From lojbab@lojban.org Sun May 07 12:03:18 2000 Return-Path: Received: (qmail 7669 invoked from network); 7 May 2000 19:03:18 -0000 Received: from unknown (10.1.10.26) by m1.onelist.org with QMQP; 7 May 2000 19:03:18 -0000 Received: from unknown (HELO stmpy.cais.net) (205.252.14.63) by mta1 with SMTP; 7 May 2000 19:03:18 -0000 Received: from bob (209-8-89-112.dynamic.cais.com [209.8.89.112]) by stmpy.cais.net (8.8.8/8.8.8) with ESMTP id PAA23235 for ; Sun, 7 May 2000 15:01:45 -0400 (EDT) Message-Id: <4.2.2.20000507144806.00ab24a0@127.0.0.1> X-Sender: vir1036/pop.cais.com@127.0.0.1 X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.2 Date: Sun, 07 May 2000 15:05:47 -0400 To: lojban@egroups.com Subject: Re: [lojban] centripetal / big endian pattern In-Reply-To: References: <20000506212358.71572.qmail@hotmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed From: "Bob LeChevalier (lojbab)" X-Yahoo-Message-Num: 2567 At 01:52 PM 05/07/2000 +0000, PILCH Hartmut wrote: >There is a reason to opt for the big endian (2000-07-05) approach: >Lojban has, like Chinese, a loose tanru structure where the determining >part comes before the determined part, as in > > The 2000th year's 7th month's 5th day > >Chinese (and Lojban) don't need a genitive (apostrophe s) particle, which >is why Chinese has the structure 2000-07-05 and so should Lojban. > >The elision argument has no weight, because one can elide both at the end >and at the beginning. Except that, as a matter of convention, Lojban does NOT generally elide beginnings, but rather endings. The orderings of place structures are geared towards end of sentence ellipsis, too. Ellipsis in beginnings often requires marking for grammatical purposes of some kind of place filler, which means that you aren't really eliding, but rather at best abbreviating (witness Jorge's semicolon "pi'e"s) > If it had weight, why does no advocate of the >little endian sequence apply it to the time, e.g. saying "it is 45:09" ? >Or to numbers, saying, eightteen, nineteen, twenty, one-twenty, >two-twenty, .. ? Because of normal language habits. I don't know if German is an exception, but in every language I know, it is very rare to give time in minutes without also stating the hour, and time in seconds is never used (which would be proper "little endian"). We do in English sometimes say "ten after" or " a quarter till", which still has an marked if unstated hour; we don't answer what time it is with "10" unless we mean "10 o'clock" (i.e. hours). Lojban is not trying to change the way language does time. I did originally go for this with my base 12 times, but was soundly voted down in favor of "what other languages do". I am not interested in what computers do or what is efficient or optimal, but rather in what people do in other languages, and that includes ellipsis. >Evidently little endian / centrifugal expressions don't stand up to any >strong testing. They are based on a habit that works only for simple >tasks. > > www.lojban.org > >still seems ok, but nobody would apply little endian to the directory >structure: > > http://paragraph3/section3/chapter2/part4/www.lojban.org > >which would be the only logical consequence of using little endian in >internet URLs. Internet URLs are designed for computers and are not a human language, hence are irrelevant. I will concede that computers, that have no common sense and thus rarely guess elided information based on context, may choose a standard different from that of human beings who normally know what year and month it is, and just don't remember the day. But as you note, Lojban is not monolithically little endian or big endian, because it does clock time differently from dates, so arguing whether one is better than the other is the wrong argument. >Not only should dates be centripetally (big endian) structured, but so >should personal names and addresses of all types. Lojban is not concerned with what "should be". This is not a language reform project. the question is what is internally consistent with the language design, which is not optimal nor is it trying to be. >The little endian is an >illogical mess, not only in Lojban but also in English, whose messy >conventions have unfortunately achieved world-wide dominance and has even >entered the realm of URLs and email addresses. Lojban is not the ideal place to introduce reform. 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 (newly updated!)