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

RE: [RIF-APS] Rules Sign

From: Hassan Ait-Kaci <hak@ilog.com>
Date: Thu, 4 Sep 2008 08:24:15 -0700
Message-ID: <9FC9C6B2EA71ED4B826F55AC7C8B9AAB01F3364D@mvmbx01.ilog.biz>
To: "Paul Vincent" <pvincent@tibco.com>, <public-rif-wg@w3.org>

I don't know whether to feel alarmed or amused at the sort of
symbol quibblings that have been exchanged regarding RIF PS syntax

I thought it was understood by all here that with a Presentation
Syntax for rules, this WG was *not* designing a new universal
rule language. We are simply defining shorthands that are
somewhat easier to read and write by humans than their normative
XML counterparts. The RIF PS is meant to help a human agent read
and write RIF examples and allow the automated generation of the
XML form from the PS form. Thus, such a PS should:

  (1) be simple and unambiguous to parse by software; and,
  (2) easy, by not excessively so, to manipulate by humans.

Why "not excessively so"? Because - again - we are *not*
designing a new universal rule language! We are just defining
(relatively) less ugly shorthands for (absolutely)ugly XML! :-)
Thus, PS is only rough syntax encompassing many potential rule
languages. What makes sense for one intended sementics does not
necessarily for another. AT this level, punctuation is important
only to ease reading and parser generation. Let us not make it
ridiculously peculiar. It should be clear and familiar to most.
It is silly trying to push matters of personal (dis)tastes. The
two criteria above are all we need.

At present, there are three levels of PS:

(1) "pure" PS (without Axel's extensions)
(2) PS ("pure" PS with Axel's extensions)
(3) APS (PS with Adrian's shorthands)

Adrian's abbreviations are more like macros that "desugar" APS
into PS.

In order to build a parser for (A)PS and generate its XML form,
[http://www.w3.org/2008/08/19-rif-minutes.html#action08], the
quick final settlement of all these syntactic issues is highly


Hassan At-Kaci  *  ILOG, Inc. - Product Division R&D

-----Original Message-----
From: public-rif-wg-request@w3.org on behalf of Paul Vincent
Sent: Thu 9/4/2008 2:51 PM
To: public-rif-wg@w3.org
Subject: RE: [RIF-APS] Rules Sign


'<= ' should be used in backward-chaining dialects, and '=>' in
forward-chaining ones. 


Paul Vincent

TIBCO | Business Optimization | Business Rules & CEP



From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org]
On Behalf Of Patrick Albert
Sent: 04 September 2008 12:54
To: Adrian Paschke; public-rif-wg@w3.org
Subject: RE: [RIF-APS] Rules Sign


Most Production Rules system have rules in the form  IF/WHEN
<conditions> THEN <actions>.


I'd recommend we adhere to the existing practice which is to have the
conditions first followed by the actions. 

Replacing the IF .. THEN by a '=>' as in '<Conditions> => <Actions>'
would be ok.





From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org]
On Behalf Of Adrian Paschke
Sent: lundi 1 septembre 2008 22:09
To: public-rif-wg@w3.org
Subject: [RIF-APS] Rules Sign




With respect to the abridged presentation syntax there is still an open
issue about the sign to distinguish the head and the body of a rule.


Currently, we use ":-" in the examples e.g. in UCR and PRD, which is
well-known in the logic community but not so much in others including
production rules.


I shortly discussed this issue with the BLD/FLD editors Michael and
Harold and we came up with this unambiguous proposal to distinguish
classical implication and rules head and body.


<== for PRD and BLD

<-- for classical


<== and <-- might be also reverted ==> -->





Received on Thursday, 4 September 2008 15:25:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:47:52 UTC