From xod@sixgirls.org Mon Mar 11 10:38:26 2002 Return-Path: X-Sender: xod@reva.sixgirls.org X-Apparently-To: lojban@yahoogroups.com Received: (EGP: unknown); 11 Mar 2002 18:38:26 -0000 Received: (qmail 98374 invoked from network); 11 Mar 2002 18:36:17 -0000 Received: from unknown (216.115.97.167) by m8.grp.snv.yahoo.com with QMQP; 11 Mar 2002 18:36:17 -0000 Received: from unknown (HELO reva.sixgirls.org) (216.27.131.50) by mta1.grp.snv.yahoo.com with SMTP; 11 Mar 2002 18:36:17 -0000 Received: from localhost (localhost [[UNIX: localhost]]) by reva.sixgirls.org (8.11.6+3.4W/8.11.6) with ESMTP id g2BIaEB09274 for ; Mon, 11 Mar 2002 13:36:14 -0500 (EST) Date: Mon, 11 Mar 2002 13:36:12 -0500 (EST) To: lojban@yahoogroups.com Subject: Re: [lojban] Programming Languages for Lojban In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII From: Invent Yourself X-Yahoo-Group-Post: member; u=1138703 X-Yahoo-Profile: throwing_back_the_apple X-Yahoo-Message-Num: 13615 On Mon, 11 Mar 2002, Robert J. Chassell wrote: > If one were to try to parse Lojban and use it for the internal workings of > an inference engine, would Lisp or Prolog be more appropriate? > > Lisp is perceived as a more general purpose language than Prolog, so > it more closely matches Lojban, which is a completely general purpose > language. > > Also, people have written interpreters for many different programming > languages in Lisp; I don't think as many have been written in Prolog. > So someone who is willing to stand on the shoulders of others will > have an easier time with Lisp. > > Am I right in thinking that you plan to write most of your inference > engine in Lojban, and that you plan to use the Lisp or Prolog as just > a boot strap mechanism? I don't know a thing about parsers, LALR and yacc, AI and inference engines, Lisp or Prolog. Hence, I am asking zero-level questions, like a choice of language. But although there's been plenty of talk about Lojban-comprehending software, none of the people with the appropriate education have started any code so I'll have to do it. I need an inference engine, and I think its internal language should be Lojban. It seems elegant to do as much of the coding in Lojban itself, but I really have no idea how realistic that is. > I am looking forward to your write ups about turning Lojban into a > humanly speakable programming language. That's not me. I have almost no interest in speaking to computers. But I would like to be able to type language and be "understood". And I would like them to be able to communicate to each other about reality in Lojban instead of, say, some XML dialect such as RDF or DAML+OIL. > What is involved in making type checking optional? (The Lojban grammar > makes it so, since you have a choice of including or not including a > restrictive clause, i.e., one that tells you the type of the argument.) > Under which circumstances will people prefer to use an imperative > format, in which the computer is a robot that responds to orders, or > prefer a format in which the computer answers questions? Lojban seems like a strongly typed language; "cusku le gerku" is widely understood to be meaningless. The back-introduction of (the computer language concept of) strong typing into a human language is interesting, and might provide benefits, although I have bristled at pedantry in the past. I would think a decent agent would understand both sentence formats. Some sentences convey knowledge, while others issue orders. -- When a system is in harmony with the Tao, the compiler makes applications and utilities. When a system goes counter to the Tao, accounting logs fill the root directory.