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

Re: [PRD] PRD TF telecon Tuesday 14 October

From: Christian de Sainte Marie <csma@ilog.fr>
Date: Tue, 14 Oct 2008 10:22:14 +0200
Message-ID: <48F456B6.8050604@ilog.fr>
To: Gary Hallmark <gary.hallmark@oracle.com>
CC: RIF WG <public-rif-wg@w3.org>

Gary Hallmark wrote:
> 
> I find it rather annoying that Adrian and I have made major changes (I 
> think in a good direction) but there are still many rough patches and 
> lots to do to make it ready to publish, yet we aren't discussing these 
> changes and assigning actions to fix issues that are in the spec now.

I completely agree that your proposals go in the right direction.

So much so, that I propose a complete reorganisation of the document to make the best of them [1] (do not worry, if you do not like it, we can revert :-)

Initially, I just tried to iron out the rough patches, so we can discuss the changes you and Adrian introduced more productively.

But, then, for Adrian's change in the semantics to make sense, some abstract syntax had to be introduced first, and, as you commented, that part should be at the begining, even before the XML syntax; not after.

Then I found that your proposed change to the semantics of actions required the same kind of change, and that the abstract syntax for actions was easier to specify after the model-theoretic semantics of condition than prior to it.

So, I moved to an organisation where we have:
1. Condition: abstract syntax, concrete XML syntax, model-theoretic semantics
2. Actions: abstract syntax, concrete XML syntax, semantics of actions
3. Rules, rulesets etc: abstract syntax, concrete XML syntax, operational semantics
4. Presentation syntax

Separating the semantics in three different parts make sense, it seems to me, because they are three different kinds of semantics: pure model-theoretic for condition; pure operational for rules and rulesets, and kind of hybrid, making the bridge between them, for actions (where we use a model-theoretic kind of semantics to specify the transition relation that is the basis for the operational semantics).

Notice that I made a couple simplifications to Adrian proposal, but without losing anything essential (Adrian and all: please check that it still makes sense).

Regarding your proposed semantics, I translated it in terms of entailment: the meaning should be the same, but it seems rather easier to understand (well, to me, at least).

I did no changes in the XML syntax parts, and I did not update the operational semantics of rules and rulesets yet. I also did not touch the introduction nor the presentation syntax, either.

If you like the new organisation, the next step could be to move the XML syntax out of the way, and gather it all in a final section (yes, I know, this is what you have been saying all the way, Gary :-)

[1] http://www.w3.org/2005/rules/wiki/PRD

Cheers,

Christian
Received on Tuesday, 14 October 2008 08:24:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:33:55 GMT