Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!wuarchive!uunet!mcsun!ukc!edcastle!aiai!aiai.ed.ac.uk!jkk From: jkk@aiai.ed.ac.uk (John Kingston) Newsgroups: comp.ai Subject: Re: NEXPERT: Its Use In Larger System Development Keywords: Nexpert frames commercial objects Message-ID: <3759@skye.ed.ac.uk> Date: 15 Nov 90 13:20:36 GMT References: <1990Nov14.192306.27466@PacBell.COM> Sender: news@aiai.ed.ac.uk Reply-To: jkk@aiai.uucp (John Kingston) Organization: AI Applications Institute, University of Edinburgh Lines: 74 In article <1990Nov14.192306.27466@PacBell.COM> sire@PacBell.COM (Sheldon Rothenburg) writes: >We are looking to develop the requirements for a new system utilizing >object oriented analysis. We are looking at several alternatives. >Colleagues have suggested that we investigate Nexpert, which we know >has some of the object and/or frame features we desire. Yes, we also >know it sells the most copies of any shell. >The underlying concern is can this tool be considered in the same >class as Kee and Art for handling large, complex applications, or >is it a general purpose tool that occupies some mid-range position >as far as functionality and performance? There was a discussion on the relative merits of CLIPS and NEXPERT a while ago which I saved a copy of. I've included a summary below. Mail me for a full version, Shelley (with several mail addresses if possible). From what I know of NEXPERT, it is a 'mid-range' tool rather than being in the same class as ART or KEE. Other tools in the same category include ART-IM (Inference), KAPPA-PC (IntelliCorp), and ADS (Aion). There are several other tools which might be claimed to fit this category, including ProKappa (IntelliCorp), which is probably near the top of this range, based on hearsay (I haven't seen it yet). CLIPS Pros ------ C Source available makes it very flexible Easy to learn - you can "Get going" right away Easy to use for embedded applications (with good documention) Easy to call C routines from within CLIPS Uses Rete algorithm Cons ------ Not object-oriented (changing in 5.0) One user found it slow and a memory hog on the Sun 3/50 (this was with using only 10 rules) Lack of backward chaining ---------------------------------------------------------------------------- NEXPERT Pros ------ Easy to embed. Agenda Monitor to observe inference engine Graphical Interface Cons _____ This was a big complaint - Documentation hard to understand (suggest taking formal training) One user found forward chaining "tortuously" slow" with medium to large rule base (doesn't use Rete algorithm) Object-Oriented features overrated because does not have message sending capabilities Rule editor is "cumbersome" Debugging painful * I fear for the freedom fighter who chose the bloody road * * Who tried to harness evil to try to lift an evil load * * And I understand the righteous wrath that drove him to what he's done * * But forgiveness lies in nail-scarred hands, not in the hand with a gun * -- Garth Hewitt, **Road To Freedom** John Kingston, AI Applications Institute, University of Edinburgh, 80 South Bridge, Edinburgh EH1 1HN, Scotland E-mail jkk@uk.ac.ed.aiai, phone 031-225 4464 ext. 229 FAX: 031 226 2730 Arpanet: J.Kingston%uk.ac.ed@nfsnet-relay.ac.uk TELEX: 727442 UNIVED G