Re: [LEAPSECS] The real problem with leap seconds

From: Michael Sokolov <msokolov_at_IVAN.HARHAN.ORG>
Date: Sun, 8 Jan 06 01:56:05 GMT

Ed Davies <ls_at_EDAVIES.NILDRAM.CO.UK> wrote:

> UTC is expressible as a real number in just the same way that
> Gregorian dates (with months with different lengths and leap
> days) can be with the Julian calendar.
>
> There's no difference in principle between converting from a
> TAI time in seconds since some epoch to a UTC date/time in
> days, hours, minutes, seconds and fractions of a second [...]

You have dodged the problem so conveniently! Of course I know how to
convert UTC to TAI or vice-versa, but that is not the problem statement
at hand. The problem statement at hand is to express UTC *itself* as a
real number, rather than convert it to some other time scale. For UTC
itself must be expressible as a real number in order to be called a time
*scale*. If you admit that this cannot be done, then you should revise
TF.460-6 to remove all use of the word "scale" and openly admit to UTC
being a time non-scale. Then no one will use UTC as the civil time
scale since it'll be obvious that as a non-scale it is not suitable as a
scale of any kind.

I stand by my assertion that the current ITU-R spec for UTC (and its
previous CCIR versions) is a clever scam, a parlor trick designed to
sell a non-scale to civil philosophers wanting a SCALE of civil time.
The manner in which it was adopted in 1970 by CCIR, a shove-down-the-
throat move reminiscent of the current leap hour scheme, does not help
them look any better.

MS
Received on Sat Jan 07 2006 - 17:56:20 PST