Re: UT1 confidence

2007-01-18 Thread Rob Seaman

Steve Allen wrote:

The plots by Arias indicate how well UT1 could have been predicted
over two and three year intervals for the 40 year interval starting
around 1960.  It is based on those plots that I have voiced no
concerns for the pointing of our telescopes if leap seconds were
published five years in advance.  I'm not ready to go for ten.

It appears to this observer also that a consensus for extending leap
second scheduling from a six month interval to a five year interval
should be straightforward to achieve.  This is a factor of ten
improvement, could be later lengthened as the state of the art
allows, is likely already supported under the standard, and might in
practice require no coarsening of the 0.9s maximum tolerance -
especially if the five year lookahead were combined with the freedom
to schedule each intervening leap second at the end of any month.

Whether this would be the consensus - or whether some other
scheduling algorithm - the first step would be to take the divisive
and dangerously naive leap hour proposal off the table.  It is
exhausting and distracting to have to continually fend off this rabid
turkey.  The wisdom of taking the time and making the effort to form
a robust consensus BEFORE changing civil timekeeping policies cannot
be overestimated.

I would, however, like to better understand what Arias means by
empirical linear prediction.  This appears not to rely on any
physical model of the rotating Earth.  Surely the plots referenced
should be taken as worst cases?  One would also want to combine these
predictions with the details of specific scheduling algorithms to
start to understand the trade-offs and what the practical limits
would be on the range of DUT1.


The Martian Chronicles

2007-01-15 Thread Rob Seaman

On Fri 2007-01-12T18:35:55 +, Tony Finch hath writ:

According to the slides linked from Dave Mills's Timekeeping in
the Interplanetary Internet page, they are planning to sync Mars
time to UTC.

There seems to be an unwarranted assumption that a spacecraft always
knows its own position.  I also wonder whether it might be productive
to consider closing the NTP servo loop in velocity (frequency) in
this case, rather than position (phase).

Regarding the choice of UTC - it sounds like this project is
ongoing.  Dave Mills is nothing if not responsive to NTP issues.  We
might consider raising the issue with him.

On Jan 12, 2007, at 12:26 PM, Steve Allen wrote:

From an engineering standpoint a variation of 2 ms in a year on
Mars is certainly better than any time scale that could be
established there in lieu of landing a cesium chronometer.

Any comments on the practicality of space-rating such timepieces?
Power source, radiation hardening, hand-off maintenance procedures, etc?

Also - what are the actual use cases requiring a common time scale,
rather than establishing a separate Martian civil cesium standard and
simply tracking the deltas?  I suspect we're all bemused to
contemplate issue terrestrial leap seconds on Martian bases.  How
does the LOD vary on Mars?  No significant moon - but then, leap
seconds are needed on Earth even if the tidal slowing is detrended.

Can something as naive as POSIX time_t really serve all such
applications, even the ones on earth, for the next 600 years?

One would argue that deprecating leap seconds will make the naive
POSIX model even less acceptable.  Some applications (and perhaps
we'll be surprised to learn how many) most certainly do require an
approximation to actual UT, not just to I'm going to lie and call it
UTC.  With leap seconds, this approximation is maintained within the
civil time standard and POSIX doesn't need to worry about it.

Without leap seconds, this approximation is maintained through an
explicit application of a DUT1 correction - a correction that would
start to grow without meaningful bound.  POSIX exists to support
applications of interest to users.  The existence of significant
numbers and classes of users requiring knowledge of actual UT (mean
solar time) would suggest that POSIX would, in turn, need to start
accommodating DUT1.

Rather than defining a perceived problem out of existence, one could
actually find that a real problem has been defined into existence.
Or perhaps I'm wrong.  Demonstrate why.

Rob Seaman

Re: Introduction of long term scheduling

2007-01-07 Thread Rob Seaman

Warner Losh wrote:

Actually, every IP does not have a 1's complement checksum.  Sure,
there is a trivial one that covers the 20 bytes of header, but that's
it.  Most hardware these days off loads checksumming to the hardware
anyway to increase the throughput.  Maybe you are thinking of TCP or
UDP :-).  Often, the packets are copied and therefore in the cache, so
the addition operations are very cheap.

Ok.  I simplified.  There are several layers of checksums.  I
designed an ASCII encoded checksum for the astronomical FITS format
and should not have been so sloppy.  They do it in hardware could
be taken as an argument for how time should be handled, as well.

Adding or subtracting two of them is relatively easy.

Duly stipulated, your honor.

Converting to a broken down format or doing math
with the complicated forms is much more code intensive.

And should the kernel be expected to handle complicated forms of
any data structure?

Dealing with broken down forms, and all the special cases usually
multiplcation and division, when tend to be more computationally
expensive than the checksum.

Indeed.  May well be.  I would suggest that the natural scope of this
discussion is the intrinsic requirements placed on the kernel, just
as it should be the intrinsic requirements of the properly traceable
distribution and appropriate usage of time-of-day and interval
times.  Current kernels (and other compute layers, services and
facilities) don't appear to implement a coherent model of
timekeeping.  Deprecating leap seconds is not a strategy for make the
model more coherent, rather, just the timekeeping equivalent of

Having actually participated in the benchmarks that showed the effects
of inefficient timekeeping, I can say that they have a measurable
effect.  I'll try to find references that the benchmarks generated.

With zero irony intended, that would be thoroughly refreshing.

If by some limp attempt you mean returns the correct time then you
are correct.

It's not the correct time under the current standard if the
timekeeping model doesn't implement leap seconds correctly.  I don't
think this is an impossible expectation, see http://, starting with the Levine and
Mills PTTI paper.

You'd think that, but you have to test to see if something was
pending.  And the code actually does that.

Does such testing involve the complex arithmetic you describe above?
(Not a rhetorical question.)  The kernel does a heck of a lot of
conditional comparisons every second.

Did I say anything about eviscerating mean solar time?

Well, these side discussions get a little messy.  The leap second
assassins haven't made any particular fuss about kernel computing
issues, either, just previous and next generation global positioning
and certain spread spectrum applications and the inchoate fear of
airplanes falling from the sky.

The probability of the latter occurring seems likely to increase a
few years after leap seconds are finally eradicated - after all,
airplanes follow great circles and might actually care to know the
orientation of the planet.  Hopefully, should such a change occur
courtesy of WP7A, all pilots, all airlines and all air traffic
control centers will get the memo and not make any sign errors in
implementing contingent patches.  It's the height of hubris to simply
assume all the problems vanish with those dastardly leap seconds.  (I
don't suppose the kernel currently has to perform spherical trig?)

Note that the noisy astronomer types on this list are all also
software types, we won't reject computing issues out of hand.

I'm just suggesting that some of the suggested ideas have real
performance issues that means they wouldn't even be considered as
viable options.

Real performance issues will be compelling evidence to all parties.
Real performance issues can be described with real data.

True, but timekeeping is one of those areas of the kernel that extra
overhead is called so many times that making it more complex hurts a
lot more than you'd naively think.

Either the overhead in question is intrinsic to the reality of
timekeeping - or it is not.  In the latter case, one might expect
that we could all agree that the kernel(s) in question are at fault,
or that POSIX is at fault.  I have little sympathy for the suggestion
that having established that POSIX or vendors are at fault that we
let them get away with it anyway.  Rather, workaround any limitations
in the mean time and redesign properly for the future.

If, however, the overhead is simply the cost of doing timekeeping
right, then I submit that it is better to do timekeeping right than
to do it wrong.  Doing it right certainly may involve appropriate
approximations.  Destroying mean solar time based civil time-of-day
is not appropriate.

Of course, we have yet to establish the extent of any problem with
such overhead.  It sounds like you have expertise in this area.
Assemble your 

Re: Introduction of long term scheduling

2007-01-06 Thread Rob Seaman

Warner Losh wrote:

leap seconds break that rule if one does things in UTC such that
the naive math just works

All civil timekeeping, and most precision timekeeping, requires only
pretty naive math.  Whatever the problem is - or is not - with leap
seconds, it isn't the arithmetic involved.  Take a look a [EMAIL PROTECTED]
and other BOINC projects.  Modern computers have firepower to burn in
fluff like live 3-D screensavers.  POSIX time handling just sucks for
no good reason.  Other system interfaces successfully implement
significantly more stringent facilities.

Expecting to be able to naively subtract timestamps to compute an
accurate interval reminds me of expecting to be able to naively stuff
pointers into integer datatypes and have nothing ever go wrong.  A
strongly typed language might even overload the subtraction of UTC
typed variables with the correct time-of-day to interval
calculations.  But then, what should one expect the subtraction of
Earth orientation values to return but some sort of angle, not an


Re: Introduction of long term scheduling

2007-01-06 Thread Rob Seaman

Warner Losh wrote:

Anything that makes the math
harder (more computationally expensive) can have huge effects on
performance in these areas.  That's because the math is done so often
that any little change causes big headaches.

Every IP packet has a 1's complement checksum.  (That not all
switches handle these properly is a different issue.)  Calculating a
checksum is about as expensive (or more so) than subtracting
timestamps the right way.  I have a hard time believing that epoch-
interval conversions have to be performed more often than IP
packets are assembled.  One imagines (would love to be pointed to
actual literature regarding such issues) that most computer time
handling devolves to requirements for relative intervals and epochs,
not to stepping outside to any external clock at all.  Certainly the
hardware clocking of signals is an issue entirely separate from what
we've been discussing as timekeeping and traceability.  (And note
that astronomers face much more rigorous requirements in a number of
ways when clocking out their CCDs.)

Well, the kernel doesn't expect to be able to do that.  Internally,
all the FreeBSD kernel does is time based on a monotonically
increasing second count since boot.  When time is returned, it is
adjusted to the right wall time.

Well, no - the point is that only some limp attempt is made to adjust
to the right time.

The kernel only worries about leap
seconds when time is incremented, since the ntpd portion in the kernel
needs to return special things during the leap second.  If there were
no leapseconds, then even that computation could be eliminated.  One
might think that one could 'defer' this work to gettimeofday and
friends, but that turns out to not be possible (or at least it is much
more inefficient to do it there).

One might imagine that an interface could be devised that would only
carry the burden for a leap second when a leap second is actually
pending.  Then it could be handled like any other rare phenomenon
that has to be dealt with correctly - like context switching or

Really, it is a lot more complicated than just the 'simple' case
you've latched onto.

Ditto for Earth orientation and its relation to civil timekeeping.
I'm happy to admit that getting it right at the CPU level is
complex.  Shouldn't we be focusing on that, rather than on
eviscerating mean solar time?  In general, either side here would
have a better chance of convincing the other if actual proposals,
planning, research, requirements, and so forth, were discussed.  The
only proposal on the table - and the only one I spend every single
message trying to shoot down - is the absolutely ridiculous leap hour
proposal.  We're not defending leap seconds per se - we're defending
mean solar time.

A proposal to actually address the intrinsic complications of
timekeeping is more likely to be received warmly than is a kludge or
partial workaround.  I suspect it would be a lot more fun, too.

Kernels aren't written in these languages.  To base one's arugments
about what the right type for time is that is predicated on these
langauges is a non-starter.

No, but the kernels can implement support for these types and the
applications can code to them in whatever language.  Again - there is
a hell of a lot more complicated stuff going on under the hood than
what would be required to implement a proper model of timekeeping.


Re: Introduction of long term scheduling

2007-01-05 Thread Rob Seaman

Tony Finch wrote:

you need to be able to manipulate representations of times other
than the present, so you need a full leap second table.

Which raises the question of how concisely one can express a leap
second table.  Leap second tables are simply a list of dates - in ISO
8601 or MJD formats, for example.  Additionally you need an
expiration date.  An ISO string is really overkill, MJD can fit into
an unsigned short for the next few decades - but this is really more
than you need for the current standard since not all MJDs are
permitted, only once per month.  Also, we don't need to express leap
seconds that are already known (or never existed), so there is a
useless bias of ~54000 days.  If we start counting months now, a
short integer will suffice to encode each leap second for the next
5000+ years - certainly past the point when monthly scheduling will
no longer suffice.

So, let's see - assume:

   1) all 20th century leap seconds can be statically linked
   2) start counting months at 2000-01-31

We're seeing about 7 leapseconds per decade on average, round up to
10 to allow for a few decades worth of quadratic acceleration (less
important for the next couple of centuries than geophysical noise).
So 100 short integers should suffice for the next century and a
kilobyte likely for the next 500 years.  Add one short for the
expiration date, and a zero short word for an end of record stopper
and distribute it as a variable length record - quite terse for the
next few decades.  The current table would be six bytes (suggest
network byte order):

   0042 003C 

A particular application only needs to read the first few entries it
doesn't already have cached - scan backwards through the list just
until you pass the previous expiration date.  Could elaborate with a
checksum, certificate based signature or other provenance - but these
apply whatever the representation.

To emphasize a recent point:  DUT1 is currently negligible for many
applications.  Which is the same thing as saying that the simple
table of quantized leap seconds is quite sufficient for civil
purposes.  The effect of the ALHP is to inflate the importance of
DUT1 - not just for professional purposes, but for some list of
civil purposes that have yet to be inventoried, e.g., tide tables,
weather forecasts, pointing satellite dishes, aligning sundials (see
article in the Jan 2007 Smithsonian), navigation, aviation, amateur
astronomy, whatever.  I'm not arguing here that these are
intrinsically sufficient to justify retaining leap seconds (although
I believe this to be the case).  Rather, I'm arguing that even under
a caves of steel scenario of Homo sapiens inter-breeding with
Condylura cristata, that there will be applications that require a
explicit DUT1 correction - applications that currently can ignore
this step since UTC is guaranteed to remain within 0.9s of GMT.

So the current requirement is merely to convey a few extra bytes of
state with a six month update cadence.  This suffices to tie civil
epochs (and a useful approximation of Earth orientation) to civil

The requirement in the post-leap-second Mad Max future, however,
would be to convey some similar data structure representing a table
of DUT1 tie points accurate to some level of precision with some as-
yet-unspecified cadencing requirement.  The most natural way to
express this might be the nearest round month to when each integral
step in DUT1 occurs, but it should be clear that the requirement for
maintaining and conveying a table of leap seconds is not eliminated,
but rather transmogrified into a similar requirement to maintain and
convey a table of DUT1 values.

Rob Seaman

Re: Introduction of long term scheduling

2007-01-05 Thread Rob Seaman

Ashley Yakeley wrote:

As the author of a library that consumes leap-second tables, my ideal
format would look something like this: a text file with first line
for MJD of expiration date, and each subsequent line with the MJD of
the start of the offset period, a tab, and then the UTC-TAI seconds

As an author (and good gawd, an editor) of an XML standard and schema
to convey transient astronomical event alerts - including potentially
leap seconds - I'd have to presume that XML would do the trick.

The thread was a discussion of appending enough context to an
individual timestamp to avoid the need for providing historical leap
seconds table updates at all.  Someone else pointed out that this
didn't preserve the historical record.  I wanted to additionally
point out that the cost of appending the entire leap second table to
every timestamp would itself remain quite minimal for many years, and
further, that even getting rid of leap seconds doesn't remove the
requirement for conveying information equivalent to this table (on
some cadence to some precision).

The complications are inherent in the distinction between time-of-day
(Earth orientation) and interval time.  The intrinsic cost of
properly supporting both types of time is quite minimal.


Re: how to reset a clock

2007-01-04 Thread Rob Seaman

Peter Bunclark wrote:

Indeed isn't this Rob's ship's chronometer?

Actually, I think it was Mr. Harrison's.  (And Steve Allen has been
basing his arguments more recently on this distinction.)  This
healthy debate between astronomical time and clock time has happened
before.  The answer is the same as before - both types of time are
needed.  (Some things never change.)  I'm sure Pete is more familiar
with this story than I am, but others may not be.

