From edward.cherlin.sy.67@aya.yale.edu Mon Jun 18 10:12:08 2001 Return-Path: X-Sender: Edward.Cherlin.SY.67@aya.yale.edu X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_1_3); 18 Jun 2001 17:12:08 -0000 Received: (qmail 38916 invoked from network); 18 Jun 2001 17:10:45 -0000 Received: from unknown (10.1.10.142) by l10.egroups.com with QMQP; 18 Jun 2001 17:10:45 -0000 Received: from unknown (HELO mta7.pltn13.pbi.net) (64.164.98.8) by mta3 with SMTP; 18 Jun 2001 17:10:45 -0000 Received: from mcp.aya.yale.edu ([216.103.90.93]) by mta7.pltn13.pbi.net (Sun Internet Mail Server sims.3.5.2000.03.23.18.03.p10) with ESMTP id <0GF400KM3Y7B7X@mta7.pltn13.pbi.net> for lojban@yahoogroups.com; Mon, 18 Jun 2001 10:07:03 -0700 (PDT) Date: Mon, 18 Jun 2001 10:19:38 -0700 Subject: Re: [lojban] Re: hexadecimal and lojban In-reply-to: X-Sender: cherlin@postoffice.pacbell.net To: lojban@yahoogroups.com Message-id: <5.1.0.14.0.20010618094840.00b08248@postoffice.pacbell.net> MIME-version: 1.0 X-Mailer: QUALCOMM Windows Eudora Version 5.1 Content-type: text/plain; charset="us-ascii"; format=flowed From: Edward Cherlin At 08:58 AM 6/18/2001, pycyn@aol.com wrote: >In a message dated 6/18/2001 2:39:19 AM Central Daylight Time, >thinkit8@lycos.com writes: > > >>anyway, i was hoping the creators really >>made dau-vai with the hope of hexadecimal as default in the future. > > >As the specification of bases makes clear, Lojban considers decimal as >fundamental. I'm sorry to hear that. It should only be a default. >Anyway, duodecimal is the most convenient form for humans >(finger counting aside). Humans have uses for binary, octal, decimal, duodecimal, hexadecimal, base 20, base 60, and all sorts of mixed bases, starting with 24 60 60. I would prefer to see less cultural bias on this point, and less insistence that one or another representation is inherently superior. The best handling of base representations and evaluations is in APL and J, which have them both as primitives called encode and decode, respectively. Given the time of day in seconds as T, we can evaluate the J expression 24 60 60 #. T to encode the time as hours, minutes, and seconds. Given time in a vector (aka array) HMS, we can convert to seconds with 24 60 60 #: HMS This also works for some of the truly weird bases, such as factorial representation, non-integer bases, and complex bases, which can be found in the writings of Knuth and Iverson. Fibonacci representations require a bit more work.