Re: [LEAPSECS] The real problem with leap seconds

From: M. Warner Losh <imp_at_BSDIMP.COM>
Date: Sat, 07 Jan 2006 13:32:16 -0700 (MST)

In message: <43BFD852.23658.6B516D4_at_dan.tobias.name>
            "Daniel R. Tobias" <dan_at_TOBIAS.NAME> writes:
: On 7 Jan 2006 at 16:02, Poul-Henning Kamp wrote:
:
: > Civil time is in the hands of individual governments, and they
: > tend to expect their computers to use the same time as the
: > rest of their country.
:
: And, in many countries (including the United States), the legally-
: defined civil time is the mean solar time at some particular spot, or
: a fixed or seasonally-variable offset from it. Any use of UTC-based
: time scales for determining civil time in such places is merely an
: approximation, currently to within a second, but perhaps varying by
: greater amounts if some new timekeeping plan is adopted. Once UTC
: stopped being a sufficiently-close approximation to the solar mean
: time at the Prime Meridian (with "sufficiently close" possibly being
: of differing values depending on the particular purpose), it would be
: necessary to either stop using UTC for determining civil time in such
: countries, or to change the law to base the civil time on UTC instead
: of a solar standard.

The US's legislation leaves it to the commerce department to define
what the mean solar time is (or rather states that it is the mean
solar time, as determined by the secretary of commerce). This is what
presumably gives us leap seconds and the like in our civil time and
when we insert them. This is a political function: Up to a second of
tolerance is allowed and leap seconds are inserted whenever everybody
else does. The same political decision could be made to say something
else, since it is the folks in DC could say that a minute is close
enough to solar mean and that's what we're going to do.

Warner
Received on Sat Jan 07 2006 - 12:34:25 PST