[fitsbits] Re: leap second alert

[Rob Seaman]

I'm glad to see that this discussion is getting an airing.  I started
the thread since I was concerned that the backdoor communication that
NOAO received was the extent of the time and frequency community's
interest in our opinion.

Steve says:

> It was just 2 years ago that the FITS community declared that UTC
> shall be used in the value of the 'DATE' keyword.  This decision
> was based on the presumption that UTC would remain the most
> widely-available timescale for earthbound FITS file writers.

Perhaps we could convince them to give up the idea of dropping leap
seconds using the "once FITS, always FITS" rule ;-)

Note that the real reason I posted to this group is that sci.astro.fits
is a de facto ADASS communications channel.  In addition to the obvious
FITS issues, I knew most of the world's astronomical software types
would be in touch with this group either directly or at most by one
"degree of separation".

Mark's analogy of this as a Y2K type error in NTP is probably getting
closer to the true motivation behind this strangely political initiative.

Nobody would care about purely astronomical time systems (and we're big
enough boys and girls to deal with such ourselves).  The pressure to
muck with something as internationally fundamental as time must be coming
from some widespread community that has a mandate to agree closely with a
legally negotiated time standard.  Are there any other candidates than NTP?
(Or perhaps the telecommunications community that spawned NTP?)

I might suggest that the astronomical community in general, and the FITS
community in particular, should reach a unified consensus to present to
the faceless time and frequency bureaucrats.  (For some reason, the movie
"Brazil" springs to mind.)  Asking for input via some poll (especially a
poorly distributed poll) is just a way of dividing and conquering.

First - what does astronomy need?  (And therefore what must FITS support?)
Second - what are the resulting implications of this for civil time?

And if the motivation behind this is really just to keep some internet
nerds happily free of the obligation to track leap seconds, can we offer
any help improving the NTP infrastructure?

Smithsonian Magazine has a picture of Harrison's Number 4 clock on the
cover of the last issue of the millennium.  Villanovans who took two
semesters of the history of Astronomy from the good Augustinian Father
Edward Jenkins may no longer be the only ones with a knowledge and
fondness for Harrison's clock.  And astronomers should think twice
before relinquishing their role as the world's original timekeepers.

Rob

-- 
seaman at noao.edu, http://iraf.noao.edu/~seaman
NOAO, 950 N Cherry Ave, Tucson AZ 85719, 520-318-8248