[Reaction-tg] advantages of Situated LP for Reaction RuleML

[Benjamin Grosof]

Hi Gerd and all,

At 10:33 PM 2/22/2004 +0100, Wagner, G.R. wrote:
> >  more general and clean than current production or ECA systems
>
>But is there any (standard) semantic account of production and ECA r
>ules to which you compare?

No, not that I'm aware of.  Providing it is part of becoming more clean.
And expressively more generality is provided as well, e.g., wrt binding
restrictions, prioritization (e.g., arbitrarily partially ordered), and NAF 
(e.g., cases covered by well founded semantics).


> >  translation approach to production rules (Jess) from {core RuleML
> >     and Prolog and SQL RDBMS}, bidirectionally;
>
>But this is restricted to the assert action fragment of Jess, right?

Yes and no -- depends what one calls "core" here -- sorry, I should have 
been clearer.

Yes it is restricted that way for Horn RuleML, SQL RDBMS, since those don't 
permit effectors.
Yes also it is restricted that way currently for Prolog too, since the 
translation available in SweetRules and theoretically as well does not 
(yet) support side-effectful procedural attachments in Prolog.

But no it is not restricted that way if SCLP is viewed as part of the core 
of RuleML, since for that case effectors are also translated 
bidirectionally.  The SweetJess translation already supports translation 
with effectors between IBM CommonRules and Jess.

Benjamin


>-Gerd

________________________________________________________________________________________________
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