[fitsbits] Re: leap second alert

[Peter Bunclark]

Steve Allen wrote:

>
> But in any case, to get back to the two questions posed
> by Rob Seaman <seaman at noao.edu> on Thu 1999-12-16T20:05:37 +0000
> > First - what does astronomy need?  (And therefore what must FITS support?)
>

It seems there are several levels of what astronomy needs:
1. High -end professional, eg Gemini TCS:
    totally rigorous system, timekeeping will be `right' no matter what
    bizarre actions the standards people take;

2. Smaller/older professional and high-end amateur - particularly
     those running equatorial telescopes, may well just use  UTC as
     delivered by NTP, since the error is much smaller than their
     hour-angle encoder setting error (both being cancelled simultaneously
     by observation at the start of night).   Data-aquisition systems on such
     systems will also tend to use UTC on the grounds that for most
     time-integration work, accuracy of <1second is precise enough; but
     these applications would be unhappy with UTC drifting away.

3.  Data reduction systems;  need access to `civil' time for bookeeping
      purposes.

All of these would be best served by an NTPng that serves TAI and
 the UTC offset (and leaving the leap-seconds in).  The high-end guys
  could just use it to check their internal time perhaps; others would
   have access to TAI without needing hardware; and existing
   applications would guaranteed not to break.


>
> > Second - what are the resulting implications of this for civil time?
> is there anything which hasn't already been covered here?

The measures of technical advancement of civilisations have been

1. The accuracy of PI

2. The correctness of the calendar


Hey, let's redefine PI to be 22/7, it'll make working out
circumferences and stuff a lot easier!

Pete.