There is a blackboard shell link on the Franz site. If you don't mind switching lisps, you have 2 choices: You could use KnowledgeWorks, which is a version of Lispworks (related to Lucid). You can use GoldenHill Common Lisp.
You can do backward chaining w/a forward chainer if you must. So then any lisp forward chainer would do: (Lisa, OPS5, or ART etc, if available). ART: InferenceCorp (then Brightware, now Mindbox.com) had a language that spawned CLIPS, (which is what Jess tries to be compatible with). It was called ART. They rewrote it around the time NASA wrote CLIPS, because Lisp machines where scarce. I think these companies still have it, and often sell the new version (Art*Enterprise) with consulting. The simple forward-chainer OPS5 could work, but you would want to work the OPS layer so it is closer/more-integrated with the objects. I think an ai/es/lisp? faq has a link to a person that has a rete-matching-network right on CLOS instances (what I think Lisa is trying to do), but it is not free like Lisa. LOOM is the lisp version of PowerLoom, just like ART/Lisa is of Art-Enterprise (~CLIPS). You could try SOAR (which also has a C version now), but it might be even more difficult to use? I'd be interested in finding out what you found was the easiest/best path for you. Oh yeah, for builing up a class-hierarchy/(Knowledge-Base), you might check out: http://protege.stanford.edu/download/prerelease/index.html -Mike B. (ps. If you need contract work/help, you can contact me:) If you have trouble finding any of the links I mentioned, I can look them up for you later. On Friday, June 21, 2002, at 03:36 PM, Ronald Taylor wrote: > Hi folks, > > Our lab will be developing a large object-oriented (CLOS-based) > knowledgebase for alcoholism research in Lisp (Franz Allegro Common > Lisp). We are looking for a package that will add forward & backward > chaining inference to the classes, class hierarchies, and object > instances that will be in the knowledgebase. We've taken a look at > the SourceForge Lisa project (Intelligent Software Agents in Common > Lisp), but apparently Lisa does not do backward chaining, which will > be important for us. Does anybody have any recommendations? Is anybody > working on a full Lisp version of Jess? > > (I am aware that Jess is a very fine package, which is why I subscribe > to this list! But we were hoping for something that is Lisp-based, for > integration.) > > Ron Taylor > > --------- > //=\ Ronald Taylor Phone: (303) 315-1091 > \=// Director of Gene Expression Analysis Fax: (303) 315-1089 > //=\ Center for Computational Pharmacology > \=// University of Colorado School of Medicine, room 2633 > //=\ Dept of Pharmacology, Campus Box C236 > \=// 4200 E Ninth Ave > //=\ Denver CO 80262 Email: [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 owner-jess- > [EMAIL PROTECTED] > -------------------------------------------------------------------- > > -Michael [EMAIL PROTECTED] http://homepage.mac.com/bobak/ -------------------------------------------------------------------- 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] --------------------------------------------------------------------
