Please note that I never intended to say that RIF core is RuleML (still can't see where that could be mixed up). My apologies if I did.
For the translation between RIF core and both production systems: I thought I'd start there, so far so good. Admittedly the rules I experimented with could be more complex, hence the brackets around supported. I'll see what the next steps will be when I get enough data (including feedback). All the best, Pierre On Sun, 29 Aug 2010 15:30:35 +0200, "Wolfgang Laun" <[email protected]> wrote: > It's a mess. RIF Core isn't RuleML; neither supports the full range of > either Drools or (I suspect) Jess. > > Jess and Drools are Production Systems; so you also have to come to grips > with RHS coding as well. Which means, in any case, that any DSL restricts > RHS coding to the subset defined in the DSL. (I concur, that this might be > a > good thing.) Also, note that W3C also have completed RIF-PRD, which is > intended to bridge between various Productions Systems. We'll see. > > Drools provides a DSL implementation system that uses regex parsing (for > DSL > rules) to create a context-free language expression, which is also bound to > have limitations. The same would be true for such a parser to translate > into > Jess. > > Having said all that, it might be worth the manageable effort to implement > a > generic DLS system for implementing a DSL with Jess/CLP as its target > system. If you strive for anything close to natural language, you will have > to make amends. But I also concur with the notion that promising a DSL is a > nice marketing spiel for RBS vendors, but that the famous "domain experts" > rarely cotton to writing the rules all by themselves. Part of which may, of > course, be due to the fact that error diagnostics in such a generic system > are bound to be somewhat less than "user friendly". > > -W > > > On 29 August 2010 10:25, <[email protected]> wrote: > >> Hi, >> >> I'm currently working on a common interface for multi-rule-engine >> distributed systems. Part of my work has so far been to develop XSLTs to >> allow RIF core (W3C spec) translations to and from JessML. DroolsML is >> also >> "supported" so it's possible to exchange rules between Jess and Drools >> through RIF core. >> >> In addition to that, I've got a PhD student currently working on a user >> friendly UI for RIF core in Flex (Adobe stuff). >> >> I'm hoping we get to present a prototype of this at the RuleML challenge >> this December. I'm also hoping I'll be able to share the XSLTs with the >> community at some point after that. >> >> It's August and everybody is on holidays here (including me) hence why I >> can't give firm release dates :) >> >> All the best, >> Pierre >> >> >> On Sun, 29 Aug 2010 04:04:17 +0200, "Ernest Friedman-Hill" >> <[email protected]> wrote: >> > Be careful to distinguish RuleML, which is a proposed standard rule >> > language, from JessML, which is an XML-based rule format natively >> > supported by Jess. The point of JessML, really, is that it's easier to >> > manipulate XML than the Lisp-like Jess rule language in code, so if >> > you want to write some kind of translator, GUI, or other tool for >> > manipulating Jess code, using JessML will make it much easier to do. >> > >> > So in any case, JessML exists so that's it's very simple for you to >> > create your own special-purpose front ends like what you describe. >> > >> > >> > On Aug 24, 2010, at 12:42 PM, Donald Winston wrote: >> > >> >> What's the possibility of some sort of rule "translator" for Jess. >> >> One of the reasons for using a rule engine in a typical business app >> >> is to separate the business rules from the rest of the application >> >> so non programmers can work with the rules. (Although in my >> >> experience we end up with programmers developing the rules anyway) I >> >> think it's going to be difficult to convince my clients to let me >> >> use Jess if this is a priority for them. I believe most will be put >> >> off by the Lisp language that Jess uses. Even if software engineers >> >> will be writing the rules from some sort of documentation they might >> >> not like it. >> >> >> >> I know there's a RuleML effort. But to be really good the >> >> translation would have to go both from ML -> Lisp and Lisp -> ML. In >> >> addition the rules developer would not be editing the ML directly >> >> but an english like language to express the rules. This would >> >> probably put a straight jacket on how Jess is used in these kinds of >> >> apps(which may be a good thing). >> >> >> >> (I've decided I like Jess especially because these days Blaze >> >> Advisor costs six figures! Drools is a frigging elephant) >> >> >> >> I'm going to use it for a prototype app I've had in mind. >> >> >> >> >> >> >> >> >> >> >> >> >> >> -------------------------------------------------------------------- >> >> 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] >> >> . >> >> -------------------------------------------------------------------- >> > >> > --------------------------------------------------------- >> > Ernest Friedman-Hill >> > Informatics & Decision Sciences Phone: (925) 294-2154 >> > Sandia National Labs >> > PO Box 969, MS 9012 [email protected] >> > Livermore, CA 94550 >> > http://www.jessrules.com >> > >> > >> > >> > >> > >> > -------------------------------------------------------------------- >> > 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]. >> > -------------------------------------------------------------------- >> >> >> >> >> -------------------------------------------------------------------- >> 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]. >> -------------------------------------------------------------------- >> >> -------------------------------------------------------------------- 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]. --------------------------------------------------------------------
