FreeBSD handles leapsecond correctly
oberman at es.net
Mon Jan 2 14:19:53 PST 2006
> Date: Mon, 2 Jan 2006 22:19:56 +0100
> From: Matthias Andree <matthias.andree at gmx.de>
> Sender: owner-freebsd-current at freebsd.org
> On Mon, 02 Jan 2006, Poul-Henning Kamp wrote:
> > In message <m3psnagxrb.fsf at merlin.emma.line.org>, Matthias Andree writes:
> > >And tell me one reason why the leap second must be discontinued while
> > >the leap day (Feb 29th) can be carried on. It's the same story,
> > >irregular rollover, inserting one particular unit of time.
> > You are clearly not thinking rationally here.
> > I know already now that year 2048 will be a leap year, but I still
> > don't know if there will be a leap second on june 30th 2006.
> And you can predict the DST rules for all major countries for 2048? Who
> says the EU won't discontinue DST effective 2008? We don't know yet.
They can, but FreeBSD and any POSIX system keeps time in UTC. DST has no
impact on UTC, GMT, TAI, or the like.
> You suggest UTC needs to be used because civil time matters, yet at the
> same time UTC were broken, and thus POSIX were broken, but could not be
> blamed for picking UTC.
Civil time matters for many of us. UTC is certainly not perfect, but
it's more practical than GMT and more real than TAI. It provides a
consistent, very accurately defined second which is VERY important to
many things which don't care an iota what time it is.
> Leap days (called leap year, to compensate for earth orbiting the sun),
> leap hours (called daylight savings time, completely artificial); aren't
> questioned, but leap seconds are.
Leap years are an artifact of a lousy calendar that has origins over 2
millennia ago. Many calendars have been proposed which fix this, but
calendar reform is simply not going to happen in our society, but leap
years are a known, non-varying and trivially calculable issue. No magic
and trivial to handle.
Leap seconds occur because the earth does not spin quite
uniformly. There is no way to know exactly when one will be needed,
although we are pretty sure that they will be needed more often in the
future. Their only purpose is to keep GMT and UTC close.
Are they needed? Get an astronomer and a physicist in the same room and
ask them. (Make sure that they are physically separated, first, if you
want to avoid any bloodshed.)
> Is it just me who sees inconsistencies in your argumentation here?
There are no known good solutions to this problem, but your arguments
are mostly orthogonal to the problem.
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman at es.net Phone: +1 510 486-8634
More information about the freebsd-current