Return-Path: <@SEGATE.SUNET.SE:LOJBAN@CUVMB.BITNET> Received: from SEGATE.SUNET.SE by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0sYp9B-0000ZFC; Thu, 20 Jul 95 09:31 EET DST Message-Id: Received: from segate.sunet.se by SEGATE.SUNET.SE (LSMTP for OpenVMS v0.1a) with SMTP id 8DD95323 ; Thu, 20 Jul 1995 8:30:37 +0200 Date: Thu, 20 Jul 1995 02:29:58 -0400 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: Re: conceptually related gismu X-To: jorge@PHYAST.PITT.EDU X-cc: lojban@cuvmb.cc.columbia.edu To: Veijo Vilva Content-Length: 718 Lines: 12 I agree that thesaurus-like classifications are troublesome for Lojban. The concept was not desisgned for a predicate language, but for a "word" language. In general thhe classifications are thus based on the x1 of the predicate. It should be possible to extend the Roget-like work to all places of all predicates and you would come closer to an accurtae thesaurus of Lojban, but no one hhas ttackled this. I don;t think the thesaurus work is too strongly based on the English keyowrds so myuchh as it is on the x1-to-predicate relationship in whatever language, though of course Athelstan's early work was spotty in this because the keywords have been more stable than the poace structures for some words. lojbab