Re: All Encompassing / Loose Coupling

Hi Adrian!

You are surel;y making good points:

1. Interoprability of (exisityng or future) rule languages is an 
essential issue for the SW.

2. An approach to interoperability of rule langhuages based on messages 
is very appealing on the Web and SW.

I also like the vision of "Rule Spaces" you suggest in Section 2 of  
http://www.w3.org/2004/12/rules-ws/paper/19/, although I doubt it is 
realizable today as you describe3 it: "A logical model- or 
fixpoint-theory defining exactly what should be inferred from any given 
set of rules and facts." Having such aq theory in a both human and 
machiune processible form is, I believe, not achievable today - maybe in 
a feew decades.

I strongly believe that the SW calls for new rule languages and not for 
those already around in industry and academia. There are two reasons for 
this:

1. The rule languages developped so far do not sufficiently take into 
account specific traits of the Web and Semantic Web like distribution, 
decentralised management, and access to partly specified data (needed 
because the data do not habe a strictly enforced common data model).

2. Rules languages deveklopped spo far, especially in industry, are no 
longer state-of-the-art: A big deal has happend during the last 5 to 10 
years in research.

-- 
Francois

Received on Wednesday, 29 June 2005 07:24:53 UTC