Harrison attempted to build a perfect clock to win the Longitude
Prize.  Folks who haven't read Sobel's book should do so - my
classmates at Villanova and I learned the story from an Augustinian
priest who appeared old enough to have known Harrison personnally.
Harrison's first glorious shipboard clock failed to take the prize
due to a lack of compensation for centrifugal effects on a sailing
vessel that must tack when sailing against the wind (or must wear
through an even larger angle, bringing the wind across its stern).
Compensation was needed for relativistic effects, if Newtonian rather
than Einstein.  (Some things never change.)

Harrison invented or improved a variety of familiar mechanical
doodads like the roller bearing and bimetallic temperature
compensation.  He likely could have succeeded in solving this
particular problem, but there would always have been another physical
improvement needed.  (Some things never change.)  Each improvement
would have made the clock more complicated and eventually too fragile
to possibly work on a constantly moving platform buffeted and often
bathed by the salty sea.

He created a second clock and was working on a third round of
improvements when the idea we're discussing first occurred to him.
He had been using a pocket watch as a mechanism to transfer time from
stationary standard clocks (many built by himself) to his portable
prototypes.  He would reset the clock in one place and physically
carry it to where the time was needed.  If a roundtrip correction
were needed, presumably he would note the time on either end and
halve the difference.  This is the standard synchrony or
conventionality of simultaneity of special relativity - familiar to
anyone who has looked under the hood of NTP.  (Some things never

What Harrison recognized was that he didn't need to build a perfect
clock - he merely needed to quantify and log the error inherent in
the clock.  By replacing a large and finicky better clock, with a
small and robust, but more even-tempered, one, the rate of the clock
could be regularized and its random and systematic errors could be
minimized.  That the rate of the clock was now guaranteed not to
match the rate of the spinning Earth was no longer a bug, but a
feature.  By carefully calibrating the clock rate before leaving on a
voyage, and checking it against astronomical observations throughout
the voyage, it was possible to compute the mean solar time at the
home port.  (Some things never change.)  Comparison with the local
time, measured by sextant, then recovered the longitude directly.

And, of course, a ship would not carry a single clock, but two or
more.  Friendly ships meeting at sea would also exchange clock
readings - creating the first ensemble time scale.  (Some things
never change.)

Thus was the chronometer born - and thus did Britannia rule the waves.

The point is that time isn't just an unending count of seconds - it
is the epoch of when the count was zero.  That epoch often has
significance in some periodic natural phenomena, usually related to
Earth orientation.


Re: Introduction of long term scheduling

2007-01-02 Thread Rob Seaman

Daniel R. Tobias replies to Poul-Henning Kamp:

Has anybody calculated how much energy is required to change
the Earths rotation fast enough to make this rule relevant ?

Superman could do it.  Or perhaps he could nudge the Earth's rotation
just enough to make the length of a mean solar day exactly equal
86,400 SI seconds.

Only briefly.  Consider the LOD plots from
~sla/leapsecs/dutc.html.  The Earth wobbles like a top, varying its
speed even if tidal slowing is ignored.

Actually, rather than being merely a troublemaker, the Moon serves to
stabilize the Earth's orientation.  The Rare Earth Hypothesis makes
a strong case that a large Moon and other unlikely processes such as
continental drift are required for multicellular life to evolve, in
addition to the more familiar issues of a high system metal content
and a stable planetary orbit at a distance permitting liquid water.
Without the Moon, the Earth could nod through large angles, lying on
its side or perhaps even rotating retrograde every few million
years.  Try making sense of timekeeping under such circumstances.

Rob Seaman

Re: A lurker surfaces

2007-01-02 Thread Rob Seaman

Magnus Danielson wrote:

If you do want a new timescale, I think rubber seconds isn't going
to be the solution.

One might point out that many time scales do rely on rubbery seconds,
e.g., sidereal time and apparent solar time.  If might be
enlightening to step back from the tendentious and tedious tug-of-war
between UTC and TAI and reflect that even UT1 - a mean solar time
scale - intrinsically has rubber seconds.  Sexagesimal notation is
clearly revealed as a way to express an angle - of Earth orientation
in this case.  The whole point of UTC is to permit Earth orientation
to be approximated while using SI seconds.

Rob Seaman

Re: Introduction of long term scheduling

2007-01-02 Thread Rob Seaman

Poul-Henning Kamp wrote:

That's an interesting piece of data in our endless discussions
about how important DUT1 really is...

The point is that by allowing it to grow without reasonable bound,
DUT1 would gain an importance it never had before.

Happy New Year!

2007-01-01 Thread Rob Seaman

Rather than reply in detail to the points raised in the latest
messages - believe me, you've heard before what I was going to say
again - I'd simply like to wish everybody a happy new year.  I am
grateful to everybody who has ever contributed to this list and
consider it a mark of the importance of civil timekeeping that the
conversation continues.

Since there are new voices on the list, I might simply direct
interested readers to my own thoughts, unchanged at their core in
more than five years:

In short, the current standard has a lot of life left in it.

That said, I have no problem whatsoever with schemes that lengthen
the six month reporting requirements to several years.  Steve's five
year plan, recently quoted again, or the decadal scheduling that has
become something of a standard talking point on the list, are each
already entirely legal under the standard if the 0.9s limit on DUT1
could be extrapolated that far in advance.

Perhaps someone on the inside could comment on the current state of
the art of multiyear predictions?  The most notable feature of leap
second scheduling has been the seven year gap from New Year's Eve
1998 to New Year's Eve 2005.  Otherwise figure 1 from my link above
(and attached below) shows a phenomenological slope close to 7 leap
seconds per decade.  The question is not whether significant
excursions are seen from this general trend - the question is how
well can they be predicted.  Looking at my figure 2 (you'll have to
click through for this one), one will see that a vast improvement in
the state of the art of making short term predictions has occurred
since Spiro Agnew had his the residence at the USNO.

Nobody should be surprised to learn that I will continue resolutely
to oppose the embarrassing and absurd notion of embargoing every 3600
leap seconds into a so-called leap hour.  Why 3600?  Since this
represents an intercalary period - that appears, only to promptly
disappear from the clocks of the world - why not any other random
number of seconds?  How about 1000?  Or 1066, to commemorate the
theft of UTC from its Greenwich origins just like the Normans stole
England from the Saxons who stole it from the Celts?  Some may join
me in thinking 666 might be the appropriate embargo.

Bottom line - nothing about the current standard forbids scheduling
(and reporting, of course) multiple leap seconds several years in
advance.  Obviously it would take at least N years to introduce a new
reporting requirement of N years in advance (well, N years minus six
months).  I suspect it would be exceptionally interesting to
everyone, no matter what their opinion on our tediously familiar
issues, to know how well these next seven or so leap seconds could be
so predicted, scheduled and reported.  If the 0.9s limit were to be
relaxed - how much must that be in practice?  Are we arguing over a
few tenths of a second coarsening of the current standard?  That's a
heck of a lot different than 36,000 tenths.

Rob Seaman

Re: A lurker surfaces

2006-12-31 Thread Rob Seaman

Poul-Henning Kamp wrote:

Rob, If you feel uncomfortable with calling leapseconds
discontinuities, then we can use the term arrhythmia instead.

Which raises the question of why projects requiring an interval time
scale lacking in such arrhythmias would have selected UTC in the
first place.  And why timekeepers who understand these issues would
focus on remediating (i.e., eviscerating) UTC as the cure.
Astronomers are among the power users for interval time as well as
time-of-day.  Helioseismologists ( needed an
interval timescale that would be even tempered over years or even
decades (a solar cycle is eleven years - the magnetic field flips at
solar max, so a complete sample would require 22 years) - so they
selected GPS, not UTC.

But actually, I think we should call leap seconds what they are -
intercalary events.  My wife works at the Arizona-Sonora Desert
Museum.  We have family visiting and decided to spend the day at the
museum - a good way to end a year.  I especially recommend the raptor
free flight program - the ferruginous hawk is especially impressive.
My point is that a javelina is not a pig, a coatimundi is not a
raccoon, and a ringtail cat is not a cat.  A kangaroo rat is, of
course, neither.  And a leap second is a not a discontinuity.
Imprecision in terminology leads to poor decision making.


Re: A lurker surfaces

2006-12-30 Thread Rob Seaman

Jim Palfreyman wrote:

With my time hat on, having time that is discontinuous pains me. It
doesn't make sense in my heart. But at least these
discontinuities are in whole seconds.

Any discontinuities must be regularly done. So they are part of all
computer systems and are tested and used all the time. Don't let
them build for a decade - that is bad bad news.

Just a reminder that UTC has no - none - nada - discontinuities.
Various computer mis-implementations may, but the standard is very
carefully constructed to avoid spring-forward or fall-back gaps or do-

This is just one of many flaws of the notion of leap hours.  A leap
hour (like a leap second or leap day) is an extra intercalary
temporal unit inserted into the continuous flow of time.  A leap hour
is NOT an unmatched fall back do-over - the day in question would
have 25 regular, ordinary, permanent, unique hours - and the extra
hour would occur contemporaneously worldwide.  It would not involve
an easy floating 2 am Sunday local clock reset.

So, for example, if the leap hour is 2606-12-31, 24:00:00 to 24:59:59
UT, it would fall BETWEEN 18:59:59 and 19:00:00 EST, just like a leap
second today is 23:59:60 UT or 18:59:60 EST, also falling between
18:59:59 and 19:00:00.  In each time zone in turn, the leap hour
would fall between different otherwise sequential clock ticks - a non-
issue with a leap second, not so easy to ignore with a leap hour.
Still not a discontinuity, but certainly a real pain for anyone who
is trying to keep the events of the day straight.

More than likely, the hour would be labeled the same worldwide, so
the EST clock would run 18:59:58, 18:59:59, 24:00:00, 24:00:01, ...,
24:59:58, 24:59:59, 19:00:00, 19:00:01, ...


Re: Mechanism to provide tai-utc.dat locally

2006-12-29 Thread Rob Seaman

Tony Finch wrote:

You need to do so in order to implement an accurate clock, since
the clock produces interval time and you need a way to convert its
output to time of day.

As Steve Allen has pointed out, it is in the nature of a clock to be
reset on occasion.  What is NTP but a mechanism for managing these
resets?  What is any clock discipline?  What else does it mean to
maintain traceability?  What is a change in rate, but a reset
schedule carried to a limit?  (Although Steve might replace reset
with maintain a list of offsets.)

I don't disagree that maintaining updated access to a master list of
resets (leaps) or rate changes provides one avenue to implementing
an accurate clock, i.e., synchronizing one clock to another.  But
even today this often is, can be, will be, managed by resetting one
clock as needed, manually if necessary.  And even with a detailed
long range list of leaps in hand, there is still a responsibility to
implement each leap correctly as a 61s or 59s minute.  Otherwise this
6 month or 10 year or 600 year lookahead is no better than having
Harold Lloyd reset your clocks by hand.

One might also point out that the clocks in most PCs are far less
even tempered than Madre Tierra.  I don't suppose anybody has thought
to run the various DUT1 possibilities past the NTP v4 working group?
One could do worse than to adopt the NTP clock discipline writ large
as a baseline leap second scheduling algorithm.  As discussed
recently, there is a natural scale to the tempering of DUT1.  Working
with that, rather than with some arbitrary forget the whole thing
strategy, is likelier to converge on consensus.

And just to stay on message, let me point out that nothing in the so-
called leap hour proposal covers any of these issues.  The only way
to create a new workable consensus on civil timekeeping is to address
the key issues forthrightly.

This is a static offset.

No, it is subject to arbitrary political variations.


However, whimsically redefined is not the same as changing in a
secular fashion.


Design - a Tufte decision

2006-12-28 Thread Rob Seaman

On Dec 27, 2006, at 12:02 AM, M. Warner Losh wrote:

Calculating time intervals for times 6+ months in the future can be
the least of one's worries when one tries to code up a library to deal
gracefully with these different failure modes.  The trivial case where
one has perfect knowledge of TAI-UTC and one can keep that knowledge
current is very simple in comparison.  Dealing with this case is very
simple, and is the way most people think about leap seconds.  But
dealing with the edge cases can be difficult because there are so
many, and so many that people forget to test or conceive of until the
call from the field comes in with a failure...

A lot of these edge cases are really firmly centered in issues of
real-time programming.  Few versions of Unix are equipped to deal
with real-time issues in even a rudimentary fashion.  In any event,
these cases have very little to do with leap seconds or any other
aspects of the representation of time quantities.

That said, I've found the current discussion immensely refreshing.
If there is to be any common ground found between the different
factions on this list (including the lurkers who actually have a vote
on ITU matters), it will be located by focusing on the actual
technical design process, not some quick fix gimmick.

   1) Who are the stakeholders for civil timekeeping?  (A discerning
eye might note that all this time we have NOT been arguing about TT,

   2) What minimal inventory of timescales will satisfy all stakeholders?

   3) What economic, legal, historical, cultural, scientific, etc.,
requirements are placed on the delivery of said timescale(s)?

   4) What technical solution(s) satisfy the requirements?

   5) If change is deemed to be warranted, how best should it be
accomplished?  (Funded, scheduled, designed, implemented, deployed,
maintained, supported, etc?)

My initial position for #2 is that there must be at least two
timescales, representing interval time and time-of-day, but in the
extreme one could even imagine a coherent position stating that NO
common international civil timescale is needed at all.  (Whether one
holds this point of view may say more about ones view of civil
society than it does about civil timekeeping :–)

In the latter case, let UTC continue as it is, complete with leap
seconds.  Let facilities derived from ITU deliberations start
distributing, for example, a refinement of GPS time.  And let some
new consortium adopt the distribution of UTC, perhaps with many of
the improvements folks have suggested.  (This would provide a chance
to do over - wouldn't that be lovely?)  And then let the market

In the U.S., one would expect funding proposals for a new-and-
improved UTC to result.  Those of us who aren't Co-I's would likely
be referees...

Whatever the solution, implementing it is unlikely to prove a
panacea.  In the seven years this list has been in existence, we've
only scratched the surface of the complexities inherent in civil
timekeeping.  In short:

   6) Where should the lines of elegant design and rational compromise
be drawn?

Rob Seaman

Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

Poul-Henning Kamp wrote:

I seriously don't belive you do equality comparisons at the 1msec
level in real world software.  Please provide examples.

You know you're in trouble when PHK and I agree.  One would think a
(double precision) floating point epsilon test might be what you
want.  In those cases that demand some sort of archival query, an ISO
8601 string might be appropriate, but one would typically expect
queries to be issued on a window about the desired timestamp, or
perhaps given a range specification from 10:03:01.933 to
10:03:02.008 (whether a string, integer or floating point - binary
or sexagesimal or BCD for that matter).


Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

John Cowan wrote:

I assume you mean 23-hour or 25-hour LCT days?  True.  It does work
against UCT days, though, since they are uniformly 1440 minutes long.

Not should leap hours replace leap seconds.

Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

I am talking about time intervals; you are talking about periodic
events.  Two different things.


Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

M. Warner Losh wrote:

And avoiding the ugly 61 or 59 second minutes to define away the

It was the time lords who decreed that rubber minutes were prettier
than rubber seconds.  We're now to skip right over rubber hours to
rubber days?  Their aesthetic sense seems strangely malleable.

Problems that are merely defined away rarely stay away.


Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

Poul-Henning Kamp wrote:

It is not an æsthetic issue, it is an issue of practical

Well, it's both.  The big question is practical implementation of what?

