[LEAPSECS] Cost: getting rid of GMT & discontinuing leap seconds

[Warner Losh]

> Do you really consider it reasonable to change the definition to

> which the law refers? When complying with the law proves too

> cumbersome, shouldn't legislators consider a change in the relevant

> law?


Yes.  It happens all the time.  GMT has gone through at least a dozen
redefinitions over the years, depending on how you count things.  UTC
has also evolved since it started in the late 1950's.  Even UT1 has
been refined (and why is it a better realization of mean solar time
than UT2?).  The prime meridian has moved at least twice since it was
fixed.  All these adjustments happen all the time either due to human
error, a better understanding of the problem, or changing needs for
time.

There's nothing magical about leap seconds.  They are not a natural
property of time.  They are merely a means by which the angular
position of the earth relative to the sun can be kept in sync with the
more regular oscillations of cesium atoms.  They are merely a latest
scheme in a long-line of schemes to keep things in sync.

The scientific community long ago decided that having a fixed length
second was more important than having the day divided into 86400
divisions that tick off.  A second, for better or worse, has not been
defined as a part of a day for a very long time (even the pre-atomic
second was defined as a small part of the mean tropical year).


> UTC may (or may not) be the only game in town, but there's no ban on

> new games whatsoever.


True.  But the regulations that kicked off this discussion are quite
explicit that UTC be used, despite the occasional hick-up caused by
leap seconds.

Warner