[LEAPSECS] nails in the coffin of mean solar time
M. Warner Losh
imp at bsdimp.com
Sat Jun 16 18:50:27 EDT 2007
In message: <46731716.9090206 at cox.net>
Greg Hennessy <greg.hennessy at cox.net> writes:
: > Can we please stop using totally bogus arguments ?
: > The chances that N future generations would use unmodified vesions
: > of whatever technical designs we come up with, be it timescales
: > or legal arrangements is NIL.
: That depends on how bad of a job we do. Of course if we do a *bad* job
: now, we can be sure it won't be used.
: "Ignore leap seconds since they are hard to get right." is a solution
: that ends up with noon being in the middle of the night, although not in
: my lifetime. That doesn't mean I want to change time standards though.
It doesn't necessarily end there. Since there will be a quadratic
acceleration of leap seconds even doing one every 6 months will
eventually lead to that as well. In a few thousand years, we'll need
hundreds a year. That's not going to work at all.
A simple work around is to move time zones over one every few hundred
years. That's a lot easier than getting leap seconds right as most
places do DSL. A single year without a DSL transition would suffice.
And each nation could choose the year they do it too, as there's
really no need to coordinate. This is even easier than the leap hour
suggestion because it keeps the civil time more or less aligned, and
it uses the same granularity that people are used to today.
More information about the LEAPSECS