- From: Michael Kifer <kifer@cs.sunysb.edu>
- Date: Wed, 24 Aug 2005 13:39:23 -0400
- To: Sandro Hawke <sandro@w3.org>
- Cc: jos.deroo@agfa.com, public-rule-workshop-discuss@w3.org
> > > At any moment when NAF is computed the set of rules and facts is known to > > the inference engine. What is your problem? > > Is NAF useful in rule sets which are combined with unknown, > unrestricted other rule sets (potentially also give true information > about the same domain of discourse) before any inference is done? Sandro, You probably didn't understand the essence of my previous message. The thing you are talking about doesn't exist -- hasn't been defined yet as far as I know. It is certainly not any form of the NAF that I am familiar with. This is why I compared "unscoped NAF" with the Unicorn. If you define it rigorously then I could look at it and give you my subjective opinion as to whether this new notion is useful or not. > If so, could you give me an example? If you define what you have in mind rigorously and if this definition will define a non-empty set of concepts then, perhaps, we could talk about examples. --michael
Received on Wednesday, 24 August 2005 17:39:32 UTC