[LEAPSECS] 2008-12-31T23:59:60Z

[Steve Allen]

On Tue 2008-12-23T20:08:06 -0800, Brian Garrett hath writ:

> examples of how the leap second

> to be thrust upon us next week is affecting list members' current

> projects.


On-campus we have just suffered a backplane failure in the main disk
array server, followed by a complete shutdown of power to the building
last weekend, to be followed by a transfer from the grid to the local
generator next weekend.  We're nowhere near at the .99999999 level
of reliability that we need to notice a leap second, there are a lot
of bigger issues, and we expect everything to be robust.

On-mountain we do not yet have the fully-robotic telescope which will
be driven by Windows-based software for which we won't have the source
code to tell whether it cares about leap seconds, the older telescopes
point only a little better than battleships, and the leap happens
while the sun is still up.

At Keck the leap happens so nearly at midday that the telescope
drive systems are often offline.

The most evident place for logging the leap would be if a calibration
exposure were in progress.  The recorded elapsed time would not match
the recorded start and end times, but this is common.  None of the
instrument specifications demanded a means of precise shutter event
timing.  The only such event systems in place are the ones that I've
retrofitted into the software, and the constraints of the other system
components mean that they can't guarantee one second accuracy.

In short, we took no precautions at the end of 2005, and we will take
none this year.  I expect to be sitting with my girls listening to
WWV and WWVH.

--
Steve Allen                 <sla at ucolick.org>                WGS-84 (GPS)
UCO/Lick Observatory        Natural Sciences II, Room 165    Lat  +36.99855
University of California    Voice: +1 831 459 3046           Lng -122.06015
Santa Cruz, CA 95064        http://www.ucolick.org/~sla/     Hgt +250 m