In these days of heavily computerized infrastructure, we need more
than half a years warning about discontinuities in the timescale.

We've had this discussion before.  There are no discontinuities in
the timescale.  Leap seconds are a question of data representation.
I'm not trying to minimize the issues by saying that, rather to point
out that we can't solve a problem until we state it correctly.

We can get that only by increasing the DUT tolerance.

We all understand the trade-offs.  Presumably the guys who have
suggested degrading the tolerance to the point that it will outlive
our grandchildren's grandchildren - and simultaneously removed any
requirement for the ITU to distribute DUT corrections - understand
the trade-offs, too.

I don't care if you want to implement leap-milliseconds, as long
as you tell me 10 years in advance when they happen.

Again - with no intent to minimize the issues - what supports this
assertion?  Is there any reason to believe that 10 years advance
notice would encourage projects and vendors to do anything other than
ignore the requirement entirely?  A statement that 10 years, or 600
years, notice is all that is needed to resolve all the problems,
smooth over all the complications, is entirely too glib.

Rather than starting from a bunker mentality of repeatedly fending
off an absurd non-solution, perhaps it would be better to design from
clearly stated use cases, responsive requirements, coherent risk
analyses, a reasonable deployment schedule, a fair-minded budget.
We're not going to successfully define the real world out of existence.

Rob Seaman

Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

M. Warner Losh wrote:

Let's turn the question around.  What would the harm be if |DUT1| were
1.1s?  1.5s?  2.0s?  Contrast this with the harm and difficulty that
the current 6 month scheduling window affords.

Indeed.  Go for it.  I look forward to reading your report.  Who and
what interests are adversely affected in each case?  How are these
effects mitigated as a function of the limit on DUT1?  Also, contrast
what benefits accrue.  One would think that the responsibility for
quantifying the implications of a change to a standard would fall on
the parties proposing said change.


Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman

John Cowan wrote:

It can't possibly be.  Nobody can know what a change is going to
cost except those who are going to have to pay for it (or not
pay for it).

Are you really suggesting that the planning of technical projects is
impossible?  One might expect some investment of time and money in
standard planning activities to be made first, rather than
immediately jumping to a narrowly and rather randomly conceived
notional position unsupported by even the slimmest of white papers.
It is crazily unprofessional to abjure any responsibility for
quantifying costs, benefits and risks.

And even their word cannot necessarily be trusted.

Um.  Which edge of the sword are we talking about here?


Re: Mechanism to provide tai-utc.dat locally

2006-12-28 Thread Rob Seaman
 designed, non-conforming UTC implementations failing to handle
leap seconds - and none with civil time-of-day falling further and
further away from actual mean solar time.  One might rest easier,
however, knowing that any effort whatsoever had been invested in
searching for potential risks on both sides.

so maybe some other means of distribution is
necessary...  And is 100ms really good enough?

Excellent questions.  Might I suggest that they be appropriately
answered before UTC is removed from life support?

Rob Seaman

Re: Mechanism to provide tai-utc.dat locally

2006-12-27 Thread Rob Seaman

Clive D.W. Feather writes:

WG14 is willing in principle to make changes to time_t, up to and
completely replacing it by something else. *BUT* it needs a
complete and
consistent proposal and, preferably, experience with it.

This is at the heart of my distaste for the so-called leap hour
proposal.  There is no coherent proposal, no implementation plan, no
discussion of adverse effects, no budget, no collection of pertinent
use cases, no exploration of requirements - no technical design
discussion at all.

Meanwhile, the astronomical community (like other prudent
communities, presumably) has instituted long range planning efforts
like the U.S. Decadal Surveys.  Major (and minor) telescope and
instrumentation projects involve multi-year design commitments with
significant budgets of their own simply to develop coherent and
complete proposals to submit to the funding agencies.  The U.S.
national centers have recently been subject to an NSF Senior Review
process that is likely to have a major affect on all our operating
budgets to free up funding for new initiatives.  No pain, no gain.

On the other hand, we have a proposal to change the fundamental
underpinnings of world-wide civil timekeeping.  A (publicly
unavailable) proposal that can't even be bothered to suggest how DUT1
will be conveyed in a future in which this quantity would assume
vastly greater importance.  It's an embarrassment.

Any proposal has got to deal with a whole load of issues, many of
haven't been properly documented. For example, it should be
possible to add
and subtract times and intervals (e.g. what time is 14 months and
87 days
from now?).

Poul-Henning Kamp wrote:

... which is exactly the kind of thing you can not do with any
{origin+offset} format, due to leap seconds.

Leap seconds are an effect, not a cause.  The intrinsic difficulty is
with mapping interval time to Earth orientation (mean solar time).
Whatever mechanism is used to synchronize civil time to the sun -
including embargoing leap seconds as leap hours - there will always
be this complication.

Consider an ISO 8601 compliant date/time representation, e.g.:

   % date -u +%FT%TZ

The first part is the (Gregorian) calendar date - the second clearly
represents a fraction of a day.  From my point of view, this is the
beginning and end of the argument establishing an identity between
civil time and mean solar time.  Others are willing to permit a slow
secular drift - in the calendar, too, of course, not just in the
clock.  Mucking with leap seconds is equivalent to redefining the
concept of a day.

The point is that over a long enough period, a broad enough temporal
horizon, we all agree that civil time must be synchronized to solar
time.  The emergence of the absurd leap hour proposal from among
folks who loathe leaps of any sort demonstrates that.  They weren't
eager to center their notional position around leap hours - rather,
they felt obligated.  In short, there is no escaping the need to
grapple with the fundamental distinction between Earth orientation
and atomic interval counts.  Just as, as one enlarges ones spatial
horizon one cannot fail to run into relativistic effects.

Timekeeping is a subtle business.  Others on this list surely
understand that better than I.

Rob Seaman

Re: Mechanism to provide tai-utc.dat locally

2006-12-25 Thread Rob Seaman

Good discussion.

On Mon 2006-12-25T15:42:12 -0500, Keith Winstein hath writ:

Even if a program is able to calculate TAI-UTC for arbitrary points
in the
past and near future, what should a library do when a program asks to
convert between UTC and TAI for some instant further than six
months in
the future?

One might start by compiling some use cases.  Who is asking and why
do they need this information?  The one thing we should be able to
get out of the disagreements on this list over the years is that
different people need timelike quantities for different purposes.
There is no one size fits all answer.

On Dec 25, 2006, at 3:08 PM, Steve Allen wrote:

This is something missing from most systems purporting to have clocks
that was there on almost all 19th century ships.  The navigator has a
chronometer, and the navigator's log has an estimate of the offset and
rate of the chronometer.  Nevertheless, until the ship next sails into
a port near an observatory, there's no way to be sure what time it
actually is.  When the ship gets to port the navigator can go back and
correct the navigational measurements after the fact, and thus
establish better coordinates for the islands and reefs.

Right, although under good conditions they were also able to conduct
lunar observations and observations of Jupiter's satellites from
shipboard that could be used to correct the chronometers in transit.
Obviously a stationary and stable observatory on land is preferred,
but the innumerable interlocking cycles of the solar system provide a
unique check on any clock.

Rob Seaman

Re: Equitable estoppel

2006-12-19 Thread Rob Seaman

On Dec 17, 2006, at 11:48 AM, Poul-Henning Kamp wrote:

Regarding an intenational treaty as a contract is not only pointless,
it is downright silly.

Regarding me as an expert on international law is what would be  
silly :–)

The point was to elaborate, in the context of UTC, on some issues  
raised by Parrish.  I'd be delighted to see real lawyers consulted on  
the real world implications for real civil timekeeping.  Evidently  
this is non-trivial.  To not seek legal counsel prior to instituting  
an historically unprecedented change in the nature of civil time  
would be absurdly foolhardy.

The concept of estoppel is quite general, however, and one might  
expect that it would apply even more to treaties than to contracts.   
Two or more parties negotiate some complicated freeform agreement.   
One of the parties presses for a certain provision.  In good faith,  
the others reluctantly agree.  Simple playground rules (i.e., what  
one might hope to apply at the level of governments :–) suggest that  
those other parties have a veto over future changes to that provision.

But you could conceiveably argue the point, that ITU-R only controls
time, as far as it pertains to telecommunication and radio  

of time signals,

No.  I argue that interval time (TAI, etc.) and time-of-day (civil  
time) are two different things.  This is a point of fact, not of law.

And I argue that while ITU-R has some authority over the UTC standard  
as pertinent to the transmission of time-of-day, it does not over its  
fundamental nature as a flavor of Mean Solar Time.

Time-of-day is what the name says - and is defined by cultural norms,  
statutory and common law and international agreements arising out of  
the Meridian Conference, for instance.  The worldwide civil time  
consensus based on GMT that has proven so durable has far more  
reality than the hazily imagined mistaken best interests of some  
telecommunications multinationals.

And no, I don't think embargoing leap seconds for 600 years  
corresponds to a practical approximation to time-of-day.  It is a  
ridiculous shell game trick.

and that each country is free to use another timescale for civilian  

Well, yes.  If not, you're going to need a heck of a treaty to  
implement a viable leap hour instrumentality six centuries hence.   
Few of today's signatories may even exist.  Others may arise whose  
opinion of early 21st century timekeeping is less than nil.

Civil clocks display subdivisions of the civil calendar, and thus of  
the mean solar day.  Wishing won't make it otherwise.


Equitable estoppel

2006-12-17 Thread Rob Seaman
I've had a great time reading the Parrish temporal brief.  If I  
didn't have a massive deadline looming, I'd now start digging into  
some of the more intriguing references from the footnotes.  Maybe in  
a couple of months.  In the mean time (so to speak), I have some  
comments on a variety of the points raised.  Let's begin at the end,  
and end near the beginning.

Parrish closes with a paragraph starting, Time does not simply  
regulate our activities, followed by philosophic words like minds  
and souls, self-identification, religious experience,  
mysterious and elusive.  The point here is that these aren't the  
words of your euphuistic, periphrastic interlocutor, but rather of a  
staid law professor who chose these words with the care of the  
director of a law library to summarize the findings of a detailed  
legal review containing more footnotes than text.

Two paragraphs prior we are told that, Knowing the correct time is  
as important to the well-being of modern people as any prerogative  
outlined in the Bill of Rights.  Presumably law professors invoke  
the Bill of Rights with as much care in their writings as physicists  
do Schrödinger's equation, or theologians the Sermon on the Mount.   
What is the correct time, however?  This is the fallacy of the  
complex question - something is taken for granted that ought be  
regarded as doubtful, as one definition has it.  Is there a single  
class of correct time?  Or is interval time, for instance, distinct  
from time-of-day?

Jumping backwards once again to the previous section, Congress  
Finally Acts, we find that the enemies of daylight saving scorned  
golf, again and again, as the wasteful indulgence of a parasitical  
class.  Reporting, as I do, from one of the few states where the  
enemies of DST triumphed, I'm taken by the thought that both sides  
got what they wanted in Arizona.  There certainly are a lot of golf  
courses around here...

The mapping of timekeeping onto the Legal Process Theory seems very  
apt:  Time determination started as a private matter, became a group  
concern, ultimately required the intervention of [...] the courts,  
[...] and finally was settled by the [...] Congress, who delegated  
[...] implementation [...] to [...] the Interstate Commerce  
Commission.  The question that comes to mind is whether we've  
reached a point of diminishing returns over the regulation of  
timekeeping.  I'd say yes, that UTC as a representation of Mean Solar  
Time provides a natural arbiter for any disputes, legal or otherwise,  
that continue to arise.  On the other hand, a pseudo-UTC realized as  
some static offset from TAI would sacrifice the traceability of civil  
time to this ultimate apolitical standard.  You can't argue with the  
Sun, but (obviously) there is no difficulty in arguing with the ITU :–)

The starving the jury discussion was new to me - I would have  
thought that the cost of sequestering a jury would always have been  
borne by the government (that is, by the people, jointly).  Subtle  
legal issues - and religious, historical, social and economic issues  
- should be addressed head-on in any realistic proposal for change.   
It will all be for the best is not a plan, and the project  
requirements aren't all technical in nature.

Finally and firstly, the construction against drafter rule is  
exquisitely apt regarding UTC.  It is a general rule of  
interpretation that an expression is to be interpreted most strongly  
against the party responsible for its drafting.  If we regard UTC as  
a contract between the precision timing community and - well -  
everybody else, then surely it is the interests of everybody else  
that are controlling.  Which is to say, that having convinced us to  
adopt leap seconds three decades ago, the drafters of the UTC  
standard should certainly be expected to live up to this provision of  
it now.

Rob Seaman

A type of estoppel that bars a person from adopting a position in  
court that contradicts his or her past statements or actions when  
that contradictory stance would be unfair to another person who  
relied on the original position.

Re: what time is it, legally?

2006-12-13 Thread Rob Seaman

On Dec 12, 2006, at 5:56 PM, [EMAIL PROTECTED] wrote:

To avoid such failures in the future, Tom Van Baak has agreed to
take over its management and he is now working on the technical
issues involving the migration.

Thanks for looking into that.  Thanks to Tom for accepting another
(nearly) thankless chore.  Thanks to everybody who checked their mail
folders.  I'm relieved to find the issue appears archival only, not a
problem with the initial distribution.  Would hate to think of all of
you being deprived of my pellucid wisdom :–)


Re: what time is it, legally?

2006-12-12 Thread Rob Seaman

On Dec 12, 2006, at 9:38 AM, Zefram wrote:

...a lot of stuff I agree with.

Standard timezones have replaced solar time in general use

But this needs a clarification.  Standard time replaced local
apparent solar time in several steps.  First, clock (mean) time
replaced apparent time for civil purposes.  As you can see from the
proliferation of railroad standards, these were both still local to
one place or another.  Later, local time was referenced to standard
localities such as Greenwich.  Still later, a loose international
consensus was formed regarding a common time zone system with a
single standard prime meridian.

All of these remained solar time.  Mean solar time of some remote
location is still a flavor of solar time because there is no secular
drift.  The important issue is the continuity of still recognizing
mean solar time as the foundation of civil time.  Leap seconds are
simply one possible mechanism for achieving this.  The notion of a
leap hour fails to preserve mean solar time in any practical fashion.

Rob Seaman

Re: what time is it, legally?

2006-12-12 Thread Rob Seaman

On Dec 12, 2006, at 1:57 PM, M. Warner Losh wrote:

I view the same data differently.  I see it as a progression:

Local Solar time - mean local solar time -
  timezone as mean local time at one point used for many - UTC
- ???

Clearly, we're moving away from solar time and towards something else.

Only if you successfully make that last leap into the unknown.

Rather, clearly we've refined our understanding of solar time over
the last couple of centuries.  As demonstrated previously, all
parties agree (even if they don't agree that they agree :–) that
short of some caves of steel science fiction scenario, civil time
must mimic mean solar time closely.  We disagree over the definition
of closely, but it is better than one second per day unless one
chooses to pretend that our customers – that is, all the citizens
of all the countries of the world – would suffer a leap hour every
decade (365 x 10  3600s).  Consider that this one second (or much
tighter) tolerance is almost two orders of magnitude smaller than the
annual variation in the length of the apparent solar day.

From my point of view, closely mimic should be regarded as is –
that is, as a mathematical identity, not just a boundary condition to
be met by forcing equality once every millennium.  On the other hand,
the precision timekeepers who proposed the absurd leap hour proposal
apparently deem a few milliseconds per day secular trend as being
tolerable slop.  (Rather strange coming from technologists who dote
on nanoseconds.)  My position is:

   Civil Time == Mean Solar Time   (i.e., time-of-day)

The time lords assert:

   Civil Time = Mean Solar Time + epsilon   (something masquerading as

Which begs the question of what Civil Time is, if it ain't MST but
must be approximated so.

