W3C home > Mailing lists > Public > public-rif-wg@w3.org > September 2008

Re: [RIF-APS] Issue about implicit quantification

From: Michael Kifer <kifer@cs.sunysb.edu>
Date: Mon, 1 Sep 2008 16:27:36 -0400
To: <public-rif-wg@w3.org>
Message-ID: <20080901162736.5e0c3688@kiferserv>

I second Adrian's proposal.

The early versions of what was then called "Core" did not require that
everything is explicitly quantified. As far as I can see, this is a gratuitous
requirement, which has a very bad consequence: if one omits the quantifiers
then it is not BLD, and there is no way to even define this omission as a
shorthand, because it would stand in direct contradiction to what BLD
explicitly requires.

As a result, the test cases for BLD would be written in a language that is not
BLD, which is an absurdity, IMO.

michael

On Mon, 1 Sep 2008 22:14:59 +0200
"Adrian Paschke" <adrian.paschke@biotec.tu-dresden.de> wrote:

> Hi,
> 
>  
> 
> Currently BLD says:
> 
>  
> 
> "All free (non-quantified) variables in the rule must be quantified with
> Forall outside of the rule (i.e., Forall ?vars (head :- body))."
> 
>  
> 
> http://www.w3.org/2005/rules/wiki/BLD#The_Presentation_Syntax_of_RIF-BLD_as_
> a_Specialization_of_RIF-FLD, Point 5
> 
>  
> 
>  
> 
> To have a very compact format the Abridge Presentation Syntax currently has
> the following convention:
> 
>  
> 
>    *  If not explicitly stated variables are universially quantified
> (Forall)
> 
>  
> 
> To enable this convention in the APS I would propose to drop that
> quantification requirement and say there is implicit quantification.
> 
>  
> 
> -Adrian
> 
>  
> 
>  
> 
Received on Monday, 1 September 2008 20:28:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:54 GMT