in reply to Rules-based Perl?
The key of a rules-based system as I learned it (back in high-school, so please be aware that this is rusty!) is that there is no definitive progression of events through a system, unlike typical procedural or OO programming (and, what seems to me to be the same as functionaly programming). The rules as defined above trigger 'instantly' when the Given sides are matched by facts in the 'cloud'; rules can be given a numerical priority, but for rules of the same priority, the one that fires should be random if both Givens are successfully matched. While in most instances of rules-based programming, the initial state of that cloud is fixed, and thus one can follow how rules fire and facts appear and disappear, a most robust rules-based system would allow facts to appear at any time in the cloud (say from a client-server like interaction), and thus the progression would be difficult to monitor, much more like debugging some GUI elements.
As I'm reading this functional programming features, it feels more like they handle data in an abstract a way as possible to avoid evaluation until the last possible moment, thus allowing 'infinite' lists to exist within a few bytes of memory (As an example, the Functional.pm module has an example of calculating the first 10 primes starting from the infinite set of integers).
Mind you, with CLIPS, the operations part of each rule used a functional-like language to print out, store, or do other things with data. (part of CLIPS' name came from the resemblence to LISP). Certainly, if I were to work on developing this for perl, I'd include those functional elements in the abilities of the package. Also, it would not be hard to maintain a history of facts as princepawn suggests, given that as facts are 'deleted' they go to a 'fact heaven' with ties to what rule deleted them as well as what facts it birthed.
Dr. Michael K. Neylon - mneylon-pm@masemware.com || "You've left the lens cap of your mind on again, Pinky" - The Brain