:34:26 UTC