The mere fact of the ALHP arising among those who clearly loathe leap
seconds is a tacit admission that mean solar time rules now – and
will continue to do so in the future.  The leap seconds don't get
legislated out of existence, after all, they are merely embargoed for
several centuries.  The proponents of the ALHP are not suggesting a
way to escape from this constraint, rather they are acknowledging it
precisely by the form of their Machiavellian non-proposal.

One might suggest that if 1) the ALHP were taken off the table such
that we didn't have to keep batting the smelly thing away, and 2) its
proponents would deign to participate in a dialog, then 3) progress
might actually be made on solving the real problem of conveying BOTH
interval time AND time-of-day to the precision time users (i.e.,
people) of the world.

It isn't revolutionary to suggest that you look before you leap

Rob Seaman

Re: how posterity will measure time

2006-12-04 Thread Rob Seaman

On Dec 4, 2006, at 4:19 AM, Poul-Henning Kamp wrote:

Although, likely as not, when some future arkæologist finds the
inscriptions he will look at them without any formal training in
any kind of physics or natural science and conclude that they
probably have religious significance which is the default
explanation in that branch of history.

Many inscriptions from the last millennium do have religious  
significance, of course.

Current archeology involves a goodly amount of both high and low tech  
natural knowledge, from ground penetrating radar to a deep  
familiarity with the life cycles of various large and small  
scavengers and decomposers.  Archeologists also benefit from being  
familiar with the subjects of their investigations.  That nothing  
much is known about the builders of Stonehenge and other neolithic  
sites isn't the fault of the archeologists – they've done wonders  
with diverse cultures from Rome to Maya to Colonial America.  (The  
perennial questions of grave robbing and treasure hunting seem  
peripheral to the current discussion.)  Future archeologists, short  
of a discontinuity major enough to recycle a large fraction of the  
Earth's crust, will have vast and detailed information to work with  
regarding our civilization.  It's also hard to imagine that they  
won't be motivated to do a good job of interpreting the semiotics of  
the detritus of our fallen world girdling civilization.  More shame  
on us, but perhaps we'll at least be able to serve as a good object  
lesson to our gill-breathing, three-eyed, land-flounder descendants.

The existence of archeologists, of course, implies a civilization of  
a certain level of sophistication.  The inscriptions on the WIPP  
aren't intended for educated readers like archeologists, but rather  
as a big Yuck! sign as used on children's medications.  Any group  
of hunter-gatherers who stumble on WIPP and think to raid it as they  
will likely have been raiding landfills and other fin de millénaire  
treasure troves, will first have to pass the threshold of being  
capable of gaining physical entry.  If they are skilled enough to do  
this, they are skilled enough to make some simple astronomical  
observations (assuming the stars are still visible through the  
cesspool we will have made of the atmosphere).  One imagines the  
corpses of previous diggers will serve as an even better warning sign  
for the successive neolithic survivors of repeated discontinuities.   
A hundred centuries is a long time.

The PP in WIPP stand for Pilot Project.  The point of a pilot project  
is to teach something to current engineers, archeologists and other  
scientists, the public, and maybe, just maybe – to a few politicians.

I'd vote, myself, for using a subduction zone for this purpose,  
although having a goal of launching our waste into the Sun might  
serve to invigorate the space program for a few decades.

Rob Seaman

Re: how posterity will measure time

2006-12-04 Thread Rob Seaman

On Dec 4, 2006, at 9:41 AM, Rob Seaman wrote:

Any group of hunter-gatherers who stumble on WIPP and think to raid  
it as they will likely have been raiding landfills and other fin de  
millénaire treasure troves, will first have to pass the threshold  
of being capable of gaining physical entry.  If they are skilled  
enough to do this, they are skilled enough to make some simple  
astronomical observations

I suppose I might comment on the implied leap second topic, too.  No,  
I don't think this particular issue overlaps much with UTC policy  
questions.  On the other hand, it is a good example of the  
distinction between absolute and relative (interval) time  
references.  UTC as currently constituted (and universal time in  
general) represents an absolute time reference keyed, as with the  
WIPP signage, to astronomical observations.  An amateur astronomer  
with a Celestron, the Astronomical Almanac and an atlas can recover  
UTC anywhere on Earth.  An advanced amateur with the same telescope  
and a slightly larger reference library could do so for the next few  
thousand years.  A dynamically trained astronomer could do so from  
first principles and some boundary conditions obtained, for instance,  
from archeology.

Unplug the atomic clocks for a few seconds (which may be taken as the  
definition of a discontinuity in higher civilization), and even the  
professional timekeepers who built the devices would be unable to  
recover TAI.

Also, many amateur astronomers have designed and built their own  
telescopes (and sundials).  Even a tribe of hunter-gatherers could  
afford their shaman enough free time to build a serviceable 'scope  
assuming naked eye observations weren't sufficient.  The instructions  
could also be carved in granite if no issues of SkyTelescope survive.

Building an atomic clock posits maintaining a continuously lofty  
level of civilization indefinitely.

One imagines the corpses of previous diggers will serve as an even  
better warning sign for the successive neolithic survivors of  
repeated discontinuities.

A civilization that affords freak show proprietors the opportunity  
to display plastinated cadavers mimicking macabre everyday  
activities might even more reasonably pre-position a few such  
plasticized corpses for the purpose of preserving its descendants  
from agony and a lingering death.  Instead of Sagan's etched plaque  
depicting friendly humans, include a tableau of friendly humans  
themselves, complete with simulated pustules, vomit and blood  
spurting from strategic orifices.  The display's feng shui could be  
modeled on the more heart-rending family scenes found at Pompeii.


Re: how posterity will measure time

2006-12-04 Thread Rob Seaman

On Dec 4, 2006, at 4:27 PM, Ed Davies wrote:

Do you really mean UTC here?

Well, I mean any of the various approximations of Universal Time as a  
synonym for Greenwich Mean Time.  As continental drift becomes  
important, the job gets harder.  (But then, to return to the original  
topic, PHK and I would rely on continental drift to dispose of  
nuclear waste.)

I can see that an amateur with a
Celestron could recover UT (for some flavour of UT, I'm not sure
which - UT0?, then presumably UT1 after traveling around a bit)
but where does the delta T come from to get UTC?

From a knowledge of a (useful) policy for issuing leap jumps.  If we  
stick with individual leap seconds, we'll be within +/- 0.9s.  Good  
enough for government work – considering we're assuming that multiple  
governments will have toppled in the mean time.  More to the point,  
good enough for recovering time series for astronomical and planetary  
science work. etc.

Actually, assuming somebody remembered to make a note of
TAI-UTC before forgetting to put a shilling in the meter for
the atomic clock TAI is exactly as recoverable as UTC in the
short term when it's possible to work out the number of leap
seconds which would have been inserted or removed.

Well, sure, I'm willing to reboot TAI from UTC (w/ leap seconds) –  
kinda makes my tediously familiar case.  Ditch leap seconds for the  
nonsensical notion of leap hours, however, and we'd be in real  
trouble vis-a-vis scientific opportunities in our post-apocalypse  
scenario.  Hoo-boy!  Watch out!

Longer term it would be harder, of course, but why would that matter?

I was just taking the opportunity to stay on message, of course.  The  
underlying point is that interval time and time-of-day are entirely  
distinct concepts.  My position, of course, is that civil time should  
remain time-of-day.  Judah Levine and David Mills have already solved  
our problems, of course:


The fine print

2006-11-29 Thread Rob Seaman

I was rummaging around to see what web service based UTC clocks I
could find for a small data-taking project.  (Small project, large
data, actually.)  I happened to visit:

and found this disclaimer:

   This web site is intended as a time-of-day service only.
   It should not be used to measure frequency or time interval,
   nor should it be used to establish traceability to NIST or the

During this Holiday Season (sorry, Mr. O'Reilly), I'm reminded of
Miracle on 34th Street:

   Your Honor...every one of these letters is addressed
   to Santa Claus.  The Post Office has delivered them.
   Therefore, the Post Office...a branch of the federal
   government...recognizes this man, Kris Kringle...
   to be the one-and-only Santa Claus!

If NIST and USNO, official agencies of the United States government,
declare time-of-day to be distinct from time interval, who are we to
disagree?  As the New York State Supreme Court rules in the play:

   Since the United States government...
   declares this man to be Santa Claus...
   this court will not dispute it.

Case dismissed!

Rob Seaman

Re: Titan Time

2006-10-26 Thread Rob Seaman

Zefram wrote:

the radian is not a very practical unit.

There's nothing at all wrong with the radian - but there is a reason
calculators let you switch between degrees and radians.  Each is best
for particular purposes, just as interval time and time-of-day are
best for different uses.  See also natural and common logarithms.  Or
even the continuing value of Newtonian mechanics in the age of

I prefer to use the circle: 1 cr = 2 _pi rad.  Similarly, for solid
angle, the sphere: 1 sf = 4 _pi sr.

...leading to the need to rely on pico, femto or atto spheres on a
day-to-day basis?

83 mcr (millicircles).  That gives me a good image of where the sun

Fine.  Glad for the acknowledgment that the position of the sun in
the sky matters.

I prefer to cut everything decimally, including the circle and the
Actually I might prefer to cut everything octally, but decimal is the
dominant standard so let's use it.  Consistency is the key.

Point the Google at consistency emerson for the canonical quote
on this subject.

For angles, I'll merely point out that you appear to be willing to
sacrifice not only the 24 hours in a day (what would Jack Bauer
think?) – but also the ability to express right angles and a
multitude of other special values of trigonometry, leading to
expressions like cosine (166.667 mcr) = one-half.  And my kid thought
trigonometric identities were dubious before...

I doubt even the ITU would claim authority over the definition of pi :–)


Re: trading amplitude for scheduling

2006-08-04 Thread Rob Seaman
Hi Tom,Careful not to confuse rate with acceleration andpropagate a common misconception that leapseconds are due to an acceleration/decelerationeffect (as in "leap seconds are due to the earthslowing down").In fact, leap seconds are simply due to the earthbeing slow. How it got to be "slow" and whetherit is "slowing" are another issue.There are valid scenarios where the earth could bespeeding up and yet we would continue to have leapseconds -- because the earth is still slow (comparedto the SI second).Rate is not the same as acceleration; leap secondsare directly due to a rate mismatch, and only indirectlydue to deceleration.You appear to be addressing me, which suggests that my message (half of it, anyway) didn't parse for you into the exact same statement as yours that I was attempting to make.  See various messages of mine over the years, decades and centuries this list has been in existence.  See also discussion at have two clocks.  Their rates are already different (and will diverge further whatever the ITU does).  Therefore, one of the clocks must be reset occasionally.  Over timescales of interest to anybody attempting to understand the Earth and its place in the cosmos, however, the fact that we have a big dang moon is not negligible.You say "directly" and "indirectly".  I might choose "proximally" and "ultimately".  After all, the length of the second was supposed to scale to the rotation of the Earth, just as the length of the meter was supposed to scale to the size of the Earth.  The fact that we live on a charmingly wobbly planet is not something our policymakers should be encouraged to attempt to ignore.Rob

Re: trading amplitude for scheduling

2006-08-04 Thread Rob Seaman

John Cowan wrote:

1) We have leap seconds because the Earth rotates more slowly
than once every 86,400 SI seconds.

2) Leap seconds will become more frequent in the future because
the Earth is decelerating.

3) Leap seconds occur irregularly because the Earth's deceleration
is not constant and in fact changes unpredictably.


Right.  One might, however, choose to restate #1:

1) We have leap seconds because the SI second is shorter
than 1/86,400 of a mean solar day.

The SI second rather matches the length of the day c. 1820.

And in a post-leap hour world, one could also say:

1a) We have leap hours because the Earth rotates more slowly
than once every 24 SI hours, or

1b) We have leap hours because the SI hour is shorter
than 1/24 of a mean solar day.

2) Leap hours will become more frequent in the future because
the Earth is decelerating.

3) Leap hours occur irregularly because the Earth's deceleration
is not constant and in fact changes unpredictably.

Presumably one could identify 1 SI hour as  3,600 SI seconds.
The problem with this is that an hour has always meant 1/24 of
a day, so one is really redefining the concept of dayness.
And a day has always meant a subdivision of the calendar, so
one is redefining the calendar.

If an Earth day has nothing to do with the rotation of the Earth,
why should a Mars day have anything whatsoever to do with


Re: trading amplitude for scheduling

2006-08-04 Thread Rob Seaman

John Cowan accepts the blame:

1) We have leap seconds because the SI second is shorter
than 1/86,400 of a mean solar day.

Post in haste, repent at leisure (I've been going with too little
sleep lately, for reasons unknown...)  I actually do know that
the earth rotates in less than 1 mean solar day.

Blame for what?  I'm left wondering.  Are we now fretting about
the distinction between sidereal and solar time again?  Or perhaps
about the so-called equation of time - itself merely a cumulative
effect of the roughly +/- 30s excursions of length of day.  I thought
John's explication of the two differing clock rate issue was pretty
sound - other than my usual comments on point of view.

We have seen lots of examples of how even experienced users
of time can find themselves saying things that are imprecisely or
incorrectly phrased.  To some extent, this whole debate is really
a question of whether it is therefore better to try to sweep these
complications under the carpeting - or rather, whether such an
attempt would be kind of - well - nutty.

A leap hour is just 3600 embargoed leap seconds.  The best
interests of precision timekeeping will surely be better served
by seizing 3600 separate opportunities to educate the public
about the need for precision timekeeping.  Consider that the
recent leap second made the front page of major newspapers,
generated dozens of magazine articles, and even produced
discussion on TV and radio (and certainly the internet).  A least
one artist has taken the leap second debate as inspiration.

I'm skeptical that the ALHP can succeed.  But imagine it does.
What would the result be?  One big result will be that precision
timekeeping will vanish almost completely from public discourse
and international policy making.  Where is the benefit?


Re: trading amplitude for scheduling

2006-08-04 Thread Rob Seaman

John Cowan wrote:

I accidentally specified sidereal rather than mean solar days by
using the wording the Earth rotates.

Rotate is as perfectly good a word to use relative to our nearby star
as to the distant ones :-)  The solar system is chock full of nifty
periodicities and resonances.

No, that won't last forever, but neither will any
other scheme -- when we get to the 36-current-hour day, the connection
with solar time will *have* to be broken, unless we have evolved (or
evolved ourselves) to cope with very different sleep-wake cycles.

We have evolved, we are evolving, we will evolve.  Unlike other species,
however, we have the opportunity to control the context of that
This doesn't have to rise to the level of Morlock versus Eloi to be
One can well imagine that a diurnal wake-sleep cycle will be selected
at a significant level.  The corresponding adaptations may be varied and
wonderful between species, however.

BTW, are we now in a position to give a reasonable figure for the
mean and standard deviation of the Earth's deceleration, or do
we not have enough data yet?

The day was 23 SI hours long during the Jurassic and about 22 SI
during the Devonian period, the age of the fishes about 350 Mya.
outrageous license, one can estimate that the day will be 36 SI
hours long
about 2100 Mys hence (no compensation in this estimate for the lessening
of the effect as the Moon recedes).

But, as Steve Allen has pointed out, the length of the day has always
exactly 24 hours by definition.  The notion of SI hours is spurious.

Since we've been arguing for seven years about a daily effect at the
of 2 ms, not 12 hours, one can assume that a consensus will need to be
formed some time sooner than the time it will take new evolutionary
as significant as the fishes or reptiles (including birds) to appear
some Futurassic period to come.

