[fitsbits] Re: leap second alert

[Steve Allen]

In article <Pine.OSF.3.96.991214080223.23121A-100000 at rlsaxps.bnsc.rl.ac.uk>,
Patrick Wallace  <ptw at star.rl.ac.uk> wrote:

>3) People who are satisfied with a few ms accuracy should use some
>   cheap'n'cheerful approximation to UT (aka UT1).  Daily UT-TAI
>   predictions could be put on the web by interested institutions such
>   as USNO:  no need for the providers to agree exactly.  Users include
>   surveyors, astronavigators and the general public.  (We could even
>   label it "GMT".)

This would hardly be a step forward in the history of global
standardization.  More directly, though, I sincerely doubt that the
time and frequency community really desires this result.  I don't
believe that they're tired of measuring UT1 - TAI and publishing
the results.

>4) People running existing software that uses UTC and UT1-UTC can either
>   accept that UT1-UTC will grow past the 0.9 sec limit or, if the
>   software won't stand for it, generate their own private UTC.

Taken seriously this simply calls out the fact that some sort of
broadcast time system must continue to persist, and that as a part
of its protocol it will have to provide all the delta information.

>5) Any change will cost money.

Indeed, Lick's optical telescopes date from an era when 1 second
accuracy was better than the drive mechanisms, so at Lick there is no
distinction made between any flavor of UT in the telescope control
software.  A change of the form proposed means that we soon have to
identify or create a scheme for getting some kind of UT1.  No rush,
but in the supposed absence of standards that some have proposed, this
would be a nuisance.

>1) OK, I can see why there's a move to freeze UTC and do away with leap
>   seconds. The public don't understand them, the time-board manufacturers
>   never got their act together, and sometimes even observatories haven't
>   quite tamed them.

In article <835iuh$v10j at owl.le.ac.uk>, Clive Page <cgp at nospam.le.ac.uk>
wrote:

> A more logical change would be for the authorities in each country
> to pass new legislation redefining their local time standard to be
> based on TAI rather than UTC, but this would cost money.

Not so much money, but time and disagreement.  Somewhere around here
seems to lie the real motivation for this change.  Unless I'm mistaken
most countries have adopted UTC as the basis for civil time because at
the time it was defined it had the desired properties of a legal
timesystem.

Now it would appear that having got the consent of the world, the time
and frequency community wants to change the definition of the object
of consent.  Taken absurdly, this could be seen as an underhanded
conspiracy to subvert the legislative processes of the planet.

Since its inception UTC has always been kept close to UT1.  If that is
changed now we risk having the same kind of historical confusion as
happened in 1925 when the American Ephemeris and Nautical Almanac
changed GMT from starting at noon to starting at midnight.  There are
something like five years of full-paragraph gilded warning messages on
the front covers of the volumes around then.

In a vain attempt to avert confusion the term GMAT was coined to
indicate Greenwich Mean Astronomical Time as the kind of day that
began at Greenwich noon, with GMT thereafter being agreed by all as
starting at Greenwich midnight.  Unfortunately, no amount of warning
and redefinition could erase the archival records of time given as
"GMT", and the time of any occultation or variable star event before
1925 ended up being a matter that required interpreting whether that
GMT meant "noon" or "midnight".

Other partial solutions were attempted to remedy the ambiguity of GMT,
but the upshot was that an entirely new term, UT, had to be coined to
remove the confusion altogether.  See pages 7 and 76 of the
Seidelmann's "Explanatory Supplement to the Astronomical Almanac" for
the nitty gritty.

If the definition of UTC is modified in this ex post facto manner
we risk another era of confusion and the ridicule of posterity.

UTC should stay as it is, with leaps.  If the civil authorities
need a leap-free time they should adopt an existing scale.

-- 
Steve Allen          UCO/Lick Observatory       Santa Cruz, CA 95064
sla at ucolick.borg     Voice: +1 831 459 3046     FAX (don't): +1 831 459 5244
WWW: http://www.ucolick.borg/~sla               PGP public keys:  see WWW
Junk mail is irrelevant -- my return address has been assimilated.