W3C home > Mailing lists > Public > public-rif-wg@w3.org > February 2009

Re: Action 692: http://www.w3.org/2005/rules/wg/track/actions/692

From: Dave Reynolds <der@hplb.hpl.hp.com>
Date: Wed, 04 Feb 2009 18:30:36 +0000
Message-ID: <4989DECC.4020208@hplb.hpl.hp.com>
To: Changhai Ke <cke@ilog.fr>
CC: RIF WG <public-rif-wg@w3.org>

Changhai Ke wrote:
> The action consists in re-factoring the BLD schema, extract the part 
> that can become Core schema, and have BLD and PRD schema to include Core 
> schemas.
> 
>  
> 
> Before working at the schema level, I have detected several differences 
> or inconsistencies in the BNF (Core: 18/12/2008, BLD: 22/09/2008). They 
> need to be confirmed or fixed before we can work on the schemas. Those 
> who are experts on these BNFs, can you comment and provide suggestions?
> 
>  
> 
> Changhai
> 
>  
> 
> - BNF for the rule language is the same for Core and BLD
> 
> - Core’s ATOMIC does not have “subclass”, while BLD’s does. 

Correct - no subclass in Core.

> And proceed with a refactoring for Core’s FORMULA BNF.
> 
> - UNITERM: difference to confirm
> 
> Core: UNITERM ::= Const '(' (TERM* ')'
> 
> BLD: UNITERM ::= Const '(' (TERM* | (Name '->' TERM)*) ')'

Correct - no named argument uniterms in core.

> - Equal: difference to confirm
> 
> Core: Equal ::= TERM '=' ( TERM | 'External' '(' FUNC ')' )
> 
> BLD: Equal ::= TERM '=' TERM

Correct but subject to change once we finalize where External can be 
used. The issue was that we agreed to restrict Externs to not be nested 
in predicates and only used via an assignment type equality. Hence the 
separation of TERM into TERM and GENERAL_TERM and the change to equality.

However, Gary requested that this decision be reversed and we allow 
nesting of equality in both conditions and conclusions. I believe we 
agreed to do that post-WD but the change hasn't been made.

> - GENERAL_TERM defined in Core but not used. To confirm.

It is used in FUNC which in turn is used in Equal as above.

> - TERM: difference to be confirmed.
> 
> Core: TERM ::= IRIMETA? (Const | Var)
> 
> BLD: TERM ::= IRIMETA? (Const | Var | Expr | 'External' '(' Expr ')')

Correct, see above.

> - Implies: difference to confirm
> 
> Core: Implies ::= IRIMETA? ATOMIC ':-' FORMULA
> 
> BLD: Implies   ::= IRIMETA? (ATOMIC | 'And' '(' ATOMIC* ')') ':-' FORMULA

Correct, to ensure compatibility with PRD.

Dave
-- 
Hewlett-Packard Limited
Registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England
Received on Wednesday, 4 February 2009 18:31:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:34:03 GMT