On Friday 04 April 2003 00:21, [EMAIL PROTECTED] wrote: > You'd want to have a separate instance of the engine (the jess.Rete > class) for each rule set. Alternatively, you could have a separate > one for each agent, but with 400 agents in one process, that's > probably too heavyweight a solution.
OK. > The trick to keeping the various agents "trains of thought" separate > even though they're sharing an engine is for agent-specific facts to > have an agent-specific tag; i.e., you can't use facts like > > (agent-network-port 6543) > > Instead they would use something like > > (agent-network-port (id 123) (port 6543)) > > Where 123 is a unique identifier for a single agent; that way there > could be separate agent-network-port facts for each agent. This way, > there can be just -one- run, and all 200 agents sharing an engine can > "think" at the same time! Magnific! I really thank you very much for your help. It would have been very difficult to see these things myself just reading the manual. I would like to say too that I like very much how Jess is designed. You get a very clean, natural interface to an underlying complex system. That's sign of hard work. Thank you again, and best regards from sunny Barcelona, -- fxn -------------------------------------------------------------------- To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]' in the BODY of a message to [EMAIL PROTECTED], NOT to the list (use your own address!) List problems? Notify [EMAIL PROTECTED] --------------------------------------------------------------------
