[LEAPSECS] it's WP7A week in Geneva

M. Warner Losh imp at bsdimp.com
Thu Sep 10 11:21:40 EDT 2009


In message: <DE4E112B-D8DA-4E92-9458-EA89C8DBC124 at noao.edu>
Rob Seaman <seaman at noao.edu> writes:

: M. Warner Losh wrote:

:

: > There are a number of solutions to the current leap-seconds problems

: > that don't completely decouple UTC from the sun.

:

: As well as non-solutions like leap-hours that don't actually eliminate

: leap-seconds, but rather accumulate them for later release in one

: colossally indigestible lump.


Actually, that solution does keep UTC coupled to the sun. I tend to
agree that large adjustments like that would never happen. But any
insertion of time into the time scale would


: > There's some that do in the recognition that UTC really is going to

: > be viable at most a few hundred to a few thousand years anyway due

: > to the quadratic acceleration of leap second timing.

:

: The quadratic acceleration will apply to any proposed solution

: relative to atomic timescales where the second is of a fixed length.

: This issue is not specific to the current UTC. In particular, it is

: not addressed at all by the proposal to embargo leap-seconds.


Actually, it *IS* addressed by the embargo on leap seconds. The idea
there would be to publish the UT1 and not worry that it grows without
bound. Then you don't have to coordinate the insertion of leap
seconds, just publish a measurement that says what the delta is.

What isn't addressed is keeping them in sync. They are two different
things, and we shouldn't lose sight of that fact. And embargo on
leap seconds forever is exactly the same thing as saying UTC is no
longer tied to the sun. That's a change that should be considered,
even if it ultimately proves to be not a viable change.

This is a proposed solution, and one that may have merit. However,
only if the users of time for which it actually matters that DUT1 is <
1s can get the larger corrections from somewhere else, retool, etc.
These costs may be smaller than continued insertions of leap seconds
(they could very well be larger too: nobody has done a public,
comprehensive study here).


: > Anyway, as with all engineering issues, the practical problems

: > should be discussed as widely as possible, and the real requirements

: > for the system should be continually reevaluated to ensure that the

: > system is meeting the real needs of its users.

:

: Amen!

:

: And with engineering best practices, no single power bloc should seek

: to force adoption of their preferred option. We would have gotten

: much more done (and I would have written many fewer messages to

: everyone's benefit :-) over the past 10 years if there weren't a need

: for constant vigilance to fend off incessant attempts to hurry a

: decision.


Correct. Unfortunately, the power block that was in control in the
late 60's and early 70's has shifted now so that the relative
importance of each of the factions has changed. The importance for
various factors going into the leap-second decision has changed. The
biggest one is now almost all terrestrial navigation is done with GPS
and the DUT1 error correction no longer matters since people don't do
things by hand anymore. Yet the fundamental solution hasn't. If
there had been no action, no leap seconds, etc invented to harmonize
the atomic and civil time scales (this would go back into the late
50's), then we'd come up with a different solution today because the
playing field is so radically different. Some say different enough to
pay the penalty of disturbing the status quo, others are not so sure
so there's friction between the different groups. This tends to lead
to preservation of the status quo. The current stalemate likely won't
change until a plane crash can be tied to leap seconds...

Warner


More information about the LEAPSECS mailing list