Re: [LEAPSECS] Software requirements

From: Daniel R. Tobias <dan_at_TOBIAS.NAME>
Date: Thu, 22 Dec 2005 08:48:59 -0500

On 21 Dec 2005 at 21:33, Poul-Henning Kamp wrote:

> I think the difference between you and me as a programmer is that
> if you make a mistake, some radio telescope bungles up an observation
> or worst case runs against a mechanical end-stop. If I make a
> mistake in FreeBSD millions of machines may bungle up things I have
> not even dreamt about them doing.
> For Windows the problem is two orders of magnitude higher.

I'm a programmer, myself, and I'm pretty confident that all of the
computers (Windows or Linux) that I have anything to do with running,
maintaining, or writing programs that run on them, will fail to have
any adverse effect due to the presence or absence of a leap second at
the end of this year or at any other time. This confidence comes
from the fact that none of these computers, nor the software running
on them, has any need for continuously-accurate real time syncronized
with a world standard, something I know from the fact that their
system clocks can at times get many seconds (or even minutes) removed
from correct time due to their own inaccuracy, requiring pretty big
jumps forward or backward each time they are reset to the standard
time (which happens automatically at intervals on some machines, and
is done manually from time to time on others). While the system
clock is inaccurate, all software on the machines continues running
with no problem other than that system log timestamps are not exactly
correct. This can be annoying when one is trying to track down a
network problem by comparing logs on different systems, but it is not
life-threatening. Any leap seconds that occur are "lost in the
noise" of the adjustments that need to be made to the system clocks
every time the time is re-synchronized.

I suspect more "real world" computers are in roughly this situation
as opposed to being absolutely dependent on being correct to the
millisecond or microsecond at all times; the system clocks of
practically all computers are just not sufficiently accurate for
that. If computers start having an actual atomic clock on board,
maybe things will be different.

== Dan ==
Dan's Mail Format Site:
Dan's Web Tips:
Dan's Domain Site:
Received on Thu Dec 22 2005 - 06:04:03 PST

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