Would suggest that a consensus on the character of the problem be
established, before a consensus on its solution is sought.  In
if we're interested in evolutionarily significant spans of time, one
that we should not kowtow to the imaginary needs of makers of early
millennium technical geegaws.

As far as the measuring the slope, try starting with the first figure
from  Three slopes are
on the data from the past 2500 years.  The most shallow, 1.4 ms/cy,
corresponds to the backwards extrapolation from recent behavior.  The
most steep, 2.3 ms/cy, is derived from direct measurements of Lunar
recession - the angular momentum has to balance.  A fit to the
overall trend
is intermediate at 1.7 ms/cy.  The short term and long term shape to the
trend line (with the interesting hint of a ~1200 year period) is
affected by numerous geophysical issues such at the continuing rebound
of the continents since the glaciers receded.  It can also give you
some idea
of an intrinsic width of any estimate.  Hard to know how to comment on
standard deviation when there are clearly effects yet to be
accounted for.

Would think that the lunar value provides the best handle on the long
behavior since there are orders of magnitude of natural smoothing in the
orbital angular momentum versus the rotational.  I doubt the point need
be emphasized that I'm no expert :-)  Would be delighted if this
list's silent
experts were to correct my gaffs and omissions.


trading amplitude for scheduling (was Re: [LEAPSECS] leap seconds in video art)

2006-08-03 Thread Rob Seaman
Brian Garrett wrote: the mini-lectures did imply that leap seconds compensate for secular deceleration of the earth rather than seasonally accumulated differences between UTC and UT1. To the extent that I understand the point you are aiming for, this statement conflates two issues:  1) that the long term secular deceleration is only perceptible as a baseline trend hidden beneath large amplitude, short period, effects, and  2) that leap seconds aren't the result of new slowing (or speeding up), they are the result of cumulative clock corrections required due to previous slowing.I agree with both, but just because we're bleeding off the cumulative terms seasonally (which I take to mean "over short term periods of whatever duration and connected to whatever physical cause"), it doesn't mean the secular trend is not pertinent.  We're required to synchronize two clocks whose rates differ.  We cannot adjust the rates ourselves and they are slowly drifting further out of step due to the inconvenience of physical reality (or a "charming fact of life in the solar system" as someone said).The impact of the short term effects isn't to either increase or decrease the net number of leap seconds, it's to play bloody hell with their scheduling.  I perceive this as a fact of life that should be accommodated (and that serves as job security for folks like Daniel Gambis).  Others perceive this as an awkward reality to attempt to ignore (although the ALHP will result in twice as many leap seconds than otherwise when the first leap hour is announced in 2606 by Obi Wan Gambis).The charm of Felicity's work is in the implied distinction between the trees of the experts and the forests of the public.Rob SeamanNOAO 

the case for created time

2006-07-18 Thread Rob Seaman
We have all been so utterly wrong!  The scales dropfrom my eyes ( theory of evolution for the creation of the solar systemseems less than satisfactory in regard that the Earth andMoon appear to generate interrelated time cycles.A prize (well, a beer when next we meet) to the bestexplanation of what the heck this guy is on about..although, there may be an object lesson here.What seconds God hath leaped, let no man put asunder.Rob SeamanNOAODrop from my eyes, ye scales of time's applying - from A Poet's Hope, William Ellery Channing II ("not the most illustrious of the transcendentalists")

Re: PT Barnum was right

2006-07-06 Thread Rob Seaman
Steve Allen wrote:In the this week is a press release for a clock thatautomatically tracks leap seconds.Anybody volunteering to tell these guys that their product is about to be orphaned?  Sounds like a lawsuit in the making.  Would think the ITU lawyers would be interested in their own liability.But the only output is a liquid crystal display, and liquid crystalshave response times around 10 ms.  That's 1/100 s, not 1/1 s.The key word here is "only".  Nothing wrong with including a display, even if the precision is lessened.  The issue with the display isn't precision, it's accuracy - in that a correction for the display's response time is unlikely to have been included.  The digits will appear something like 10 ms too late.  Worse, the response time may well depend on the value reported, may vary from digit to digit, and may change with the age of the unit, etc.  Anybody have an opinion on the correct statistical distribution to use for modeling LCD transition behavior?This seems akin to all the complaints about GPS receivers whichdisplay a time that is off by about 2 seconds.  I've never botheredto dig on that, but my impression is that they probably also displaya position of where they were 2 seconds ago.Hmmm.  Does this apply to the kinds of GPS carried by airplanes, etc?  Or is this purely a problem for consumer grade units snapped up by the millions for $100 at Walmart?  A constant offset will tend to drop out of the equation (to first order) when any sort of calibration procedure is followed.  Higher order effects will emerge when the unit is moving at high speeds, or if direction or speed changes frequently.  Most troubling would be if two moving platforms are depending on GPS units with differing delays, e.g., two airplanes following neighboring flight paths.  How far does an airplane move in 2 seconds?  What is the minimum separation required by the FAA?  Again - this will preferentially tend to be a 2 second delay, never 2 seconds early. The CBS radio affiliate in the LA area very plainly is using a timecompressing/FFT pitch shifting device on the live national feed.  Thetime tone in LA always happens around 10 to 15 seconds after the hour.Classic!  Send it in to comp.risks.  (Search the archives, first.)As Steve knows, mountaintop observatories are great places to reveal unintended consequences.  One of my favorites was an interaction with a terminal window page view mode (happened to be Sunview, could happen with current technology, too).  The observer would type a command line to snap a lengthy sequence of several calibration exposures while they trotted off to dinner.  On their return, they would discover that the sequence had halted after only a few minutes and was waiting for a SAK.The problem isn't only with fixing such issues (such that they stay fixed), it's with recognizing that a problem exists and with having the imagination to comprehend contributing factors. A listener might note that the time signal was delayed, but may be unaware of the existence of time compression technology.  Invisible logistical details may also be key.Sometimes system delay is unintentional, sometimes it is intentional.Even before PT Barnum latin had a two word phrase for such products."Sucker bait"?  "Jumbo junk"?  "Electronic egression"?Somebody tell me again -- why is it thta broadcast civil time signalsneed atomic accuracy?I think you meant "atomic precision" here, even if it's less alliterative.  I think we all would like to see an improvement to the accuracy of civil timing against whatever underlying standard - accuracy averaged over typical ensembles of clocks.  Whether a particular clock is 15s fast or 15s slow, however, often does not matter.  (And your point is well taken about the policy making implications of chasing unneeded requirements.)A simple argument of regression to the mean suggests that clocks (such as radio time signals) used to set numerous other clocks should be responsive to a requirement for relatively high average accuracy.  The problem with your LA radio station is not that they are imprecise, it is that they are consistently wrong *in the same direction*.  This is actually something you might be able to get them to fix, should you choose.  The one thing the FCC appears to care about is performing station identification.  Interviews are constantly interrupted by the requirement to do so "precisely" at the top of the hour.  Send them a letter and copy the FCC.Improving accuracy often implies that precision is improved as a result.  The reverse is frequently untrue.RobNOAO

Re: PT Barnum was right

2006-07-06 Thread Rob Seaman

On Jul 6, 2006, at 12:46 PM, Brian Garrett wrote:

I was told that the station delays their broadcast in order to
enable on-the-spot editing of objectionable material.

Surely the requirement is to permit review of *potentially*
objectionable material.  A time signal is no such thing and need not
be delayed.  Proper system design would interpose time signals (and
perhaps other safe content such as weather reports or what have you)
after the delay line.  Am also skeptical that most stations employ a
24 hour censor to monitor and bleep all content before it reaches the
air.  (And who would want such a horrific job?)

it's in networks' best interests to do this even if it means
setting your watch to their time signal means being 9 seconds late.

That's just one use case, of course.  Folks who set their watches
using well synchronized time signals, or who consult their cellphones
or NTP driven laptops, will be ahead of this particular radio
station.  Nine seconds is meaningless for many purposes, but a 9s
simulcast delay would be intolerable, for instance.  Usage issues
might also be revealed when switching between stations.

My thought when we reach one of these topics is to marvel at the
chutzpah of proposing a solution like leap hours without investing
the slightest effort in characterizing potential issues.  Timekeeping
can't simultaneously be the deeply important issue we all must think
it is (or else would not be reading this :-) and also be worthy of
such abject neglect.

The 'time' as most of us know it is simply inexpensive crumbs from
the tables of the few rich gourmet consumers of time and
frequency information.

Astronomers have traditionally been not only among the most demanding
gourmets, but have also employed some of the greatest temporal
chefs.  This is indeed a pretty good analogy, although the word
inexpensive is out of place.  The point is that a crumb from the
table of a gourmet is still a gourmet crumb.  A parvenu of time can
also aspire to become a gourmet should the need or interest arise.  I
believe this describes the world we currently inhabit.

we're being served chronological junk food and most folks couldn't
care less.

This is a different analogy.  It isn't a question of most folks, it
is a question of for most purposes.  Even gourmets sometimes
appreciate a simple meal.  And on the other hand, the temporal hoi
polloi are dependent more-and-more on chronological caviar through
the offices of various technological agents.  Cellphones don't only
report high quality timing information, they and their networks
require this to operate.

But your analogy is quite apt for the world that would follow the
adoption of the Absurd Leap Hour Proposal.  All time signals would
then become junk food.  All gourmets would find themselves in the
position of dumpster divers.


Re: independence day

2006-07-05 Thread Rob Seaman

John Cowan wrote:

I regret to state that this remark appears to me no more than

Merely hyperbole intended to make a point about the art of crafting
fundamental standards.  Obviously I failed to make that point :-)

Why precisely, however, do you regret your inference?  If my
arguments were to be deemed specious, surely that would strengthen
opposing arguments (or at least remove competing options).

If the U.S. tied its legal time to the ITU, it could untie
it in future if that seems like a good idea.

and later in reply to Markus Kuhn:

Reader, suppose you were an idiot, and suppose you were a member
of Congress; but I repeat myself.   --Mark Twain (1882)

You can't have it both ways.  Either a prudent decision making
process is being followed, or it ain't.

What would be prudent?  How about taking the consensus at the Torino
meeting at face value?  A civil timescale without leap seconds should
prudently be called something other than UTC.  Heck - define it
right now and let the market decide.  If that consensus is deemed
unacceptable, what might be a prudent response?  How about holding
another - well advertised - conference to discuss the issues some
more?  Hold it in Washington so members of Congress can attend -
assuming they're capable enough to locate the venue :-)  What else
might be prudent?  How about taking our time characterizing the
problem fully before proposing solutions?  The current standard is
good for several hundred years.  What precisely is the hurry?

In any case, changing the legal definition of U.S. time from GMT
to UTC merely regularizes the de facto position, since GMT no
longer has a specific international definition.

That was my point, if only my épée hadn't rattled ineffectually
aside.  GMT has a physical definition that trumps any international
definition (although I'm a bit perplexed at why you assert the latter
to have gone completely missing).  Let me pose my argument again in
the hopes of snagging my point d'arret in your knickers:

1) Notionally, the first leap hour would occur 600 years hence should
the Absurd Leap Hour Proposal (ALHP) be rammed through by the
mumbling minions with fingers in their ears.

2) Six centuries ago, the New World was Terra Incognito - the Sistine
chapel was yet to be built, let alone painted - the Ptolemaic cosmos
and the Julian calendar were unchallenged - the Medici were in full
flower, although Machiavelli was but a potentiality inherent in his
grandparents - the Canterbury tales were written, but their publisher
was equally unborn - Shakespeare lay in the distant future, of
course, but his Prince Henry was still hanging about with Falstaff in
taverns, not the French at Agincourt - and the great Chinese eunuch
admiral Zheng He had set sail with 27000 men on 300 ships, with the
protectionist retreat of the later Ming dynasty still a century in
the future.

3) During all the time since, flavors of solar time have provided the
fundamental standard for timekeeping in Asia and in Europe and in the
Americas - and across civilizations on all the continents, and on
islands scattered across the world's seas and oceans, and on vessels
traversing those oceans and later the skies above them.

4) Who knows what changes the next six centuries will bring?  Rather
than being an argument for the timekeepers having the freedom to
follow whatever policy making whims seem expedient - this question is
instead a demand for prudence beyond bureaucracy and deliberation
beyond misconstrued self-interest.

5) A time standard rooted in solar time can be recovered at remote
times and in diverse places.  Patrick O'Brian's pugnacious ill-
educated Napoleonic era characters perform this feat daily from the
rolling quarterdeck of a Frigate with sightings taken to establish
local noon.  Their chronometers are synchronized to the observatory
at Greenwich, not to some random clock in a basement in Paris (not
coincidentally, that would have been under the control of their

6) A time standard rooted in an ensemble of clocks, on the other
hand, is subject to the vagaries of happenstance and history (like,
say, another Napoleon).  What price to ensure 24/7/365/600
reliability?  (I look forward to your riposte pointing out that the
metric system emerged from the Reign of Terror :-)

7) The calendar, and its constituent subdivisions by the clock, is
the mother of all international standards.  It deserves the respect
we show our own mothers, not the derision reserved for avoirdupois or

What in practice would stop these individuals
from leaping the clock forward or backward at will, or from changing
the rate of UTC, or for that matter from making the clocks run

The fact of being rendered irrelevant, not to say a laughingstock.
What is to prevent the IERS from issuing bogus leap second

Precisely the constraint that DUT1  0.9s.  Precisely the fact that
UTC is currently tied to an underlying physical 

Re: building consensus

2006-06-08 Thread Rob Seaman

Clive D.W. Feather wrote:

March was the first month of the year; look at the derivation of
September, for example.

Makes the zero vs. one indexing question of C and FORTRAN programmers
look sane.  I've pointed people to the whole 7, 8, 9, 10 sequence
from September to December on those (admittedly rare) occasions when
the issue has come up.  Presumably other languages agree in usage,
which would be another indicator of the age of the names of the months.

The *seven* day week was, but before then the Romans had a rigid
*eight* day week.

The latter, of course, persisted all the way into the 1960's, as
immortalized by the Beatles' song.


Re: building consensus

2006-06-08 Thread Rob Seaman

John Cowan wrote:

In the cover story, it was used as a final
defense against the Invaders and destroyed by them.  In the true
story, it was destroyed because it constituted a hazard, but I
forget exactly how.

Thanks!  But not sure true story is the opposite of cover story,
here :-)

Both versions of the book are sitting in a box somewhere in the
garage.  It must be twenty or thirty years since I read whichever.
Few writers other than Clarke had the chutzpah or cleverness to write
a viable story placed billions of years in the future.  Now that you
mention the lunar plot twist, I do remember something about it -
strange that other aspects of the story remain much more vivid.  Of
course, the other civilizations were tossing stars around, not just
moons, so it may have seemed pedestrian.


Re: building consensus

2006-06-08 Thread Rob Seaman

On Jun 8, 2006, at 8:08 AM, Clive D.W. Feather wrote:

Rob Seaman said:

Thanks!  But not sure true story is the opposite of cover
story, here :-)

I don't think John's referring to Against the Fall of Night
versus The
City and the Stars. Rather, at least in the latter, the official
story of Diaspar (sp?) and the Invaders disagrees in many aspects
with the
true story

Right.  Was merely questioning whether a story within a story within
a work of fiction could be regarded as true :-)

Re: building consensus

2006-06-07 Thread Rob Seaman

On Jun 7, 2006, at 2:01 AM, Clive D.W. Feather wrote:

Actually, the evidence from experiments is that the natural sleep-
cycle is about 27 hours long, but force-locked to the day-night
cycle (it's
easier to synchronise a longer free-running timer to a shorter
signal than vice-versa).

