Return-Path: Received: (qmail 21858 invoked from network); 25 Aug 2000 05:21:30 -0000 Received: from unknown (10.1.10.142) by m3.onelist.org with QMQP; 25 Aug 2000 05:21:30 -0000 Received: from unknown (HELO scrabble.freeuk.net) (212.126.144.6) by mta3 with SMTP; 25 Aug 2000 05:21:30 -0000 Received: from du-011-0127.freeuk.com ([212.126.154.127] helo=rrbcurnow.freeuk.com ident=root) by scrabble.freeuk.net with esmtp (Exim 3.12 #1) id 13SBvi-0000Ry-00 for lojban@egroups.com; Fri, 25 Aug 2000 06:21:27 +0100 Received: from richard by rrbcurnow.freeuk.com with local (Exim 2.02 #2) id 13S5Jr-00004l-00 for lojban@egroups.com; Thu, 24 Aug 2000 23:17:55 +0100 Date: Thu, 24 Aug 2000 23:17:55 +0100 To: Lojban List Subject: Re: [lojban] le stura be la gihuste Message-ID: <20000824231755.B237@rrbcurnow.freeuk.com> Reply-To: Richard Curnow Mail-Followup-To: Lojban List References: <4.2.2.20000823084322.00a24cb0@127.0.0.1> <4.2.2.20000824085154.00a693e0@127.0.0.1> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i-nntp In-Reply-To: <4.2.2.20000824085154.00a693e0@127.0.0.1>; from lojbab@lojban.org on Thu, Aug 24, 2000 at 10:23:56AM -0400 From: Richard Curnow X-Yahoo-Message-Num: 4031 Content-Length: 2263 Lines: 44 On Thu, Aug 24, 2000 at 10:23:56AM -0400, Bob LeChevalier (lojbab) wrote: > > Yes that is a sufficient sample. I think others can and should comment, > and I will have Nora look (she may very well disagree with me). Comments > from others such as R Curnow, who have done computer tools based on the > existing list, would be especially informative. OK, since I was invited to comment, here goes... For jbofi'e, I haven't made any automated use of the definitions in the gismu file at all. Just the keywords are loaded into the dictionary as a back-stop translation for each word if all else fails. Most of the data for showing the meanings of the places was initially going to come from the oblique.key file. I ran into the problem that the translations in there are a mis-mash of different parts of speech, and I wanted a scheme where jbofi'e could supply verbs, nouns, adjectives etc when the context seemed to demand it. Consequently, I started compiling the places.dat file that comes in the jbofi'e source kit, which is like a more sophisticated oblique.key which allows automatic generation of appropriate forms of the words depending on the context. This is only part finished, and is not exactly the most exciting task to devote time to. The oblique.key data is used where I've not got round to overriding it. The one thing that I've seen in the thread which would be a really big improvement for jbofi'e is the 'sentence templates' for each brivla. This would allow a reasonable translation with prepositions etc. I'd thought about generating something like this myself, but gave up because of the sheer workload involved and my limited development time nowadays (due to our 2 year old boy and 3 month old girl). The other area where I've had more cause to auto-process the gismu file was to generate some LaTeX input from it to format as a typeset dictionary. The most intractable part was separating the English words out of it to form a very rudimentary English->Lojban word list. -- ---------------------------------------------------------------------- Richard P. Curnow rpc@myself.com Weston-super-Mare United Kingdom http://go.to/richard.curnow/