From jimc@MATH.UCLA.EDU Fri Sep 14 11:26:37 2001 Return-Path: X-Sender: jimc@math.ucla.edu X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_3_2_2); 14 Sep 2001 18:26:37 -0000 Received: (qmail 33432 invoked from network); 14 Sep 2001 16:43:52 -0000 Received: from unknown (10.1.10.142) by l8.egroups.com with QMQP; 14 Sep 2001 16:43:52 -0000 Received: from unknown (HELO simba.math.ucla.edu) (128.97.4.125) by mta3 with SMTP; 14 Sep 2001 16:43:52 -0000 Received: from localhost (jimc@localhost) by simba.math.ucla.edu (8.11.3/8.11.3/SuSE Linux 8.11.1-0.5) with ESMTP id f8EGhmT01375 for ; Fri, 14 Sep 2001 09:43:48 -0700 X-Authentication-Warning: simba.math.ucla.edu: jimc owned process doing -bs Date: Fri, 14 Sep 2001 09:43:48 -0700 (PDT) To: Subject: Re: [lojban] Overspecifying places (was: (lojban-beginners) pi'e) In-Reply-To: <01091408194100.01115@linux> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT From: "James F. Carter" On Fri, 14 Sep 2001, [utf-8] Björn Gohla wrote (replying to Nick): > i am getting the feeling that there is quite a tendency in lojban to > overspecify things. i consider date formatting to be beyond the realm of the > language itself, just like alot of the metaphysics we tend to talk about. so > instead of embossing a specific world view into lojban we should concentrate > on making as much as possible easily and clearly expressible. Your point is not unreasonable. But a gismu is defined to be a word used as a symbol for a relation between... what? You can't have a relation, therefore can't have a gismu, without parameters that are thus related. My canonical example is, "the rat eats the cheese" differs in meaning from "the cheese eats the rat", but these differ only in the order of parameters, using the parameter <-> place assignment style most often used in both English and Lojban. Thus, it's important to nail down the _order_ of the places (parameters) for each gismu, as well as how each place participates in the relation. But in the case of /detri/, x1 is clearly the date -- there's no argument there. The gismu list does specify the precise time part order, which is what's being argued about. Here's a Modest Proposal: /detri/ = "x1 is the date of event x2 at location x3 by calendar x4; x1 to be expressed in the customary format of x4 using /pi'e/ to join time parts." Thus pc, when using /detri/, will assume x4 = the Gregorian calendar expressed in provincial format, while people who have to live with data exchange will assume x4 = the Gregorian calendar in ISO-8601 format. And the Mayans will use the special time parts unique to their culture. So I'm actually agreeing with Björn that /detri/ X1 is overspecified in the gismu list. More argument fodder (but maybe we've put enough time on this already): when dealing with "the month of Caprock" or "the year of the fruit bat", does it really make sense to "massify the unit values with /joi/"? Does it make sense with /pi'e/? How many angels can dance on the head of a pin? James F. Carter Voice 310 825 2897 FAX 310 206 6673 UCLA-Mathnet; 6115 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA 90095-1555 Email: jimc@math.ucla.edu http://www.math.ucla.edu/~jimc (q.v. for PGP key)