From lojbab@lojban.org Tue Mar 13 12:32:41 2001 Return-Path: X-Sender: lojbab@lojban.org X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_0_4); 13 Mar 2001 20:32:40 -0000 Received: (qmail 59403 invoked from network); 13 Mar 2001 20:32:39 -0000 Received: from unknown (10.1.10.26) by l7.egroups.com with QMQP; 13 Mar 2001 20:32:39 -0000 Received: from unknown (HELO stmpy-2.cais.net) (205.252.14.72) by mta1 with SMTP; 13 Mar 2001 20:32:39 -0000 Received: from bob.lojban.org (209-8-89-95.dynamic.cais.com [209.8.89.95]) by stmpy-2.cais.net (8.11.1/8.11.1) with ESMTP id f2DKWar54304 for ; Tue, 13 Mar 2001 15:32:36 -0500 (EST) Message-Id: <4.3.2.7.2.20010313151409.00c28130@127.0.0.1> X-Sender: vir1036/pop.cais.com@127.0.0.1 X-Mailer: QUALCOMM Windows Eudora Version 4.3.2 Date: Tue, 13 Mar 2001 15:35:46 -0500 To: lojban@yahoogroups.com Subject: Re: [lojban] Almost 100 (spun from Train Catching) In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed From: "Bob LeChevalier (lojbab)" X-Yahoo-Message-Num: 5813 At 12:14 PM 03/13/2001 -0500, pycyn@aol.com wrote: >I find myself in a xodish situation, trying to work out a new case by analogy >with examples in The Book, in this case dealing with "almost 100" and working >out of sections 8 and 9 of chapter 18. In the latter, apparently similar >cases, like "at least 2" and "more than 2" are handled by {su'o re} and {za'u >re}. Now, the list in that section doesn't include "almost" but does include >"approximately", a slightly more general term in the same area. The previous >section contains another case apparently related, {da'a} "all but" along with >another group ro ... so'u that seem to be relative numbers dependent upon >some groups size (the logics of these are a real pain, depending upon a set >of parameters not very clearly motivated). But, since they all belong to the >same grammatical class, it would seem that they could all be used in the same >constructions and with roughly the same pattern of meaning. Risky %^) If this were the case then "pa" and "so" which are also number/digits should also link to the following number to make such constructions, and thus every multi-digit number would be a weirdness. >So, if {da'a} >could be used like {su'o} to indicate "one less than" whatever number >followed, it seems that {so'a} could be used similarly to indicate some >number slightly less than whatever number followed, {so'a panono} for "almost >100" along with {da'a panono} for 99 (with a fuzz factor probably) and {su'e >panono} for "at most 100" the others being in fact specifications of the last >and sharing its grammar. >But Lojbab says not (at least "I don't think so"). My opinion, because we explicitly built into a couple of words like da'a means to glom onto the following number when it occurs before that number. (This dates from the time when these were NOT all one selma'o, BTW.) There wasn't a general rule. > As near as I can figure >out, his point is that the class these all belong to is that of the numerals >and that, therefeore, these words are to be interpreted numerically. Be careful here. "Numerically" as in like numerals/digits, and not numerically like "numbers". > So >that, for example, {so'a} is a number slightly less than the base, say 8 or 9 >for base 10. No, so'a as a DIGIT seems likely to mean a digit slightly less than the base. As a standalone number, so'a is a pretty large number, relatively speaking. That means that so'a prenu when talking in the context of the world probably means billions, and not "8 or 9". so'a lojbanists hopefully is a number in the hundreds. >{so'a panono} is either 8100 or 9100, with fuzz factors. That is one way I could interpret it, and the one that fits best with the digit concept. Another way might be equivalencing to set a scale (A large number; i.e. 100 or so). The point is that there is no convention. so we have to be careful. In non-initial positions, however, it seems hard to imagine a non-digit interpretation such as: > And >{panoso'a} is 108 or 109 or some other version of "almost 110" >On this version, if generalized slightly (but without changing grammatical >categories), {su'e re} ought to mean (n * 10)+2 for n from 0 to 9, >disjoined, which it >clearly not intention in TB.18.8. Clearly there is a problem in that some of the "digit" words have a special semantics at the beginning of a number. But then there is precedent in that the negative and positive signs have special meaning at the beginning of a number, and no clear meaning anywhere else. If I wanted to have su'ere be interpreted as a two digit number, I thus might have to force the non-initial digital semantics by saying nosu'ere >Since Lojbab notes that many of the strings involving things other than >regular digits but in the same general class (PAs, of which there seem to be >half a dozen varieties) Those varieties are my own informal classification, originally for sorted grouping in LogFlash and in the cmavo list. There is no official difference between a PA1 and a PA3, except that we generally want them separated in a word list. >we seem to ahve a certain amount of freedom here. >There is apparently an established (but not in TB) pattern of these critters >at the right end of strings serving as fuzzy numbers as indicated above: >{soso'a} or {soda'a} as "almost 100" =ninety almost-base or "ninety >base-but-one." It isn't in TB because Cowan and I couldn't agree that these patterns were mandated. And indeed there were other interpretations than the ones I came up with. So as usual we punted to the future per the ancient pc adage "Let usage decide." I still like the digital interpretation wherever it makes sense with possibly a different interpretation for initial position, but only if we can agree what it means because it probably has to go into the dictionary if it is in anyway different from the non-initial position version. 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