- From: Peter F. Patel-Schneider <pfps@inf.unibz.it>
- Date: Mon, 17 Jul 2006 12:35:43 -0400 (EDT)
- To: csma@ilog.fr
- Cc: public-rif-wg@w3.org
<scribe> ACTION: pfps will describe loophole he sees in the semantics [recorded in http://www.w3.org/2006/07/11-rif-minutes.html#action11] I see two loopholes in the mapping from rule conditions to the Conditions in http://www.w3.org/2005/rules/wg/wiki/Extensible_Design. 1/ There are multiple potential semantics in the Extensible Design. Worse, the different semantics are provided only by very vague references. This leaves open the possibility of using the syntax of the Extensible Design, but providing a special semantics that might, for example, interpret "And" as disjunction. Yes, this example of modifying the semantics is particulary silly, but there are indeed useful rule languages where the order of conjuncts matters. How can such deviant semantics be accommodated while disallowing silly deviations? 2/ The calls for mappings have not provided any semantic requirements for the mappings. This leaves open the possiblity that the mapping does not respect any semantics. For example, the mapping might just map disjunctions into CONJ. Yes, there has been discussion on what the semantic requirements should be, but there has not yet been consensus. See http://lists.w3.org/Archives/Public/public-rif-wg/2006Jun/0101.html for one possible way to go. Peter F. Patel-Schneider
Received on Monday, 17 July 2006 16:36:51 UTC