From lojbab@lojban.org Thu Aug 08 16:20:21 2002 Return-Path: X-Sender: lojbab@lojban.org X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-8_0_7_4); 8 Aug 2002 23:20:21 -0000 Received: (qmail 76989 invoked from network); 8 Aug 2002 23:20:21 -0000 Received: from unknown (66.218.66.217) by m3.grp.scd.yahoo.com with QMQP; 8 Aug 2002 23:20:21 -0000 Received: from unknown (HELO lakemtao03.cox.net) (68.1.17.242) by mta2.grp.scd.yahoo.com with SMTP; 8 Aug 2002 23:20:20 -0000 Received: from lojban.lojban.org ([68.100.206.153]) by lakemtao03.cox.net (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with ESMTP id <20020808232021.ZMPQ1975.lakemtao03.cox.net@lojban.lojban.org> for ; Thu, 8 Aug 2002 19:20:21 -0400 Message-Id: <5.1.0.14.0.20020808190443.00ac6830@pop.east.cox.net> X-Sender: rlechevalier@pop.east.cox.net X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Thu, 08 Aug 2002 19:14:28 -0400 To: lojban@yahoogroups.com Subject: Re: [lojban] Gismu Thesaurus In-Reply-To: <0208081711360M.18087@neofelis> References: <20020808154145.Y4556-100000@granite.thestonecutters.net> <20020808154145.Y4556-100000@granite.thestonecutters.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed From: Robert LeChevalier X-Yahoo-Group-Post: member; u=1120595 X-Yahoo-Profile: lojbab At 05:11 PM 8/8/02 -0400, Pierre Abbat wrote: >I think we need a general brivla thesaurus. Can we make one with the >dictionary database? We should be able to say that {selratni} and {cmuxu'i} >and {xu'icmu} all can be said of chemical elements but the place structures >are different, or that ro kurbita cu guzme but not the reverse. We first need to have a set of agreed upon place structures for the lujvo. And so far there has been little work at devising place structures for any lujvo, nor for verifying the place structures of the lujvo that have been structured. I haven't looked at xod's effort, but there is a lot more work needed in the gismu thesaurus, merely in thesaurusizing the oblique places. The oblique places was why we abandoned the Roget's scheme, BTW - it really doesn't work for multiplace gismu. You could thesaurusize based on the individual places (which are nounish) of each gismu, but Roget's includes categories that are nounish, verbish, and adjectivish all mixed together, and gismu themselves don't work well as verbs or adjectives without the place-limiting use of converters (i.e. the Roget's category for xe klama would likely not be the same as for klama). We really need a new concept of how to break down the gismu (or brivla) - either that or risk the malropno equation of a gismu to the sum of its individual places. lojbab -- lojbab lojbab@lojban.org Bob LeChevalier, President, The Logical Language Group, Inc. 2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273 Artificial language Loglan/Lojban: http://www.lojban.org