NavList:

   

Reply

Another recent contribution of interest from the LEAPSECS list:


-------- Forwarded Message --------
Subject: 	[LEAPSECS] current / future state of UT1 access?
Date: 	Fri, 16 Mar 2018 10:18:11 -0700
From: 	Rob Seaman 
Reply-To: 	Leap Second Discussion List 
To: 	leapsecs@leapsecond.com


Regarding Demetrios's response to Steve: did astronomers give advice
divergent to what CCIR decided?It isn't obvious how the history of this
rather typical, if somewhat esoteric, technical debate amounts to
"strong emotional bias". I reject the implication that technical
disagreements, at that time or in the current day, are nothing more than
emotion.

It is perhaps telling that Demetrios doesn't address Steve's central
assertion that it was recognized early on that two kinds of time were
needed. Two kinds of time are still needed.

Meanwhile, over the past week or two I have not been able to connect to
NIST's UT1 server:

https://www.nist.gov/pml/time-and-frequency-division/time-services/ut1-ntp-time-dissemination

Judah Levine has been very helpful in looking at things on the NIST
side, and Harlan Stenn and Martin Burnicki have been insightful
regarding NTP. Needless to say, this level of expert customer support is
atypical! There is no reason to believe that NIST cannot serve UT1, and
certainly such a brain trust (absent my fumble-fingers) could oil the
gears and get the NTP clock spinning like a top. However, that is not
currently the case from my campus. (We do see two other NTP servers in
the same rack as that NIST UT1 server.)

I'm preparing a paper on pragmatic timekeeping for an upcoming
observatory operations conference and would welcome comments from
anybody who has been using the NIST's UT1 service or any other UT1
service or related internet tables. And those who maintain NTP instances
might try connecting to 128.138.140.50 at the University of Colorado in
Boulder and report on their results.

For many purposes (even in astronomy) UTC currently serves as a proxy
for UT1 and for mean solar time in general. If leap seconds cease that
approximation will no longer be functional and the engineering
requirements and infrastructure for delivering UT1 will be stressed
manyfold. How well does it work now? What investment would be needed to
make it reliably scalable? Absent emotion, what are the proposed best
practices for providing access to the multiple timescales needed?

Many thanks!

Rob Seaman
Lunar and Planetary Laboratory
University of Arizona

   

Reply