Several questions on the email list this morning, most dealing with not enough RTFM, reading the fabulous manual. Second observation, our society (like myself and apparently the others on this list) wants immediate results and, unfortunately, immediate understanding of complex ideas. So, let's begin the beguine 1. <The advantage of Jess> Jess is declarative, not procedural. Literally, Jess "chews" on the IF-THEN-ELSE statements until no more rules can be executed, fired. Jess is non-monotonic in nature, meaning that the data can be added or deleted from the problem set without affecting the results. Jess can work with incomplete information and still reach a result, maybe not the one that it would reach with all of the possible known facts. Jess, if programmed properly by persons who are familiar with the tool and the art and science of AI, should be able to think almost like a human thinks. NEVER try to program "common sense logic" into Jess. 2. <When to use Jess> If the same problem CAN be solved in procedural code, then do it in Java, C++, etc. (Ouch?) However, if a decision tree for the process takes more than two pages of standard, letter-sized paper in fonts that can be readily seen at arm's length, then use Jess. If the problem is extremely complex, use Jess. If the rules need frequent modification, use Jess. etc., etc., etc. 3. <How to use Jess> RTFM, RTFM, RTFM. Do the examples. Do the examples. Do the examples. Then ask questions of the mailing list. 4. <How to convert from ILOG JRules to Jess> JSR 94 was supposed to be a start at this. Never happened. It is NOT to the manufacturer's financial advantage to do this so it will NOT be done until the user community forces them to do it. Give it a wee bit of thought of you will understand that making money, not community enlightenment, is the primary goal of ANY business, otherwise it is nothing more than a philantrophic organization or a government agency. However, in answer to the question, probably a parser of some kind might work, but that would be the same as parsing C to C++ or Java to C; not the best of ideas. Why? Because each engine has certain advantages and certain disadvantages and you can have a very non-efficient engine at the end of a parsing arrangement. (I used to use awk for this kind of thing, but there are some much better languages today.) Also, you might check the legality of parsing ILOG JRules to Jess. Also, almost any version of ILOG can generate the irl language that is the basis of all of their different views and XML configurations. SDG jco James C. Owen Senior Consultant "Never give up. Never give up. Never, never, never give up." From a speech by former Prime Minister, Winston Churchill, during the most trying times of the British Empire; a great leader too soon forgotten by his country after keeping it from being swallowed up by the Nazi Empire during the second world war. |
- JESS: My Two Cents... James C. Owen
- Re: JESS: My Two Cents... James C. Owen