References for this?  Your explanation makes a lot of sense and I'm
prepared to be convinced, but have been skeptical of experimental
design as applied to questions of human behavior since participating
in studies as a requirement of undergraduate psychology coursework.
And if this cycle is inferred from the behavior of undergraduates,
I'm even more skeptical :-)

So humans will cope until the solar day is about
27 (present) hours long, after which we'll probably start to move to a
system of two sleep-wake cycles per day.

Well, cope isn't the right word if this cycle is as you describe.
Also hard to imagine how one gracefully transitions from one to two
sleep cycles a day.  It would simply appear that the underlying
mechanism evolved to rely on a slightly longer free-running timer
synchronized to length-of-day.  As the day lengthens, Darwin would
predict that our intrinsic cycle would also lengthen.  This is
similar to arms races leading to other periodic natural behavior such
as prime number 13 and 17 year locusts.  (Non-primes would allow
locust predators to emerge more frequently while locking into the
phase, thus gaining an advantage.)

Also, whether or not one believes that humans have somehow escaped
the grip of evolution, it is hard to imagine our continued sojourn on
Mother Earth half a billion years hence :-)

Obvious lines of research for further sleep period investigations
would be to examine similar cycles in other animals.  One imagines
this is some function of the nervous system, so one might also
contrast strategies pursued by plants and animals without.  Also -
how is this intrinsic cycle inferred?  Could signatures of this
intrinsic cycle be preserved in the fossil or DNA record?  All sorts
of other cycles are.  Could such signatures be correlated with length-
of-day at various epochs?

Just another mechanism tying our species to time-of-day.


Re: building consensus

2006-06-07 Thread Rob Seaman

Tim Shepard replies:

Also hard to imagine how one gracefully transitions
from one to two sleep cycles a day.

It is already the norm in some places:

Thanks for the chuckle.  One is then left wondering whether our far
future, Clarkeian Against the Fall of Night/City and the Stars,
descendants will enjoy *two* siestas per day.  With inhabitants being
reconstituted every few millennia from a computer database, one might
argue that evolution will have indeed ceased - so they won't be
adapting to the lengthening day by natural selection.  Of course,
they may have artificially stabilized the Earth's rotation by that
point, too :-)  Does anyone remember if Earth-Moon dynamics plays a
role in the story?


Re: building consensus

2006-06-06 Thread Rob Seaman
Ed Davies quoted:The Gregorian calendar provides a reference system consisting of a,potentially infinite, series of contiguous calendar years. Consecutivecalendar years are identified by sequentially assigned year numbers.A reference point is used which assigns the year  number 1875 to thecalendar year in which the “Convention du mètre” was signed at Paris.The Gregorian calendar distinguishes common years with a durationof 365 calendar days and leap years with a duration of 366 calendardays. A leap year is a year whose year number is divisible by fouran integral number of times. However, centennial years are  not leapyears unless they are divisible by four hundred an integral numberof times. This is from section of the final draft of ISO8601:2000.This does not express a complete algorithm, of course, since it is not tied to the days of the week.  One presume the zero point of 1875 was selected as a practical compromise with historical realities.  One questions whether the Gregorian calendar can be used to express dates before 1582-10-15, and it is ambiguous for large parts of the world until the mid-eighteenth century.  Still in 1875, it had not reached all locales.  Etc.Zefram wrote:Pope Gregory, of course, used a different epoch.  The original Papal Bulldidn't use a well-known event such as this, but instead effectively said1582 October 5 Julian = 1582 October 15 Gregorian.This is very familiar territory since I had two semesters of the History of Astronomy from an Augustinian priest.  I had not previously read through the Papal Bull in question, however.  I wouldn't rely on this as a fundamental reference, but see the translation at: few comments.  First, the name of the Bull is "Inter Gravissimas" or roughly "Among the most serious".  These are not issues to be trifled with.  Second, from paragraph 5: "Our dear son Antonio Lilio, professor of science and medicine, brought to us a book, written at one time by his brother Aloysius [Luigi], in which this one showed that, by means of a new cycle of epacts which he had devised, and who directed his own particular Golden Number pattern and accomodated the entirety of any solar year, every [defect of] the calendar collapsed, and the constant calculations would endure for every generation. He was, thus, able to restore and explain how the calendar itself will never need published any further change."A worthy goal.  That residual discrepancies (that were presumably within the measurement error in the sixteenth century) will ultimately require currently unscheduled modifications to the Gregorian calendar, does not diminish the achievement of Aloysius/Luigi.A successful proposal to reform civil time (as with the civil calendar) should exhibit this same worthy goal of actually improving the status quo, rather than simply arranging to ignore a reality that will return with a vengeance after 600 years.Gregory also had a refreshing attitude toward the standards process:  1)  "This new project of the restoration of the calendar, summarized  in a small book, we forwarded a few years ago to the Christian princes and to the large universities so that this work, which is the business of all, is carried out with the consultation of all."  2)  "Those having expressed their agreement to us, as we had sincerely  hoped, we have, with this consensus, arranged a gathering in the Holy City, to reform the calendar, of the very qualified men on the matter whom we had chosen from the principal countries of the Christian world a long time before."  3)  "Those, after having devoted much time and attention to this night  work and having discussed between them cycles which they had  collected from everywhere, old ones as well as modern ones, and  as they had carefully studied the reflections and the opinions of  erudite men who wrote on this subject, chose and prefered this  cycle of epacts,"  4)  "adding to it elements which, after thorough examination, appeared  essential to the realization of a perfect calendar."So, they:  invested sufficient time to understand the issues - recognized the extremely broad range of stake holders - sought their advice and consent in characterizing the problem - invited representatives to help craft the solution - and more to the point, were looking for an actual solution, not some imaginary bandaid.Rob SeamanNOAO

Re: building consensus

2006-06-05 Thread Rob Seaman

On Jun 4, 2006, at 9:57 PM, M. Warner Losh wrote:

leap days have a rule, while leap seconds are scheduled.

A schedule and a rule are the same thing, just regarded from
different historical perspectives.  The leap day rule will most
certainly have to accommodate scheduling changes over the millennia.
On the other hand, I am sure we haven't exhaustively discussed
possible refinements to the leap second scheduling algorithm.  (And
ain't that a rule?)  If we have to spend all our time fending off
this silly leap hour proposal, we'll never have the opportunity to
focus on rules and algorithms (not to mention technology and

Just don't do it is not a rule.

The biggest difference between leap days and leap seconds is that
days are quantized.


Re: building consensus

2006-06-05 Thread Rob Seaman
On Jun 5, 2006, at 8:45 AM, Warner Losh wrote:Leap days have an iron-clad rule that generates the schedule on whichthey happen.  Leap seconds have a committee that generates theschedule on which they happen.Further discussion in this thread calls into question the characterization of "iron-clad rule" :-)One might ponder what standards body is responsible for the international calendar specification.  Is it the Roman Catholic church?  Or has the specification passed into the public domain?  Are individual nations each responsible for their own calendars?  If so, mustn't they then be responsible for trade and scientific purposes for providing tables of conversions between their national calendar and the international standard?  Which then returns us to the question of who is responsible for that international standard...We have discussed having some kind of rule for when leap seconds are inserted.Yes, but note that the IERS could institute a wide variety of scheduling algorithms *on top of* the current monthly (or twice yearly) leap second constraint.  If the state of the art allowed predicting UT1 for a decade in advance, a table of leap seconds could be provided a decade in advance.  This option requires even less of a change than the Absurd Leap Hour Proposal (ALHP).On Jun 5, 2006, at 8:57 AM, John Cowan wrote:On the other hand, I am sure we haven't exhaustively discussedpossible refinements to the leap second "scheduling algorithm".  (Andain't that a rule?)I thought the whole point was that while we had a rather good predictionof changes in the tropical year (viz. none), and therefore only have todink with the calendar when the current error of about 8.46 seconds/yearaccumulates to an uncomfortably large value, there is simply no knowing,in the current state of our geophysical knowledge, how the wobbly oldboulder in the sky is going to wobble next.The biggest difference between leap days and leap seconds is thatdays are quantized.Can you expound on this remark?A calendar counts days.  A day - whether from noon to noon, midnight to midnight, sunrise to sunrise, or sunset to sunset - is an atomic "quanta" of time on earth.  It also happens to be growing relative to the year.  Ultimately calendrical and clock issues are the same.  (The historical time horizons over which various effects matter for various purposes may be very different, of course.)The ALHP is an attempt to redefine the day.On Jun 5, 2006, at 9:27 AM, Zefram wrote:In the realm of calendars the terminology is "arithmetic" versus"observational".  That's one of the things I included at the start ofthis thread.  I'd also like to throw in the word "deterministic".The Gregorian calendar itself is strictly arithmetic and thus immutable.There is the alternate point of view that the calendar in actual civil usein a particular locality, changing between different arithmetic calendarsat different times, constitutes an unpredictable observational calendar.Perhaps we need a concept of "calendar zone" analogous to time zone,with a calendar zone database to match.So the calendar is either immutable - or it isn't :-)I have a hard time reconciling the notion of a "calendar zone" with the definition of "deterministic" as: "an inevitable consequence of antecedent sufficient causes"For the sake of argument, however, assume that the Gregorian calendar is immutable - leap day every four years, except for even centuries not divisible by 400.  What will then happen when the Gregorian calendar is inevitably deemed to fail to serve?  Well, we already have historical precedent.  The Gregorian calendar succeeded the Julian, just as the Julian succeeded what came before.  That Caesar was more successful than Pope Gregory at convincing the world to rapidly adopt the new standard is a result of some pretty interesting historical differences between the two eras.  The fundamental fact, however, is that a new calendar was completely substituted for the old.  One might also note that the staged and delayed politically sensitive adoption of the Gregorian calendar was possible precisely because the Julian calendar continued in force.  In fact, it continues as a standard to the present day.  The Julian calendar was deprecated, but not redefined.Compare this with the ALHP.  I might disagree quite strongly with the idea of a leap hour - but I wouldn't have quite the visceral hate and utter contempt for the idea if the proposal were to also substitute a new name.  Instead of eviscerating UTC (a coherently defined entity that the ITU simply inherited), call it "McCarthy Time", for instance.  One would think that just as the Julian and Gregorian calendars pay homage to Julius Caesar and Pope Gregory, that the eponymous "MT" would be taken as homage to its creator..and if not, ask yourself, why not?In what ways is the ALHP unworthy of its authors?RobNOAO

Re: building consensus

2006-06-05 Thread Rob Seaman

On Jun 5, 2006, at 1:05 PM, John Cowan wrote:

(ObOddity:  It seems that in Israel, which is on UTC+3, the legal
day begins at 1800 local time the day before.  This simplifies
the accommodation of Israeli and traditional Jewish law.)

I wouldn't call this an oddity, but rather an interesting and
elegant, one might even say charming, local custom.  The logic of
this accommodation between 6:-00 pm clock time and a mean sunset
demonstrates another weakness in the ALHP, since clock time would
drift secularly against mean solar time.


Re: building consensus

2006-06-05 Thread Rob Seaman

On Jun 5, 2006, at 1:38 PM, John Cowan wrote:

I found another spectacular illustration of how massive the difference
between solar and legal time can be.  Before 1845, the time in Manila,
the Philippines, was the same as Acapulco, Mexico, a discrepancy of
9h16m from Manila solar time.  This was a consequence of the
having been colonized and administered from Spanish America.
Nowadays the
standard time of Acapulco is UTC-6; of Manila, UTC+8.

Q: What happened in the Philippines on December 31, 1844?
A: Nothing.  It never existed.

One might suggest that the accommodation between civil time and legal
time is of more interest.  What does it mean to say that some nation
or locality uses the Gregorian - or any other - calendar, if some
date legally does not exist as you suggest?  The sun certainly came
up on that day and rose the following day about 24 hours later.  A
variety of activities occurred on that day that fell into bins like
weekday, weekend or holiday - or if this was some red letter
day different from all others, then the authorities must have tacked
up fliers or alerted town criers or otherwise informed the populace
of the special nature of the day in question.  When they did that,
what did they call it?  The day after December 30, 1844?  Next
Tuesday?  (Which begs the question, of course.)  Suspect rather that
legal time only applied to certain specific interactions with
colonial authorities.  Would love more details.

An observing session with a ground-based nighttime astronomical
telescope typically begins on one calendar date and ends on the
next.  For some observatories in the western hemisphere, the session
starts after 0h UT such that the entire session can be trivially
labeled with a single date.  The problem with this is that the UT
date is one day after the date on the observing calendar.  For
observatories elsewhere, the data from a single coherent session are
split between two dates whether local or UT time is used.  The
solution I have adopted for the dozen telescopes in my bailiwick is
to establish a local noon pivot.  All data are assigned to the
calendar date at the start of the night.  This 12h difference is, in
effect, the maximum possible discrepancy between a legal date and a
solar date.  As always, the question is:   what is your timekeeping

In any event, the case you are basically making is that in throwing
off the yoke of their colonial masters, the Philippines specifically
chose that their legal time should match their civil time and that
their civil time should agree with local solar time.


Re: building consensus

2006-06-05 Thread Rob Seaman

On Jun 5, 2006, at 2:47 PM, John Cowan wrote:

I'm not sure what you mean by civil time in this context.

I meant whatever we've meant in this forum for the past five years.

For some people, civil time is synonymous with standard time; for
others, it means the time shown by accurate clocks in the locality.

I presume you aren't asserting that standard time clocks can't be
accurate, but rather distinguishing between standard (timezone)
time and local mean solar time?

On the other hand, all I've ever meant by the term civil time is
that time that a well educated civilian sets her clock in order to
agree with other civilians for civilian purposes.

There was no day labeled 1845-12-31 in the Philippines.
Consequently, the year 1844 had only 365 days there, and
the last week of 1845 lacked a Wednesday.

Interesting question:  On similar historical occasions, for instance
during the transition from old style to new style dates as the
Julian calendar gave way to the Gregorian, has the sequence of
days of the week remained unbroken?  Or rather, have days of the
week been skipped as well as days of the month?  Surely the Gregorian
calendar is not just a rule for adding a leap day every four years
sometimes), but also includes the definitions of the twelve months, and
an initialization of a specific day-of-the-week on whatever date.

This was not a calendar transition, but a (drastic) time zone
involving moving the International Date Line to the east.

Not obvious that there is any difference - kind of a calendrical
Mach's Principle.


Re: building consensus

2006-06-05 Thread Rob Seaman

On Jun 5, 2006, at 4:05 PM, Rob Seaman wrote:

On the other hand, all I've ever meant by the term civil time is
that time that a well educated civilian sets her clock in order to
agree with other civilians for civilian purposes.

I should clarify this to mean the underlying internationalized
standard delta'ed to local time.  Time zones are a trivial
refinement of a unifying theme.


Re: building consensus

2006-06-01 Thread Rob Seaman
Warner Losh objects:There are several doughty people here who happen to have that opinion, but they abide with us mortals outside the time lords' hushed inner sanctum.I have spent much time explaining why leap seconds cause real problems in real applications, only to be insulted like this.Sincere apologies for my awkward statement. defines "doughty" as "marked by stouthearted courage; brave".  I wasn't questioning the knowledge or passion of folks holding views that differ from my own.  Rather I was attempting to question whether anybody actively participating on this list - holding whatever view - is also participating in ITU discussions.I see that Mr. Cowan has also parsed my admittedly opaque remarks.RobNOAO

Re: extracting leap second schedule

2006-05-30 Thread Rob Seaman

On May 24, 2006, at 6:27 AM, Zefram wrote:

I've written a Perl module that handles the conversion between TAI and
UTC.  It's Time::UTC, available at
When asked about future times where it doesn't yet know the
leap second schedule, it automatically downloads and parses  This works well if
a future leap second already scheduled, but otherwise there is a
in working out how far ahead the last TAI-UTC offset applies.

