From rizen@ispwest.com Wed Mar 13 19:02:32 2002 Return-Path: X-Sender: rizen@ispwest.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: unknown); 14 Mar 2002 03:02:31 -0000 Received: (qmail 44097 invoked from network); 14 Mar 2002 03:02:31 -0000 Received: from unknown (216.115.97.172) by m10.grp.snv.yahoo.com with QMQP; 14 Mar 2002 03:02:31 -0000 Received: from unknown (HELO ispwestemail.aceweb.net) (216.52.245.18) by mta2.grp.snv.yahoo.com with SMTP; 14 Mar 2002 03:02:31 -0000 Received: from there (unverified [66.2.47.25]) by ispwestemail.aceweb.net (Vircom SMTPRS 1.2.221) with SMTP id for ; Wed, 13 Mar 2002 19:00:05 -0800 Message-ID: Content-Type: text/plain; charset="iso-8859-1" To: lojban@yahoogroups.com Subject: Re: [lojban] lojban application in wearable computing Date: Wed, 13 Mar 2002 18:58:49 -0800 X-Mailer: KMail [version 1.3.1] References: In-Reply-To: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Ted Reed X-Yahoo-Group-Post: member; u=104181342 X-Yahoo-Profile: xrizen On Wednesday, March 13 2002 06:44 pm, Jay Kominek wrote: > On Wed, 13 Mar 2002, Ted Reed wrote: > > > > Perhaps not lisp, but scheme tends to be organized around predicate > > > > syntax and brackets to define things that are more than one word. > > > > > > not necessarily, but again if your list head symbols are defined as > > > predicates (or functions returning a boolean value) it would be like > > > you say. > > > > My question is, why do we even have to base it on a preexisting proglang? > > It seems to me that lojban already has its own syntax and trying to force > > it into the confines of a preexisting language would limit the > > effectiveness of using lojban in the first place. > > Not to be rude, but it isn't 'we'. It is whoever actually produces the > code. Certainly. > The idea with Prolog, presumably, would be to translate the Lojban > losslessly into Prolog rules which can be queried. Assuming it is > lossless, your complaint is relatively groundless. (Computer science > has plenty of practice translating machine languages flawlessly, btw.) > > Sounds like xod is going to produce some code, and he asked for > suggestions. Were you planning on helping him? :) > > (By the way, I doubt that writing a quality, fast inference engine is > easy. Hence, it would be very timesaving to use an existing one.) Had I the time, I would already be working on something like this, using a modular, expandable engine with speech recognition and such. However, I have entirely too little time to be taking on another project. > - Jay Kominek > Plus ça change, plus c'est la même chose -- rizen