Re: Leap Seconds

[Joseph P Vlietstra <joevl@concentric.net>]

I have a few recommendations/notes:

1. Subscribe to the leap second mailing list from USNO
   I don't have exact URL, since I receive leap second e-mail at work
and I subscribe
   to comp.lang.ada at home.  As you have noted, leap seconds are
painful to deal with.
   There is a proposal from the IAU Working Group on Reference Systems
(WGRS) to
   eliminate leap seconds from UTC.  If implemented, either UTC will be
redefined to
   be continuous (no leap seconds) or a new continuous time scale, UTS,
will be defined.
   Keep your fingers crossed.

2. Leap seconds are not introduced due to the Earth rotation SLOWING
down per se,
   they are more due to the fact that the Earth had already SLOWED down
slightly when
   the UTC time scale was defined.

3. If your system has exact time synchronization requirements, there is
probably a
   GPS receiver somewhere in the system.  If there is a GPS receiver,
you're in
   luck (sort of).  GPS provides a leap second correction factor to
compute UTC.
   It also provides information on future leap seconds on one of GPS
almanac pages.
   GPS seems to be the best mechanism of distributing the leap second announcement
   -- the Jan 1999 leap second was announce via GPS almost a day before
I received
   the official announcement from IERS (next note)

4. IERS Bulletin C (official leap second announcement)
   Issued at least 6 months before the leap second is introduced.
   In US, you can get it from http://maia.usno.navy.mil
   While you're at USNO, fetch the leap second file at
   ftp://maia.usno.navy.mil/ser7/tai-utc.dat
   We use this file in our leap second computations.