Note that in a leap-secondless alternate reality, exactly the same
would be encountered when trying to predict UT1 given TAI (or its
evil kirk
twin, new-UTC).

Amusingly, in the long run this will become a non-problem on its own.
In perhaps a century or so it will become unavoidable to schedule
more than one leap second ahead.  That way the final time mentioned in
tai-utc.dat will always be in the future.

Non-amusingly, in the alternate no-time-of-day universe, this never
becomes a non-issue for recovering the orientation of Earth-2.

Rob Seaman
National Optical Astronomy Observatory

Re: Precision vs. resolution

2006-05-30 Thread Rob Seaman

On May 24, 2006, at 7:25 AM, John Cowan wrote:

Can someone lay out for me exactly what the difference is between
clock precision and clock resolution?

Interesting question.  Perhaps it is the distinction between
and physical pixels that one encounters on image displays and hardcopy
devices?  (Still have to posit which is which in that case :-)

You might have more luck directing this question to time-nuts
( or perhaps the NTP WG
([EMAIL PROTECTED]) - although one would be delighted to find
this list capable of generating a knowledgeable response to any
such clearly expressed neutral question :-)


Re: ideas for new UTC rules

2006-04-15 Thread Rob Seaman

Only hours ago did I learn of the recent problems with D-Link routers.

Remarkable!  Just imagine the logical disconnect at the product
development meetings.  The marketing folks emphasizing the highly
desirable feature of NTP compliance and the tech folks tossing a list
of 50 servers into the center of the table - a list they probably
spent all of a half hour compiling immediately before the meeting.
Neither group pondering for even the briefest flicker what effect
their product and customers would have on the servers, or conversely,
what value the company was proposing to scavenge for free from using
those servers.

Even people in the internet industry appear to believe that it just
exists free for the picking.  These bozos haven't a leg to stand on.
Am especially baffled at  why it wouldn't occur to D-Link that it was
their responsibility to field their own NTP servers.  This is even
more basic than the resource discovery issue.  Hardwired host names -
bah!  Hardwired host names belonging to somebody else?  Absolutely

Not to give the slightest indication of blaming the victim, but am a
bit perplexed exactly why this devolves to an issue of Danish
infrastructure at all.  Would think the EU would be the appropriate
entity to plan, specify, fund and deploy time servers.  I heartily
applaud PHK for undertaking this volunteer commitment, but he's not
the first and won't be the last good samaritan to caught up in the
gears of commerce.


Re: ideas for new UTC rules

2006-04-14 Thread Rob Seaman
On Apr 13, 2006, at 10:41 PM, Steve Allen wrote:Today is one of the four days in the year when Newcomb's_expression_ for the equation of time has a value of zeroand it was Samuel Beckett's hundredth birthday.  Leap second as Godot:  ESTRAGON: And if he doesn't come? VLADIMIR:  We'll come back tomorrow. ESTRAGON:  And then the day after tomorrow. VLADIMIR:  Possibly.  ESTRAGON: And so on. VLADIMIR:  The point is—  ESTRAGON: Until he comes. VLADIMIR:  You're merciless.  ESTRAGON: We came here yesterday. VLADIMIR:  Ah no, there you're mistaken.I suspect that this is almost certain to offend everyone.Ok, I'll bite - you scurrilous traitor!Yesterday was also Maundy Thursday - the day Judas betrayed (or did he?) Jesus.Today, of course, is the anniversary of Lincoln's assassination.I would not be surprised to learn that the Time Lords arealready contemplating a scheme akin to this one.One suspects the Time Lords have never seriously consideredany option that would preserve leap seconds in any form, to anytolerance, utilizing any scheduling algorithm - no way no how.==Educate, educate, educatebut can you explain your scheme in under 23 seconds?The ITU-R should openly publish the UTC specification.Boy!  On this Easter weekend one has to believe that thiswould require a miracle second only to the resurrection.The reality is that the ITU-R "specification" is just a minorfootnote pertaining to obsolete technologies of time signaltransport.  One presumes nothing would stop the IERS frompublishing any scheduling algorithm such as you describe.The IERS should enter into dialog with the International VirtualObservatory Association (IVOA) and the Internet Engineering Task Force(IETF) to determine an adequately robust scheme for guaranteeing thedistribution and availability of the XML documents which give theschedule of upcoming leap seconds.This was indeed one of the notional use cases for VOEvent( specification might benefit from dedicated IETF blessedsupport for leap second "aware" semantics, but it would be trivialeven using the current standard to represent such a schedule.The emerging VOEvent transport protocol(s) could certainlypropagate leap seconds as easily as reports of Supernovae,Gamma Ray Bursts and Earth-killer asteroids.To hear the detractors talk, a leap second is made to seem asdire an event as these cosmic catastrophes.As the five year prediction scheme goes into effect this will probablymean that there will be times when UT1 - UTC exceeds 0.9 seconds, butnot by very much.Cannot simultaneously constrain look-ahead and DUT1 tolerance.Suggest that rather than pick any fixed window, it is sufficientto define a mechanism and format for reporting the schedule -whatever it may be at any given time.  After all - what if thestate of the art improves enough to permit accurate ten yearpredictions?Simply allow the IERS to announce any number of leap secondsin advance extending over any time horizon - and yes - occurringat the end of any month.  If predictability is the goal, relaxingunnecessary constraints is the solution.Actually - one presumes the IERS currently has the authority todo both of these things.  Have never heard anyone suggest thatthe next two leap seconds might not be announced simultaneously.And the ITU-R has already signed off on the monthly schedulingof leap seconds - this is the law of the land.What precisely is stopping us from implementing some variationof Steve's scheduling algorithm right now, today, this minute?All in favor, say aye!I am still not convinced that there is any need for change.  I wouldlike to see further examples of why the status quo is not tolerable.I think even this gives the current "process" too much credit.There has not been a single cogent example of any negativeconsequence of leap seconds described in a convincing andwell characterized fashion.  Moaning and moping is not thesame thing as making a coherent argument.Surely someone could draft a two page white paper describingto some (any) level whatsoever of technical, scientific, economicor sociological detail exactly what dreadful things happen to someparticular system in the presence of a leap second?  And more tothe point, to provide evidence why we should believe the cure won'tbe worse than the disease.Rob SeamanNational Optical Astronomy Observatory

Re: 24:00 versus 00:00

2006-02-27 Thread Rob Seaman

On Feb 17, 2006, at 12:30 PM, Markus Kuhn wrote:

Clive D.W. Feather wrote on 2006-02-17 05:58 UTC:

However, London Underground does print 24:00 on a ticket issued at
midnight, and in fact continues up to 27:30 (such tickets count as
issued on the previous day for validity purposes, and this helps to
reinforce it).

The tickets of UK train operators are perhaps not good examples to
common standards practice, because they deliberately print them with
highly creative *non-standard* conventions, to make fake tickets
to spot for their staff.

We were actually using this for exactly the opposite purpose - to
infer NON-standard practice.  It's easy to design a system that
handles only idealized use cases and users who always live within the
bounds of rigidly described artificial rules.  The clock on the
wall may be the world's best example of technology that has to be
adaptable to a wide range of precisely these highly creative non-
standard conventions.


Re: 1884 IMC online

2006-02-20 Thread Rob Seaman
On Feb 19, 2006, at 1:35 PM, Steve Allen wrote:A few years ago Joseph S. Myers of Cambridge University went through the trouble of scanning a copy of the proceedings of the 1884 International Meridian Conference, and I put the TIFFs online have just been alerted that Project Gutenberg has just finished checking its transcription of the document into fully machine readable and searchable form.!I bet the ITU WP7A wishes the original rules applied to contributions from LEAPSECS: "Resolved, That the President be authorized, with the concurrence of the Delegates, to request an _expression_ of the opinions of the gentlemen invited to attend the Congress on any subject on which their opinion may be likely to be valuable."...but then, I wish we had Thomson and Newcomb on the list :-)Rob

Re: 24:00 versus 00:00

2006-02-17 Thread Rob Seaman

Clive Feather wrote:

London Underground does print 24:00 on a ticket issued at midnight,
and in fact continues up to 27:30

An even better example.  We cannot expect to dissuade such usage.
Deploying systems that require it be avoided is folly.  Wouldn't
think the modulus operator would be controversial.

John Cowan wrote:

Airlines in the U.S. [...] avoid arrival or departure times of
midnight so that there is no ambiguity

Again - an even better example.  The easiest way to conform to a
standard is to avoid challenging situations.  It is trivial for
airline schedules to avoid a brief window around midnight - say,
23:59-00:01, or 2 out of 1440 minutes per day.  It should be even
easier for NTP and other UTC transport mechanisms to avoid 2 minutes
out of 365+ days.

This isn't the solution to every challenge facing civil time - but it
sure simplifies the search space.

Rob Seaman

Re: Ambiguous NTP timestamps near leap second

2006-02-16 Thread Rob Seaman

On Feb 16, 2006, at 2:06 PM, Markus Kuhn wrote:

While there is a 24:00:00, there is certainly *no*
That would be 00:00:00.0001 instead.

Says who?  Didn't we just burn a lot of calories discussing whether
UTC was a real number or a continuous function?  Time does not end at
midnight, rather, a time without a date or a date without a time are
incomplete.  Are we to believe that there is something evil about
expressing negative time values, too?  Just depends on the epoch.

Put it in angle notation.  12 hours UT = 180 degrees = pi radians,
but also = -180 degrees or -pi radians.  So what?

Would think the usual tactics apply:  interpret rules as loosely as
possible on input and as strictly as possible on output.


Re: Ambiguous NTP timestamps near leap second

2006-02-16 Thread Rob Seaman
On Feb 16, 2006, at 4:46 PM, Warner Losh wrote:UTC rules state that the time sequence should be23:59:59.7523:59:60.023:59:60.2523:59:60.5023:59:60.7500:00:00.:00:00.25Well, no.  ITU-R-TF.460-4 says nothing whatsoever about the representation of time using sexigesimal notation:  "2.2 A positive leap-second begins at 23h 59m 60s and ends at 0h 0m 0s of the first day of the following month. In the case of a negative leap-second, 23h 59m 58s will be followed one second later by 0h 0m 0s of the first day of the following month (see Annex III)."Annex III contains two diagrams indicating the "dating of events in the vicinity of a leap-second", specifically "30 June, 23h 59m 60.6s UTC" and "30 June, 23h 59h 58.9s UTC", for a positive and negative leap second, respectively.  Note the explicit combination of a date and time to represent a specific moment in history (ignoring the lack of a year for these examples), not a time-of-day.  And further, each field of the date and time is expressed as a separate value.The problem is for time exchanges that cross the 0:00:00.00 boundary.It is the word "boundary" that is the problem.  The ITU recommendation discusses the dating of historical events, not how to divide one day from the next.  We're not discussing an issue with UTC, but rather with the NTP realization of same.Inventing a new notation that is not described in the ITU UTC definition is not helpful,even if lots of other people use it informally.Um - I wasn't discussing the notation as any sort of aid in resolving this problem with NTP.  I like the idea of the server simply failing to respond in the vicinity of a leap-second.That "lots of other people" use some feature most certainly is an issue when capturing the requirements for civil timekeeping.  Nobody is suggesting that NTP generate such timestamps - but the ITU cannot keep people from specifying time any way they want.  If two representations are congruent, why should our standards care?It will create confusion because it has no precise definition in the UTC standardAs we've seen, the "UTC standard" (really, the ITU recommendation for how UTC will be constituted in practice) does not address the representation of time at all.  Is this surprising?  Sexigesimal notation applies to multiple timescales, as well as to longitude and latitude and other spherical coordinates.(note: NTP specifically states UTC, and no other standard).Lots of standards reference other standards.  Lots of standards get it wrong.Rob SeamanNOAO

Re: An immodest proposal

2006-02-14 Thread Rob Seaman

On Feb 14, 2006, at 12:50 PM, Markus Kuhn wrote:

You can, of course, define, publish, implement, and promote a new
version (4?) of NTP that can also diseminate TAI, EOPs, leap-second
tables, and other good things. I'm all for it.

But why are you for it?  Before investing large amounts of time and
money in developing and deploying a large new timekeeping system,
wouldn't one want to invest smaller amounts in exploring the issues
and options?  Heck - one has to imagine that a number of successful
grant applications are lurking around here somewhere.  Time is an
issue that cuts across every funding agency out there.

I personally would very much prefer to see a protocol specification
that clearly indicates on the wire if something other than UTC is

Sure.  Sounds like a clear requirement for whatever system comes
next.  This is equivalent to the central problem with the leap hour
kludge.  If it ain't universal time - don't pretend it is.  Surely we
could devise some safeguards, starting with limiting the testbed to a
closed network of systems dedicated to timekeeping test applications.

Anything else sounds as dangerous to me as using the same kind of
plug in countries that use 115 V and 230 V power.

And yet you can buy dumb little conversion plugs that perform no
function other than allowing US electronic devices to fit straight
into European outlets and vice-versa.  It is the devices' power
supplies that are responsible for adapting to the voltage and line
frequency.  That you can't safely use the same plug for simple high
current devices like hair driers is just another one of those
inconvenient facts of nature.

What are the classifiable, quantifiable, facts of nature for
timekeeping in the real world today?

Naive users exist, and if things appear to fit together, they will
be plugged together by someone.

And that is one of the main points for building such a testbed
system.  What are the real risks?  What features are required to
mitigate those risks?  What level of naivete should be tolerated?
Users (meaning everybody, everywhere) are expected to master the
intricacies of sexigesimal notation - in both analog and digital
formats.  What is the appropriate level of timekeeping expertise that
can be relied on for various classes of user?

This is more likely to add to the problem than to the solution.

I'm not advocating a solution.  In the absence of additional data, I
expect I will never be moved from my current position of supporting
the UTC status quo.  As your comments imply, the alternatives are too
dangerous.  If any of us want to convince the other side (whichever
side that is) to change their minds, surely assembling hard data is
the first step.

Usability experiences gained in a testbed run by a small group of
knowledgeable enthusiasts do not necessarily scale into the real

No indeed, but what greater folly to suggest that omitting such
testbed runs might somehow build more confidence in proposed
solutions.  What is the opposite of knowledgeable enthusiasts?
Should these issues rather be left to ignorant apathetes?

Besides, bastardised NTP servers that replace UTC with TAI and
UT1 have been around for quite some some; for instance Patrick
Wallace (Rutherford Appleton Laboratory) reported at the 2003
Torino meeting about his UT1P server.

Well, yeah - I didn't claim my proposal was either new or rocket
science.  Pat has lead a lot of such efforts over his career.  The
whole point is to find a simple way to start to actually experiment
with and develop new ideas for timekeeping infrastructure.

