FreeBSD handles leapsecond correctly

Kevin Oberman oberman at
Mon Jan 2 15:31:57 PST 2006

> Date: Tue, 3 Jan 2006 10:22:08 +1100
> From: Peter Jeremy <PeterJeremy at>
> Sender: owner-freebsd-current at
> On Mon, 2006-Jan-02 23:27:31 +0100, Poul-Henning Kamp wrote:
> >Interestingly, the main reason why calendar reform is a no-talk
> >issue seems to be that The Vatican owns the standardization area
> >of calendars because they have written all (relevant) standards for
> >the area in the past.
> Actually, the Vatican has only been responsible for one variant of the
> Western calendar (the Gregorian).  The Julian calendar predates the
> Christian church.  The Vatican was open to calendar reform at the time
> because trying to combine elements of both Lunar and Solar calendars
> with a relatively inaccurate year length meant that Easter was heading
> for Christmas - which was felt to be undesirable.  The actual
> algorithm was not developed within the Vatican but was promulgated by
> the Vatican because it was about the only international body which
> more than one or two countries would actually take any notice of - and
> even so, it wasn't until the 20th Century that (eg) Russia switched.
> Islam has its own calendar (with a particularly painful Leap Year
> calculation that gives very marginally more accuracy than the
> Gregorian).  I'm not sure how the Chinese, Hindu, Japanese and Jewish
> calendars handle leap years.  (I think that covers the major non-
> Gregorian calendars).
> >According to one insider, none of the possible owners of a new
> >calendar (ITU, ISO & BIPM) can bring it on themselves to ask the
> >pope permission to take over the area.
> The Gregorian algorithm is reasonably accurate - I think it's good for
> about 1 day in 3000 years.  I don't know how this compares to the
> magnitude of miscellaneous perturbations in the Earth's orbit but it's
> likely to be a couple of thousand years before a further adjustmens is
> necessary.
> The obvious solution is to shift the Earth's orbit slightly towards
> the Sun to provide an even 365 day period.  To make things even
> simpler, move the Earth a bit further to give a 360 day period and
> move the moon a bit further away to give a 30 day period.  (Of course
> the downside of the latter change is that solar eclipses would become
> even rarer and might even disappear totally).  With a bit of tweaking
> of the Earth's rotation, the need for leap seconds can also be removed.

We keep trying to define the universe and keep getting annoyed when the
universe fails to cooperate. I still want pi to be 3.14. Yeah, it would
warp the universe rather badly, but would make the arithmetic so much
easier. And I guess sin(90) would no longer be rational, either. Guess
we just can't win. :-(

I like the old "joke":
Q. How do you make God laugh?
A. Tell him your plans.
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman at			Phone: +1 510 486-8634

More information about the freebsd-current mailing list