corrections and an issue for SWRL spec esp. in abstract syntax

• Previous message: Sandro Hawke: "Re: XML syntax for SWRL"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
• Mail actions: [ respond to this message ] [ mail a new topic ]

Hi Peter and Mike and all,
Just noticed that there are some "*'"'s missing from the abstract syntax in 
the SWRL V0.6 document,
in order for it to correspond to our intention and to the XML serialization 
syntax.
In particular, the omissions are in the definitions of builtin, antecedent, 
consequent, and rule.
The abstract syntax for builtin should say

builtIn '(' builtinID { d-object* } ')'

Likewise, it should be

antecedent ::= 'Antecedent(' { atom* } ')'
consequent ::= 'Consequent(' { atom* } ')'

rule ::= 'Implies(' [ URIreference ] { annotation* } antecedent consequent ')'


QUESTION:  Any objections to making the above changes to the document?

Said has volunteered to do the cvs update.


More subtly, there's another issue in both the abstract syntax and
the XML serialization syntax:
do we want to permit a rule to have no atoms in its consequent AND no
atoms in its antecedent?  I think this was discussed a while ago
and we decided that it was OK.  But if not, we should say so somewhere
in the abstract syntax section.


Benjamin



________________________________________________________________________________________________Prof. 
Benjamin Grosof
Web Technologies for E-Commerce, Business Policies, E-Contracting, Rules, 
XML, Agents, Semantic Web Services
MIT Sloan School of Management, Information Technology group
http://ebusiness.mit.edu/bgrosof or http://www.mit.edu/~bgrosof

• Next message: Pat Hayes: "draft of SCL XML syntax"
• Previous message: Sandro Hawke: "Re: XML syntax for SWRL"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
• Mail actions: [ respond to this message ] [ mail a new topic ]

This archive was generated by hypermail 2.1.4 : 06/26/04 EST