[LEAPSECS] WP7A status and Re: clinical evidence about time and sun

M. Warner Losh imp at bsdimp.com
Thu Dec 18 17:46:47 EST 2008


In message: <18762.53127.165662.23992 at gromit.timing.com>
John Hein <jhein at timing.com> writes:

: > Solutions for "applications" can and should rely on properly designed

: > systems

:

: Indeed. And relying on a system whereby you receive six months notice

: is one of the problems with the current system of leap seconds.


I find this ironic too. "We must have a predictable future" is
fundamentally incompatible with the current system of leap seconds.
There's no predictability to the current system at all. There's no
way to know how many leap seconds that will elapse between now and
the end of 2040.

Also, trying to force a system that's been in place for 37 years with
known limitations to last 1000's of years seems like an unwise
engineering move. When Pope Gregor reformed the calendar, it was
based on hundreds if not thousands of years of observations, and was
known to be good for tens of thousands of years without anything else
ever changing, or any need for any authority to dictate leap days.
The necessity to say 'yea or nay' on leap seconds at a given time, as
opposed to a mechanical means of knowing, is a fundamental flaw with
the current system of leap seconds.

I guess the fundamental problem is that eventually, the day will be
86401 seconds.

Warner


More information about the LEAPSECS mailing list