Return-Path: <@segate.sunet.se:LOJBAN%CUVMB.BITNET@BITMAIL.LSOFT.COM> Received: from segate.sunet.se by xiron.pc.helsinki.fi with smtp (Linux Smail3.1.28.1 #1) id m0sxQLz-0000ZOC; Tue, 26 Sep 95 05:06 EET Message-Id: Received: from listmail.sunet.se by segate.sunet.se (LSMTP for OpenVMS v0.1a) with SMTP id DB0F9F3F ; Tue, 26 Sep 1995 3:51:42 +0200 Date: Mon, 25 Sep 1995 22:49:48 -0400 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: gismu list question X-To: lojban@cuvmb.cc.columbia.edu To: Veijo Vilva Content-Length: 1257 Lines: 27 Cyril asks: >Subject: gismu list format > >Can anybody explain me, what means this funny letters and numbers in >gismu list between English definition and comments? I belive it is some >info for LogFlash, but I haven't them... It just so happens that I answered this for someone else just a couple days ago: >Those are just codes for sorting - the 2-letters/letter+number is a code >tied to the order the words were being introduced in the draft lesson, >so that LogFlash could be run with words in that order. But the draft >textbook has been reordered and that code is obsolete. We now >distribute LogFlash with words in descending order by the second code ( >a number which is null for a small number of words) - that number is the >frequency the gismu appeared in the set of all Lojban text and >discussion in our archives up to some point in time, and hence >presumably gives a usefulness index that says how lilely you are to run >into the word when reading about Lojban. The data is NOT that good, and >a comple of words (since = snake) have high values due to the English >false friend which also appeared in the same texts. But it is far >better than alphabetical or random order as a basis for learning the >words. lojbab ----- lojbab