[LEAPSECS] Windows Server 2019

Steve Allen sla at ucolick.org
Fri Jul 20 21:08:56 EDT 2018

On Fri 2018-07-20T14:23:38+0000 Nero Imhard hath writ:
> Leap second smearing by an NTP server effectively means that it uses
> a different time. Without a way to indicate, within the protocol,
> what time a server uses (and there isn't because UTC is implicit) this
> seems a really bad idea.
> On another level, I regard it as some sort of "technical betrayal"
> which erodes trust in protocols and standards, rendering them less
> useful than they could be.
> Isolate your workarounds and don't let them pollute your environment.

In the arena of leap seconds, technical betrayal and dilution of the
protocols and standards began before the inception date.

Gernot Winkler of USNO is credited as one of the people who suggested
that the leap second could be the solution to the problem.  Gernot
Winkler reported that radio broadcast time signals operated by the
USNO would follow TAI and not implement leap seconds, while at the
same time the USNO Nautical Almanac tabulated events in Ephemeris Time
and UT1.  This is clearly the USNO making use of the "two time scales"
scheme that since 1948 astronomers had been saying would be necessary,
and not following international protocols and standards.
Gernot Winkler was co-author of the 1970 report to the IAU that said
leap seconds would cause problems for automated navigation systems.
At the 1970 IAU Comm 31 meeting Gernot Winkler was acting president
where in the transcript it is clear that at least one paragraph has
been omitted about the response in the IAU meeting when the CCIR
decision to implement leap seconds was announced.
At that same meeting IAU Comm 31 was led to yield that they had no
influence over the leap seconds that the CCIR had instituted, and IAU
Comm 31 was pressed to produce a statement declaring that leap seconds
were "the optimum solution."

All of the above strike me as "something is seriously wrong here."

Looking deeper into the history and memoirs by folks who were involved
it becomes clear that the inception of leap seconds was the
culmination of a 20 year game of international regulatory and
scientific agency pinball.  After the CCIR introduced them that game
continued for another 10 years as other agencies and governments were
led to approve the notion of UTC with leap seconds using words like
"parfaitment recommandable".

The game started when the IAU produced a recommendation for radio
broadcast time signals that was not implementable within available
resources.  Another agency produced a recommendation based on criteria
which were relevant to its particular mission.  At subsequent meetings
of other agencies the agreements made by previous agencies were
represented as the basis for further recommendations.  One of the
agreements used in this game was achieved not at an international
meeting, but in the living room of one of the delegates.  This process
continued until CCIR was led to approve leap seconds without any
accompanying documentation about the technical details of how they
would be implemented.

Several documents indicate that the preparatory meetings prior to the
international assemblies discussed that changes to national laws would
be required in order to broadcast a purely atomic time scale.
I have found nothing that directly explains why it was repeatedly
deemed impossible for any of these agencies to explain and recommend
the existence of two kinds of time scales, but it seems clear that
the legal considerations led toward the notion of a compromise.
I surmise that the folks who wanted to broadcast atomic frequency were
in too much of a hurry to wait for a critical mass of nations to
legislate and allow their national metrology agencies to recognize and
broadcast purely atomic time, so they took a shortcut.
Taking that shortcut meant going to the CCIR with a proposal that did
not explain the consequences and mechanisms, neither to the delegates
who voted nor to any implementors of systems using UTC.

I am not sure whether these kinds of interaction among international
regulatory and scientific agencies are perfectly normal business as
usual, or an example of systematic abuse spanning over decades.  I am
sure that the result was dysfunctional operational systems (which
still lack solid technical underpinnings) and a trail of technical
folks who had withdrawn from the arena to have their agencies continue
using time scales that were not UTC with leap seconds.  I am also sure
that after an ensemble of international agencies have been persuaded
to approve something it becomes very difficult to unapprove it even if
its initial approval was achieved under dubious circumstances with
serious unaddressed technical questions.

So we have betrayal, eroded trust, and reduced usefulness because some
folks wanted to take what looked like a politically expedient shortcut
which was full of unexplained technical complexities.  It is not clear
who can remedy things.

Steve Allen                    <sla at ucolick.org>              WGS-84 (GPS)
UCO/Lick Observatory--ISB 260  Natural Sciences II, Room 165  Lat  +36.99855
1156 High Street               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