Re: [LEAPSECS] the legacy of ephemeris time

From: Poul-Henning Kamp <phk_at_PHK.FREEBSD.DK>
Date: Tue, 23 Dec 2003 00:09:59 +0100

In message <>, "Se
eds, Glen" writes:

>Hmm... there is certainly a problem when it comes to specifying legal time
>referring to the future: sun time is not predictable into the future with a
>high degree of precision. People wanting that kind of precision in legal
>documents would have two choices:

In my 20 years in the computing business, I have yet to see one
single instance where the duration until a future legal event was
specified with second precision. I have seen many instances where
the future legal event was locked to a legal time with second

The distinction is important it be aware off.

Contracts may run until "Noon, 12 december 2012", but they do not
run "for 252460800 seconds".

And I seriously do not belive that they will, until a lot of things
change radically, including wrist watches.

On the other hand, past events are routinely timestamped, sequenced
by these timestamps and time-intervals between them calculated with
sub-second precision.

I belive the same holds true for scientific applications: The
desire to be able to calculate a delta-T in a sane manner (ie:
without lookup tables) pertains almost exclusively to events already
in the past.


Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk_at_FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
Received on Mon Dec 22 2003 - 15:10:15 PST

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