Re: [LEAPSECS] floating prime meridian

From: Steve Allen <sla_at_ucolick.org>
Date: Wed, 6 Sep 2000 16:55:14 -0700

On Fri 2000-09-01T23:11:40 -0400, John Cowan hath writ:
> On Thu, 31 Aug 2000, Rob Seaman wrote:
> > Your number 1 and number 2 are the same issue. If they aren't going
> > to care about the absence of leap seconds - why do they care about
> > their current presence?
>
> Because their unpredictability complicates the design of time software.
> That's *my* interest.

Consider the contents of /usr/share/zoneinfo on any modern Unix system.
Consider the content of URL
 http://www.merlyn.demon.co.uk/uksumtim.htm
Consider the content of URLs
 http://www.microsoft.com/australia/support/timezone/2000.htm
 http://www.austlii.edu.au/au/legis/nsw/consol_act/sta1987137/notes.html#amends

>From the first we see that during the relatively brief history of Unix
there have been untold occasions on which the rules of civil time have
been changed. From the last we see that this trend doesn't seem
likely to stop anytime soon, and that we can't even expect enough
warning for calendar printers not to have to scurry to get the
printing right. We are going to have to deal with them.

Is it really fair to say that time software design can tolerate
completely unpredictable shifts of civil time by one hour at the whim
of regional legislative bodies, but that it cannot tolerate
constrained unpredictable shifts of civil time by one second under the
command of a single global authority?

I don't think so.

--
Steve Allen          UCO/Lick Observatory       Santa Cruz, CA 95064
sla_at_ucolick.org      Voice: +1 831 459 3046     http://www.ucolick.org/~sla
PGP: 1024/E46978C5   F6 78 D1 10 62 94 8F 2E    49 89 0E FE 26 B4 14 93
Received on Wed Sep 06 2000 - 16:55:23 PDT

This archive was generated by hypermail 2.3.0 : Sat Sep 04 2010 - 09:44:54 PDT