- From: Boley, Harold <Harold.Boley@nrc-cnrc.gc.ca>
- Date: Tue, 20 Feb 2007 09:56:27 -0500
- To: "Paula-Lavinia Patranjan" <paula.patranjan@ifi.lmu.de>, "Christian de Sainte Marie" <csma@ilog.fr>
- Cc: "RIF WG" <public-rif-wg@w3.org>
On the basis of these suggestions, I've drafted a preliminary http://www.w3.org/2005/rules/wg/wiki/Core/Overview -- Harold -----Original Message----- From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org] On Behalf Of Paula-Lavinia Patranjan Sent: Tuesday, February 13, 2007 11:34 AM To: Christian de Sainte Marie Cc: RIF WG Subject: Re: [TED] The sutructure of the Core document Hi, > In addition to all the comments by Allen, Philippe, Jos and Mike, > there is one issue that I would like us to discuss at the telecon, > today, and that is the overall structure of the Core document. > > To the current structure, I would rather expect something like this: > > - Section 1: An introduction explaining how Core relates to RIF and > what is the scope of that specification (e.g. Core is a/the root > dialect of RIF; it is designed to cover about Horn; etc). > > In WD1, this should be complemented with some explanation about our > overall approach (translators, modularity, dialects and extensions, > specification at the abstract syntax level and derivation of XML > syntax from the abstract syntax, application-specific > vocabularies/identifiers etc). Possibly, that part will be moved to > the Arch document if we decide to have one, but, in the absence of > such a document along WD1, some basic explanation is needed here. > > - Section 2: Abstract syntax and semantics > * 2.1 Core rules: the abstract syntax, BNF and semantics for Core rules > * 2.2 Core Condition Language: the abstract syntax, BNF and > semantics for the Core Condition Language > > (I would prefer to have rules introduced before the condition > language; the problem is that we would have a forward reference when > specifying the semantics of rules; or maybe we could separate the > semantics in a third subsection?) > IMHO, it is better to start with Core Condition Language and then extend it to Core Rules: 1. If Section 1 presents a short overview of Core Rules, it is easier to read and understand the document without forward references. 2. The condition part of a rule is also part of production rules (of the form 'when condition do action') and Event-Conditon-Action rules (of the form 'on event if condition do action'), so it is an important rule component for developing extensions to RIF Core. > - Section 3: XML > > Only illustrative examples in WD1 I suppose that we can find a better section's title :) Paula
Received on Tuesday, 20 February 2007 14:56:36 UTC