Re: [LEAPSECS] The real problem with leap seconds

From: Mark Calabretta <mcalabre_at_atnf.CSIRO.AU>
Date: Fri, 20 Jan 2006 11:32:26 +1100

On Thu 2006/01/19 16:36:45 BST, "Deckers, Michael" wrote
in a message to: LEAPSECS_at_ROM.USNO.NAVY.MIL

> of UTC where UTC is not just taken as the (discontinuous) timescale
> TAI - DTAI.

It's possible that I have misunderstood your email (after many readings)
but it seems that we are going around in semantic circles.

Here is my original description of what happened at the recent leap
second with the addition of another scrunched-up column containing
TAI-DTAI computed in bog-standard sexagesimal:

        UTC TAI DTAI TAI-DTAI
2005/12/31 23:59:58 2006/01/01 00:00:30 32 2005/12/31 23:59:58
2005/12/31 23:59:59 2006/01/01 00:00:31 32 2005/12/31 23:59:59
2005/12/31 23:59:60 2006/01/01 00:00:32 32 2006/01/01 00:00:00
                 60.9 32.9 32 00.9
                 60.99 32.99 32 00.99
                 60.999... 32.999... 32 00.999...
2006/01/01 00:00:00 2006/01/01 00:00:33 33 2006/01/01 00:00:00
2006/01/01 00:00:01 2006/01/01 00:00:34 33 2006/01/01 00:00:01

As you can see, TAI-DTAI is discontinuous (generally an undesirable
property for any timescale!), but UTC isn't. That is "the trick".

> What leads me to the more pertinent question: if UTC really is just
> TAI in disguise, why shouldn't we adopt a new UTC as TAI - 33 s,
>without
> any disguise? Not that I am advocating the departure of civil time
>from
> solar time -- I am just checking the arguments against it.

We rely on "the trick" to tie TAI-based civil timekeeping close to UT1.
TAI-DTAI as above, if that's what you are suggesting as an alternative,
wouldn't be acceptable.

Mark Calabretta
ATNF
Received on Thu Jan 19 2006 - 16:32:54 PST