[LEAPSECS] leap smear

[Joe Gwinn]

At 5:20 PM +0000 9/16/11, Poul-Henning Kamp wrote:

>In message <20110916164012.GF9711 at ucolick.org>, Steve Allen writes:

>>On Fri 2011-09-16T10:54:24 +0000, Poul-Henning Kamp hath writ:

>>>  I can fully understand Googles workaround as a way of coping with

>>>  leap seconds.

>>

>>Unless I'm mistaken it is compliant with POSIX.

>

>Actually, I don't think it is, because it's not UTC for a lot

>more time that the leap-second.  POSIX time is UTC execept

>for during leap-seconds.


POSIX time as defined in the standard is *not* UTC.  That said, in 
practice people slave their POSIX boxes to a GPS receiver, and 
believe whatever it may say.

If the GPS receiver is configured to emit UTC, then UTC it is.  There 
will be a bit of strangeness during and a little while after leaps, 
but life goes on.

If the GPS receiver is instead configured to emit GPS System Time, 
then GPS System Time it is (and there are no leaps).

Joe Gwinn