W3C home > Mailing lists > Public > public-rif-comments@w3.org > December 2009

Re: Some thoughts on "intended Semantic Structures" section 3.8 of RIF-FLD

From: Christian De Sainte Marie <csma@fr.ibm.com>
Date: Wed, 9 Dec 2009 15:49:21 -0800
To: MOTTD@uk.ibm.com
Cc: public-rif-comments@w3.org
Message-ID: <OF0B20BAE8.DD4BD14E-ON88257687.008226D0-88257687.0082C89D@fr.ibm.com>
Dear David,

Thank you very much for your comments. Please see our comments within. 
David Mott wrote on 19 October 2009:
> I am trying to understand the section 3.8 on intended semantic 
structures. 
> I am not a deep expert in logic, so what may be obvious and implied to 
> experts is alas not obvious to me. (My goal is to develop a dialect 
> handling Naf amongst other things).
> 
> I have two comments/questions...
> 
> 1) Looking at the section 3.8 , I see the term "intended semantic 
> multi-structure". It is said that RIF-BLD does not specify what these 
> might be, and I am fine with that. However this section does not even 
> define (explicitly) what the "purpose" of such structures might be, nor 
> any criteria for knowing that you have selected the right set for a 
given 
> purpose. (Could the set be entirely random, I ask myself?)

We have added more explanations in that section, which should hopefully 
make the intent clear. The intended structures are by no means random. 
They should be defined by a trained logician who is familiar with the 
mode-theoretic semantics of rules. 
> Therefore I 
> look to the Shoham87 reference, and I read about "preferred" 
> interpretations. This gives a clear "purpose" to the relationship 
between 
> alternative interpretations, and I guess that this is what is meant by 
> intended semantic multistructure.
> 
> So is it correct to say that intended semantic multistructure  == 
> preferred interpretation? If so, could that be made explicit? If not, 
then 
> what is the "purpose"?

Yes, "intended" subsumes "preferred" in Shoham's terminology. Please see 
if the reworked explanations make the purpose any clearer. 
> 2) In the section 3.9 a definition for entailment is given, based upon 
the 
> lattice of truth values. However in the reference (again Shoham87) there 

> is no mention (as far as my non-expert eyes can see) of lattices of 
truth 
> values, only preference relations between interpretations.  So I am not 
> sure what the intended RIF-FLD relationship is between these two 
concepts.

This is a generalization of Shoham's notion to the case of multi-valued 
logics. Shoham considers only two-valued logics. Our notion reduces to 
Shohams for 2-valued dialects. 

Please acknowledge receipt of this email to <
mailto:public-rif-comments@w3.org>  (replying to this email should 
suffice). In your acknowledgment please let us know whether or not you are 
satisfied with the working group's response to your comment.

Best regards,

RIF WG
------------------------------------------------
Christian de Sainte Marie

IBM
9 rue de Verdun
94253 - Gentilly cedex - FRANCE
Tel. +33 1 49 08 35 00
Fax +33 1 49 08 35 10


Sauf indication contraire ci-dessus:/ Unless stated otherwise above:
Compagnie IBM France
Siege Social : 17 avenue de l'Europe, 92275 Bois-Colombes Cedex
RCS Nanterre 552 118 465
Forme Sociale : S.A.S.
Capital Social : 611.451.766,20 ?
SIREN/SIRET : 552 118 465 03644
Received on Wednesday, 9 December 2009 23:50:12 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 December 2009 23:50:12 GMT