Re: [LEAPSECS] Precise time over time

From: Rob Seaman <>
Date: Wed, 10 Aug 2005 23:33:59 -0700

On Aug 10, 2005, at 9:14 PM, John.Cowan wrote:

> Where on earth do these systems get mean solar time? That requires
> solar observations plus the current equation of time.

An interesting question - that I'm sure one of the time lords could
answer in great detail if at any time over the past five years they
had deigned to participate in this discussion.

The method you suggest would work, but I believe mean and apparent
solar time historically were derived from nighttime sidereal
observations. For one thing it is much easier to determine the time
of transit of a point-like star against a black sky rather than
peering at the half degree wide disk of the Sun through the daytime
glare. The fundamental reference these days is a grid of
extragalactic sources quite thoroughly fixed to the celestial sphere.

However, the question is irrelevant. One is willing to entertain the
notion of many different choices for a civil time scale. The first
question is: one standard or several or many? I suspect even folks
on the extreme opposite sides of the leap second question would today
prefer a single world wide civil time standard. Originally, of
course, there were as many civil time standards as train stations, etc.

So, imagine any single choice for international civil time. Let's
pick some option that nobody has yet championed. Some timekeeper
exhibiting periodic regularity that will be divided into 24 big hours
of 60 big minutes of 60 seconds too big to be confused with SI
seconds. How about the Lunar month? Forget about the day entirely -
just like the ITU proposal suggests. On the other hand, there is a
direct (well, rather indirect really) connection between the month
and the solar day, so the astronomers will be happy - right? So - what?

So, we've invested good solid effort in reaching a consensus before
implementing a new civil time standard, and the new standard has some
connection to some observable natural phenomenon - and that is really
all we need to know external to the civil time interface. The
details of how the standard is implemented and promulgated are opaque
outside the interface. In fact, part of the friction in all of our
discussions has been precisely that we are too willing to violate
both the TAI and UTC interfaces. One crowd knows too much about the
internals of the astronomically derived interface - and the other
crowd knows way too much about the internals of the atomic time

This special knowledge isn't necessarily helpful in deciding the
fundamental question: What international civil time scale is most
appropriate for the wide range of human activities and communities?
This is a question for use cases and requirements and risk analyses
and economic impact statements and international conferences and...

Personally, I am indeed willing to entertain all sorts of notions
about what makes an appropriate civil time scale. I suspect most of
my buddies from the astronomical software community are similarly
open minded. It has appeared since day one of this debate, however,
that exactly one option has ever been entertained by the ITU. And
whatever options ever do reach the table for discussion, it matters
greatly what process is followed to choose between those options.

Rob Seaman
National Optical Astronomy Observatory
Received on Wed Aug 10 2005 - 23:34:11 PDT

This archive was generated by hypermail 2.3.0 : Sat Sep 04 2010 - 09:44:55 PDT