Do you know of JEOPS (sourceforge)? It was the only rules engine I could find that had a reasonable license.
I also think that the future of programming will more and more involve reliance on declarative/functional approaches, especially to glue together procedural code snippets.
thanks
david jencks
On Friday, January 31, 2003, at 10:03 AM, Bill Burke wrote:
Hope you don't mind, but I have the [jboss-dev] mail list in copy.
What I think you're saying in all your email is:
1. You're interested in Rules engines and JBoss (jsr94)
2. You want to fuse AOP and a Rules engine.
Both are great ideas. We're always looking for people with initiative and
new exciting ideas. How can you join this? Implement something. Implement
something that sort of works, and we'll give you CVS access. We're pretty
free with CVS access as long you have shown some prior initiative.
I think a Rules engine would be an excellent addition to JBoss and I am
looking forward to your contributions. Let us know.
Best Regards,
XXXXXXXXXXXXXXXX
Bill Burke
Chief Architect
JBoss Group, LLC
XXXXXXXXXXXXXXXX
-----Original Message----- From: Tomas Lapienis [mailto:[EMAIL PROTECTED]] Sent: Thursday, January 30, 2003 7:58 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Are you in functional programming and JSR94? How can I join this?This letter is basically provoked by Bill's call for developers :) Hello Marc and Bill, I want to join JBoss development in the area which I think is the key area in the modern programming technologies. Actually for me it was a missing part which I was looking for years and (stupid me:-) successfully overlooked - until a few month ago. In the first glance it looks as simple as jsr94 http://www.jcp.org/ja/jsr/detail?id=94 But for me it is not just "one more" API in a growing J2EE technology set. (BTW, BEA is involved ;-) I see funtional programming as a glue for Aspects. <CV> I have 16 years of experience in programming (10 years I was paid for that). I have started from procedural (Pascal); then it was x86 assembler under DOS next I did OO in C++; device drivers in C; let myself to express in >100 lines of single sql statement in oracle (this requires futher explanation); felt advanced in C++ STL Generic and Design Patterns; got some java; hacked own XSLT based on W3 draft under Java and some lib for servleting it (Cocoon style embryo); [this is the part when I saw EJBoss for the first time - year 2000]; next years I spent under strong influence of Apache and JBoss doing Web Apps - I could not be fast enough with growing "good" technologies set - especialy Apachy's ones. </CV> And I feel puzzled in Apache - there are a bunch of nice technologies which does more or less the same but is difficult to mix and match. So I bounded myself to a very basic (IMHO) technologies and started to look up for the glue. Aspects were looking very promising [glue] except that Aspects ran into the problem - there is the need for query language which will resolve the problem how to apply (mix&match) aspects on the system. I kept thinking until I noticed functional approach to all this procedural nightmare. I started to dig deeper and that's what I noticed: Functional Approach (FnA) a) FnA is a kind of technology which now has a little effort made to combine it with Procedural Approach (in open source area especialy). BTW, Messaging System is a kind of stripped FnA (IMHO) b) I think that extended treatment of FnA will work as a glue for Aspects. Also, FnA cuts the need for aspects!!! They become a Rhs of FnA. OK, I'm wrong here :) in FnA Aspects remains as a powerful Design Patterns Set for Problem Solving. Anyway, current FnA lacks of delegation which is required part when doing distributed applications - some business logic must be applied on remote data IN THE remote site without additional coding. Resume: This is introductory part - I need feedback about JBoss Community interests and advances in FnA field. Mail me directly or tell what mailing lists to subscribe (currently this account is not subscribed to any) Regards, Tomas Lapienis P.S. Many things left not said: one of them is my study in oracle performance - rule based optimization and cost based optimization (spent 1 year full time) It discusses throughput - goes with explanation why it has 100 lines in single sql statement. BTW, EJB fundamentals clashes with this experience - the partial work around it is MBeans (sad, but the usage of MBeans is more difficult than ordinary beans). Basically, SQL is FnA, just oracle optimizes it more than Rete can. Useful links: http://www.pst.com/rete.htm - with all my respect to Mr. Forgy http://sourceforge.net/projects/clipscpp - maybe this must be ported to java - here is the begining? http://www.haley.com/ http://www.mail-archive.com/jess-users@sandia.gov/msg03506.html http://www.mail-archive.com/jess-users@sandia.gov/msg03284.html I guess JRULES-OPT uses the similar approach as Oracle's------------------------------------------------------- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com _______________________________________________ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
------------------------------------------------------- This SF.NET email is sponsored by: SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! http://www.vasoftware.com _______________________________________________ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
