From nellardo@concentric.net Fri May 12 13:05:35 2000 Return-Path: Received: (qmail 6046 invoked from network); 12 May 2000 20:04:36 -0000 Received: from unknown (10.1.10.142) by m2.onelist.org with QMQP; 12 May 2000 20:04:36 -0000 Received: from unknown (HELO uhura.concentric.net) (206.173.118.93) by mta3 with SMTP; 12 May 2000 20:04:35 -0000 Received: from marconi.concentric.net (marconi.concentric.net [206.173.118.71]) by uhura.concentric.net (8.9.1a/(98/12/15 5.12)) id QAA21690; Fri, 12 May 2000 16:04:33 -0400 (EDT) [1-800-745-2747 The Concentric Network] Errors-To: Received: from concentric.net ([216.112.226.144]) by marconi.concentric.net (8.9.1a) id QAA22151; Fri, 12 May 2000 16:04:32 -0400 (EDT) Message-ID: <391C6202.E7D93A39@concentric.net> Date: Fri, 12 May 2000 16:06:04 -0400 Reply-To: nellardo@concentric.net Organization: Herds of Wild Buffalo Girls X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: lojban@egroups.com Subject: A thought on dates: big-endian mekso and little-endian bridi Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit From: Brook Conner So there's been all this argument about big-endian vs. little-endian, and I can't help but put myself at risk of flames by being baffled about the argument, at least in respect to dates. mekso are big-endian and bridi are little-endian. With mekso, of course, you need to specify a different base, but the language allows for that. And it allows for more than one "decimal point" - I recall that time is a specific example. lojban numbers (and numbers in most human languages) are big-endian - the largest unit comes first. Whether that is thousands vs hundreds, pounds vs ounces, francs vs centimes, or hours vs minutes. Now, if there were a gismu that was an assertion of a particular point in time as existing ("date" roughly, but not quite), then (let's pretend it's "datci" - which, despite sounding like a gismu, doesn't appear to be) let's suppose that the various places mark the time steps: * datci - /x1/ seconds, /x2/ minutes, /x3/ hours, /x4/ days, /x5/ years or alternatively: * datci - /x1/ seconds, /x2/ minutes, /x3/ hours, /x4/ days, /x5/ months, /x6/ years This is of course somewhat un-lojban-like, as it doesn't have the more usual pattern of "thing,aspect,frame of reference" kind of stuff going on. However, "The time is 5:15pm": la cabna cu datci la pamu la paze Literally, "The second called now exists at minute named fifteen at hour named 17" Prefer hours then minutes? Several ways to say it: la cabna cu datci fi la paze fe la pamu marking places explicitly la cabna cu tesete datci la paze la pamu using the little-used compound conversions la cabna fi la paze fe la pamu cu datci "The time, 17:15 it is" - Yoda :-) la fi revo datci - 24:00, i.e., "midnight" la fi no datci - 00:00, also midnight la fi pare datci - 12:00, noon la midju datci - "The middle moment", could be noon, midnight, midweek, midmonth, midyear, depending on context. So how is this different from the *actual* lojban gismu mokca? mokca - x1 is point/moment/instant (dimensionless) in/on/at time/place x2 mokca, first of all, requires a more complex mekso than a simple number. Secondly, and more epistemologically, mokca requires identification of the space-time instant. x1 is that space time instant. datci does not so require an identification. In effect, the whole bridi is that identification. On the other hand, mokca covers both space time and is unit-neutral - you can use Martian hours and kilometers to specify the space-time moment. Of course, our hypothetical datci could be modified, or the units could be specified explicitly as part of x1-n (with the time ones only being the default). Some thoughts..... not meant to be a criticism of lojban, if anything, a statement of how well things fit in.... Brook