[LEAPSECS] LEAPSECS Digest, Vol 45, Issue 3

Steve Allen sla at ucolick.org
Fri Sep 3 00:22:31 EDT 2010

On Thu 2010-09-02T20:21:41 -0600, M. Warner Losh hath writ:

> And the on-time marks for the signals are UTC, with a resolution of

> microseconds. The resolution of the DUT1 data is only 100ms. This

> sounds to me like the official time is UTC, and that, oh, by the way,

> there's this extra parameter that kinda sorts relates to something

> else.

The current ITU-R TF.460 only requires accuracy of 1 ms in the
The current Circular T shows that Norway and Hungary are off
by 38 us, and China is off by 11 us.

> Also, this value is not disseminated by ntp or any other means that

> reports the official time.

That puts us back to pondering what the judges would say for a test
case between "official" and "legal". Given that a US judge just said
any item sold with an expired patent number on it is liable for a fine
up to $500 each I wouldn't expect to predict the answer.

> I'd have to say this argument is a stretch too. As Tony points out

> "traditional" GMT isn't UT1 nor UT2...

As another way of looking at that image I posted from the 1968
BIH Annual Report I suggest looking at the plot here
The coordinate origin of the pole position in the ITRF is supposed to
correspond to the old Conventional International Origin, which was
defined at the location of the pole at 1903.0. It is plain that the
location of the mean 1903 pole in the current system is notably
displaced from the origin. This is the spatial component of the
offsets caused by the adoption of new techniques and conventions
for computing earth orientation.

Every change in technology and conventions causes a drift of origin.
Janssen pointed this out at the 1884 IMC, and Simon Newcomb agreed
with him up to the point where the Department of State had him booted
out of the meeting for agreeing with the Frenchman.

The notions for UT1 and UT2 were required at the 1955 IAU meeting,
and the names were coined just in time to be used as of 1956.
The Greenwich astronomers moved to Herstmonceux in 1957, so the
meridian circle was not operational at the point when the first
cesium-based atomic time scales were coming into use.
The epoch at which the atomic time scales were syncronized to UT2
was 1958-01-01, but that epoch was not decided until mid 1959, and
the values of UT2 differed notably from one observatory to another.
In 1963 the CCIR rewrote Recommendation 374-1 to require that
broadcasts of time should be "within about 100 ms of UT2".
In 1970 the CCIR replaced Rec 374 with leap seconds in Rec 460;
they would have asserted that the new UTC should follow UT2 if there
had not been an astronomer present to point out that navigators
actually need UT1; and they did assert that the maximum deviation
from UT1 should be 0.7 s without bothering to ascertain that then
current techniques could not guarantee better than 0.9 s.

So all through this excitement there were 3 changes in the BIH
conventions along with the beginning of satellite tracking and VLBI
that ultimately resulted in globally-self-consistent, geocentric
geodesy, and there was nothing at Greenwich to say what GMT was.

Until 1972 the only people who had ever heard the term UTC or knew
that it might be used were those who worked in the time bureaus.
Until 1974 as far as the public in the US knew, the official agency
providing public time was saying "Greenwich Mean Time".

Steve Allen <sla at ucolick.org> WGS-84 (GPS)
UCO/Lick Observatory Natural Sciences II, Room 165 Lat +36.99855
University of California Voice: +1 831 459 3046 Lng -122.06015
Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m

More information about the LEAPSECS mailing list