Hey,

 

Can someone enlighten me as to what the 'compiled sequential' rule
evaluation method provided by Blaze Advisor is all about?  It is
advertised as a performance optimization over Rete.  I could pose the
question to the Fair Isaac folks but I'd like a somewhat more objective
opinion (if that is possible).

 

What is sacrificed to achieve the performance gain?  Is something
similar possible using Jess?  Does it represent an alternative to Rete,
or is it simply an optimization of Rete based on the assumption of a
FIFO conflict resolution?

 

Thanks.

-Mitch Christensen

 

Reply via email to