Hello everyone and best wishes for the new year

could someone please give me an explanation why the following two rules behave in a different way, although they express the same logical formula (i.e., not (Object ^ Pred) )? In fact, only the second rule reaches correct conclusions, as I note below.


(defrule Rule1
(not
   (and
      ?o <- (Object (name B))
      (Pred (obj ?o))
))
=>
(printout t "OK1" crlf)
)


(defrule Rule2
(or
   (not (Object (name B)))
   (and
      ?o <- (Object (name B))
      (not (Pred (obj ?o)))
))
=>
(printout t "OK2" crlf)
)

The definition of facts could be like:
(deftemplate Object (slot name))
(assert (Object (name A)))
(assert (Object (name B)))
(deftemplate Pred (slot obj))

where <obj> is the id of some Object fact, eg.

(defrule Rule
(declare (salience 10))
?o <- (Object (name A))
=>
(assert (Pred (obj ?o)))
)


What I notice is that Rule1 behaves erroneously when the Object fact exists in the KB, but not the Pred fact: although it is supposed to be triggered, it is not! Rule2, on the other hand, behaves correctly all the times.

When both Object and Pred facts exist or when the Object does not exist, both rules behave fine.

So, what is wrong with the syntax of the first rule? No variable definition is used outside the not pattern..

Thanks

Theodore


--------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users y...@address.com'
in the BODY of a message to majord...@sandia.gov, NOT to the list
(use your own address!) List problems? Notify owner-jess-us...@sandia.gov.
--------------------------------------------------------------------

Reply via email to