[fitsbits] Leap Second push poll

Rob Seaman seaman at noao.edu
Tue Jan 14 19:19:46 EST 2003


Beginning around the fall of 1999, some subset of the precision timing
community began lobbying to change the definition of UTC - in particular,
to halt the issuance of leap seconds.  Their motivations are unclear - but
are certainly not *our* motivations.

In April of 2002, Demetrios Matsakis announced:

    "The ITU/R Special Rapporteur Group met last week in Paris.  This is the
    group whose charter is to advise the International Telecommunications
    Union (ITU) on whether UTC should be redefined.   Their charter is to
    complete their report by this October, but the date may slip a bit
    (see below).  I will forward any formal announcement by the group when
    I receive it, but the following are the main conclusions, more or less:

    1.  The group voted to exclude from consideration all redefinition
    options except freezing the number of leap seconds at some so-far
    unspecified date.  This means they won't try to change the definition
    of the second from 9,192,631,700 periods of a certain hyperfine
    transition of the undisturbed cesium atom, replace leap seconds by
    leap hours, etc."

The #2 bullet from that list was to hold a public meeting to discuss the
situation.  This was finally announced a few weeks back:

    "The SRG has held several coordination and technical exchange
    meetings to generate, analyze and discuss alternative approaches to
    reduce or eliminate the operational impact of the leap second.  The
    work to date has produced within the SRG a consensual opinion that
    the SRG wishes to present and discuss with  representative parties
    of the scientific, standards and telecommunications agencies and
    groups.  For that purpose the SRG is organizing a  Colloquium for
    deliberating recommendations to the ITU-R."

Note that a "consensual" opinion has already been formed by this committee -
but it will apparently not be announced for debate before this Colloquium,
which will be held:

    ITU-R Special Rapporteur Group (SRG)
    Colloquium on the UTC Time Scale
    Torino, Italy, 28-30 May 2003

Bullet #2a was to issue a more directly targeted survey.  That survey
is appended.  Note that we're not invited to participate.

UTC is an important issue in astronomy - and thus to FITS.  If we don't
develop a common viewpoint on this issue now, it is unlikely anyone will
listen to us later.  If you have a strong viewpoint - or even a slight
interest - in the UTC standard, please consider joining the LEAPSECS
mailing list (address below).

As Matsakis says, it might also be prudent to consider the survey questions
from a standpoint of any projects we are involved in.  In particular, much
astronomical software is likely to require modification to handle DUT>0.9s.

But more fundamentally, what are the implications of decoupling civil
time from the rotating Earth?  And shouldn't the UTC standard be robust
indefinitely far into the future?

Rob Seaman
NOAO

---

> From owner-leapsecs at ROM.USNO.NAVY.MIL Tue Jan 14 16:37:46 2003
Reply-To: Leap Seconds Issues <LEAPSECS at ROM.USNO.NAVY.MIL>
From: matsakis.demetrios at USNO.NAVY.MIL
Subject: [LEAPSECS] Draft Questionnaire
To: LEAPSECS at ROM.USNO.NAVY.MIL

As many of you know, I am the Chair of a Working Group (WG) on UTC for the
International Union of Radio Scientists (URSI).  Several years ago I
distributed a questionnaire as the chair of a similar committee of URSI
Commission J.

My WG is considering sending the following questionnaire to URSI-members.
The intent is to find out what the true costs would be if UTC is redefined
so that there are no new leap seconds after N years, and what the costs
would be of continuing with the same UTC definition we have now.  I would be
interested in your comments, given the scope of the questionnaire and my
limited charter.

To my knowledge, very few of you are URSI members and therefore you will not
be asked this question by my WG.  However I predict that any information
along the lines of the questionnaire would be received with interest by the
points of contact in any of the other relevant international bodies to which
you may belong.

Demetrios Matsakis

****************************************************************************
********
1.      Name and Position

2.      Contact information

3.      URSI Commissions of which you are a member

4.      If it were decided to change the definition of UTC so that no leap
seconds would be inserted after a specified date, 5 years in the future,
would any extra effort be required to adjust any system you work on?

5.      If your answer to question 4 is yes, for each system affected please
provide the information in 5a through 5h.   If these estimates are be
difficult to formulate, you may wish to indicate the range of answers.
Please feel free to contact a member of the Working Group to discuss the
level of detail to provide so as to best help us represent your needs to
URSI.

a. Name of system

b. Brief description of system

c. Hours of extra labor that would be required and for what general purpose,
such as a software review.

d.      Extra equipment that must be purchased, and approximate cost to
purchase

e.      Extra equipment that must be developed, and approximate cost.

f.      Installation cost of extra equipment

g.      Risks involved in modifying the system.

h.      Costs in terms of system performance or final product once the
adjustments are correctly made

6.  If the decision were made to insert no new leap seconds as of today,
please indicate which of your responses to questions and 4 and 5 would be
different.

7. Is there an implementation date that would significantly decrease the
costs indicated in your response to question 5?  If so, please provide the
data and associated costs.

8.      Please provide the approximate cost to your systems of incorporating
the next leap second, should one be called for in the year 2003.

a.      Hours of labor
b.      Equipment purchase
c.      Probability that leap second will not be correctly adjusted for
d.      Costs in terms of final product or system performance if the leap
second is not correctly included
e.      Costs in terms of final product or system performance even if the
leap second is correctly allowed for.

9. Please use this space to make any comments or provide any information you
feel appropriate.

10. It is possible that we would like to publicly identify your system(s) as
relevant to the decision.  In that case, do we have your permission to fully
quote your reply?



More information about the fitsbits mailing list