Surely I'm not the only one who is dreadfully tired of hypothetical


Re: An immodest proposal

2006-02-14 Thread Rob Seaman
On Feb 14, 2006, at 2:28 PM, M. Warner Losh wrote:UTC time stamps in NTP are ambiguous.  TAI ones are not.Requirements should be kept separate from implementation.  Whatever the underlying timescale, certain external global requirements apply.  Whether NTP or some other implementation properly captures those requirements is a separate issue and should be treated as such.My understanding of the point you are making is that given an infrastructure like NTP that (more-or-less) assumes an even interval time scale, that UTC timestamps are indeterminate (to some level of precision) in the absence of a table of DTAI.  Of course, given an accurate UTC clock, TAI is indeterminate without that same table of DTAI values.UTC time stamps do not convey enough information to properly implement things like intervals, while TAI ones do.And TAI does not convey Earth orientation.  These are all well traveled issues.The NTPNG stuff that I've seen appears to consider these problems as worthy of needing a solution and they plan on solving them.Well, good!It isn't rocket science, but one has to divorce ones self from the chauvinistic view that UTC is always best.Chauvinism:  Prejudiced belief in the superiority of one's own gender, group, or kind: “the chauvinism... of making extraterrestrial life in our own image”.  Doesn't really seem that applicable...As has been repeatedly emphasized, astronomers are serious users of *both* UTC and TAI.  UTC is not always "best", rather different purposes require different time scales.  Many of us happen to think civil time - on planet Earth - should remain some variant of UTC/GMT.  That current implementations of the UTC standard have shortcomings is no surprise.For time exchange, it is not the best, and has many problems around the edges.Are there any time scales that don't require "time exchange"?  The key issue remains interval time versus Earth orientation.  It would be incorrect to assert that all timekeeping issues in computing devolve to a need for interval time.Doing NTP with TAI (and the implied requirement for DTAI) doesn't change what time is displayed for users.  It does make it *MUCH* easier to get leap seconds right for those users that need them.Perhaps.  Would be prudent to find out.  This is, however, only one aspect of timekeeping.Leap seconds aren't needed by users - they are required by UTC that is needed by users for various purposes.Anything else is madness.Now who is chauvinistic?Surely we could design a series of experimental programs to address these issues in an unbiased fashion.  Is this a controversial statement?Rob

Re: Comparing Time Scales

2006-02-04 Thread Rob Seaman

So internally, once we hit the leap second time, we step backwards.

Lovely waxing crescent moon last night.  I could see it out of the
north facing windows of our family room.  It might be simpler in some
sense to pretend that the moon (and the sun for that matter) always
rises due east and set due west, but there it was, shining in the
window because its declination last night happened to lie above the

Whatever time does, it does not step backwards.  It may be simpler -
in some sense - to build a clock or software interface that
accommodates the lunar correction to the solar day by stepping
backwards more or less contemporaneously with a leap second, but when
you get down to it, this is - well - a naive kludge.

If you freeze time, then doing a time exchange during a leap second
is going to give bogus results.

Time neither freezes...

If you step it back, then it will give good results, but there are
other bad effects.

...nor steps back during a leap second.  Just like the 29th of
February, a leap second is an extra unit of time that is interpolated
into the grand eternal sweep of history and causality.  Why not
design our computers, clocks and communications technology to
implement that simple fact - just like our calendars and palm pilots
recognize the fact of an extra day in February every four years?

Hey gang!  Let's put on a Gedankenexperimenten!  Imagine we are faced
with the prospect of a quadratically accelerating sequence of
negative, rather than positive, leap seconds.  (We'll ignore the
trifling anthropic dilemma of the moon looming ever closer in our
sky.)  Two questions to ponder:  How would we implement these?  And
would the ITU feel more driven - or rather less - to modify UTC as a

To address the first question, we can start with the statutory
requirement from ITU 460-4:

   2.2 A positive leap-second begins at 23h 59m 60s and ends at 0h 0m
0s of the first day of the following month. In the case of a negative
leap-second, 23h 59m 58s will be followed one second later by 0h 0m
0s of the first day of the following month

As we can see from the current discussion, system designers don't
necessarily even try to implement statutory requirements.  Notions
about freezing time or of stepping it back arise - it seems to me -
out of balking at the clear and explicit requirements of the task at
hand.  But in the case of a negative leap second, there would be no
additional wrinkle in time to iron out or to double over - just the
simple need to omit a second from the count.

If we were merely faced with omitting a second every year or two,
would the requirement really seem particularly onerous from the point
of Posix or NTP or our other technology?  Would GLONASS have noticed
the event even at the minimal level seen this New Year?  What is it
about a 61s minute that is deemed more herculean than a 59s minute?

We manage to deal with months containing 28, 29, 30, and 31 days.
The great majority of months (4497/4800) aren't even denumerable with
the length of our week.  Why then is a requirement that one minute
out of 800,000 accommodate one extra (or one fewer) second seen to be
such an imposition?  Especially when anybody who does find it so can
simply choose to use TAI instead?

Eppur si muove!

Rob Seaman

The nature of risk

2006-01-25 Thread Rob Seaman
 any conversion to Earth orientation
would have to be introduced explicitly.

Leap seconds are asserted to be a risk.  Does their lack present
fewer risks?  Prove it.

Rob Seaman

Re: The nature of risk

2006-01-25 Thread Rob Seaman

Leap seconds are asserted to be a risk.  Does their lack present
fewer risks?  Prove it.

No, you prove it.  Such rhetorical devices are designed to divide
and separate,

No, my rhetoric really isn't designed for that purpose.  And even if
it were so - how does that possibly undermine the idea that risks
should be explored before decisions are made?  Look before you leap
is not usually considered controversial.

rather than to understand the problems at hand.

The problems at hand have been rejected out of hand.  The initial ITU
position from six years ago has not budged an inch - simply cease
leap seconds.  Meanwhile, have significant issues been raised by the
recent leap second?  If so, no one has made them public.

On the other hand, we are simply to take it as a matter of faith that
not only do no communities outside of astronomy possibly care - but
that no possible negative outcomes can occur from blindly making such
a momentous change to the practice of timekeeping.

Imagine changing not just the definition of the meter, but the
underlying concept of length.  Wouldn't the governments of the
world first demand proof that vast infrastructures wouldn't topple?
Why then are the timekeepers so cavalier with the time they are
keeping?  If nothing else, one might imagine that the potentially
immense insurance liability would give them pause.

Other names for rhetorical devices are paragraphs and sentences.
I'll not apologize for knowing the difference between zeugma and


Re: the tail wags the dog

2006-01-24 Thread Rob Seaman

On Jan 24, 2006, at 12:50 AM, Peter Bunclark wrote:

I don't think Rob meant the above to be a complete course on

...although as a fan of Patrick O'Brian I am qualified not only to
teach navigation, but also the violin and Catalan.  You should see me
in a Bear costume.

Good example of a timekeeping decision made by a (very tiny)
minority over the majority.

The issue here is the meaning of the preposition over.  It is not
unusual for we anointed of Hephaestus (the god of dweebs) to be
placed in the position of making decisions for others - who may not
even be aware an important issue is being considered.  The more
fundamental question is whether the requirements of the majority are
properly considered and given appropriate weight.

How nice indeed, it would be, if the months were fixed to match

Quadratic despair still lurks, of course, since the month is
lengthening for exactly the same reason as the day.  Well, despair
would be lurking if we tried to match the length of the month (a
natural phenomenon) to an SI unit (such as the second).


Re: the tail wags the dog

2006-01-24 Thread Rob Seaman

On Jan 24, 2006, at 8:06 AM, Ed Davies wrote:

James Maynard wrote:

The problem is not that the SI second is not based on a natural
phenonemon (it is), but that the periods of the various natural
phenonema (rotations of the earth about its axis revolutions of the
earth about the sun, revolutions of the moon around the earth,
etc.) are
both incommensurate and changing.

Not to mention the hyperfine wibbles of caesium-133.

...and we wonder why our less technically oriented loved ones tune
out when we start to speak :-)

Point taken - these are all natural phenomena.  But then, so are all
the other issues we've ever raised.

The rotation of the Earth and the revolutions of its Moon are natural
phenomena we have little ability (and less reason) to attempt to
control.  Hyperfine wibbles are things that humans can hope to tame
in various ways.  No one disputes that our clocks have been improved
wondrously - but the point of a good clock is the point of other
technology, to tame nature in the service of mankind.  (Taming
mankind in the service of nature?  Hmm - there's a thought.)

The question on the table is whether mankind is better served by
gracefully accommodating the charming quirks of Earth and Moon - or
whether we should attempt to impose a metric standard, inappropriate
to the purpose.


Re: the tail wags the dog

2006-01-24 Thread Rob Seaman
On Jan 24, 2006, at 7:21 AM, Poul-Henning Kamp wrote:I think the crucial insight here is that geophysics makes (comparatively) lousy clocksThe crucial insight is that the Earth is not a clock at all, but rather the thing being timed.and we should stop using rotating bodies of geophysics for timekeeping.We already have.  The question is whether everybody on the planet needs to adjust their clocks to match requirements imposed by the high precision timekeeping needs of projects that already have options other than UTC.The follow-up question is whether attempting to ignore an inevitable outcome is better than simply dealing with it through the intervening centuries.  All proposals (other than rubber seconds or rubber days) face the same quadratically accelerating divergence between clock and Earth.Rob

worthy challenges

2006-01-24 Thread Rob Seaman
Ed Davies wrote:By "rubber seconds" you, presumably, mean non-SI seconds.  What do you mean by "rubber days"?  I'd guess you mean days which are divided into SI seconds but not necessarily 86 400 of them.Yes.  See for instance: a parochial note, ones suspects that the "astronomers" are seen by many as reactionaries about leap seconds.  The reality is that we're willing to consider dramatically deeper and broader changes than have been proposed.  The alternative to actually addressing the complete and complex needs of some situation, however, is often to continue to muddle through  in the mean time with an imperfect solution that has been inherited - and not coincidentally, that has been thoroughly tested.How is it possible that the precision timing community is advocating for simply throwing up their hands in disgust and attempting to sweep the whole thing under the rug for 600 years?  A difficult challenge is what engineers live for - the resulting triumph all the greater:"If I were to say, my fellow citizens, that we shall send to the moon,240,000 miles away from the control station in Houston, a giant rocketmore than 300 feet tall, the length of this football field, made ofnew metal alloys, some of which have not yet been invented,capable of standing heat and stresses several times more thanhave ever been experienced, fitted together with a precision betterthan the finest watch, carrying all the equipment needed forpropulsion, guidance, control, communications, food and survival,on an untried mission, to an unknown celestial body, and then returnit safely to earth, re-entering the atmosphere at speeds of over25,000 miles per hour, causing heat about half that of thetemperature of the sun [...] and do all this, and do it right,and do it first before this decade is out--then we must be bold."Our challenge is to build a system of timekeeping worthy of our children's children.Rob SeamanNOAO

Re: Approach to leap second discussion

2006-01-22 Thread Rob Seaman

I hope we can all continue this discussion in a more positive manner.

I'm of the opinion that messages on this list (no matter how
tricky :-) are always positive.  Timekeeping is a fundamental
issue.  It would be remarkable if there weren't diverse opinions.
Any negative aspects of this discussion are related to those who
don't choose to participate.  Which is to say, those who claim to
have decision making authority over UTC at the ITU, for instance.

The folks on this list appear to cluster into two groups (speak up if
your opinion diverges from both):

   1) Civil time should remain layered on UTC.  UTC should remain
largely unchanged.  Leap seconds should continue.


   2) Civil time should be layered on some flavor of interval time.
That timescale might be a variation of TAI called TI.  TI will not
have leap seconds.

The proposal submitted to the ITU is neither of these.  It is:

   3) Civil time should remain layered on UTC.  UTC should be modified
to no longer be a useful approximation to universal time.  Leap
seconds will be issued 3600 at a time.

You all know where I stand - but there are worlds of difference
between #2 and #3 as alternatives to #1.  All three proposals face
the same looming quadratic emergency.


Re: Risks of change to UTC

2006-01-21 Thread Rob Seaman

On Jan 20, 2006, at 10:17 PM, M. Warner Losh wrote:

Any watch that is smart enough to decode those signals would be
smart enough to add this minor correction as well.

A viable time scale could be constructed from any periodic (or near
periodic) waveform - there's nothing magic about the tick, tick, tick
of delta functions (or step functions if you prefer to think of it
that way).  A viable watch could present a different representation -
12 hour/24 hour, sexigesimal/decimal, local/universal - every time
it is consulted.  It wouldn't even have to be monotonic, as long as
enough temporal metadata is provided for context.  That metadata
(such as DUT1 and DTAI) could be provided in as circuitous and
obscure a fashion as can be imagined - encrypted, proprietary,
steganographically.  Heck - time signals might even arrive as little
blips on shortwave radio - as hard as that might be to believe.

There's viable - and then there's viable...

The mechanical watch might be a bit of a problem, but DUT1 doesn't
change enough to introduce navigation errors similar to what we
have today over the course of a year and can easily be looked up
like someone would lookup what the weather was going to be like.

...and we're back to the confusion between periodic and secular
effects.  There seems to be some thought that mean solar time is
nothing but a polite (or lately, sometimes impolite) fiction.
Greenwich Mean Time is real enough to have built the British Empire.
You're also working both sides of the equation.  A navigator observes
local apparent solar time onboard and compares it to GMT (or mean
time on any other known meridian) transported via chronometer.  DUT1
is a mechanism to correct mean solar time as reported by the clock.
The equation of time, on the other hand, is used to convert shipboard
apparent time to local mean time.  Subtraction does the rest.

Rob Seaman

Re: Risks of change to UTC

2006-01-21 Thread Rob Seaman

On Jan 21, 2006, at 12:03 AM, M. Warner Losh wrote:

WWV and most of the world's time stations broadcast DUT1.  I should
have added in my last message that some change in the signal format
would be necessary if the range of DUT1 exceeds 0.9s.

Bearing in mind that the ITU proposal would cease the reporting of DUT1.

I will note that the profile of high precision time users has
changed since 1972 when UTC was invented.  [...]  Should we
continue to tie our time up in knots because of a tiny minority of

Am fascinated by the failure of the precision timekeeping community
to perceive six and a half billion souls as users.  Shouldn't
choices related to international/civil/legal/business/historical time
be based on their needs?  No matter what the profile of high
precision time users has become - it is that entire community who
comprise a tiny minority.

How many celestial navigators are there today?

The Apollo astronauts relied on navigation by sextant - celestial
indeed.  One expects that future solar system explorers will
carefully continue to carry such fundamental instrumentation -
certainly for emergencies (think Shackleton, not just Magellan), but
also perhaps as an agile and reliable primary resource in their
toolkits.  Am not arguing that this has a direct connection to the
matter at hand - but rather that old doesn't mean obsolete.

Over the next 50 years, these two watches will be well within the
tolerance of most normal watches.

This interpretation confuses systematic effects (monotonically
diverging timescales) with random errors.  Much (one is tempted to
say, all) experimental science depends on abstracting trends from
noisy data.  No matter how large a tolerance one allows, the
diverging meanings of clock will eventually exceed it.

The approximation of civil time will be less than one minute off
during that time

A useful approximation captures asymptotic or otherwise limiting
behavior.  Where there is no limit, there may be an agreement to draw
a line in the sand - but there can be no approximation.

Einstein isn't right and Newton wrong, rather Newton's laws are
correct in the limit - the everyday limit.  High precision time
users may well place stringent requirements on fundamental
timescales.  But civil time requires a common sense everyday
compromise.  What this entire début de siecle discussion has been
about is whether ignoring the whole question for a few hundred years
is more common sensical than continuing to issue occasional small

If this is a real issue, the market will take over and produce
watches that have 'navigation time' and 'civilian time' at the
touch of a button

The market has not proven itself creative in meeting highly technical
needs.  Time and again, the market has converged on significantly
less than ideal solutions - Windows, VHS, internal combustion.  If
the magic hand of the market is the first law (conservation of
energy) of modern economic theory, the second law (entropy) is the
tragedy of the commons.

Timekeeping is pervasive in our society, but often invisible.  As
with the grand environmental challenges confronting this natal
century, market forces threaten to provide - oh so efficiently
provide - the wrong answers to timekeeping questions.

Of course, purely mechanical watches have other issues when used on
a boat.

And yet Harrison found a way around these when he invented the first
chronometers - for the express purpose of being used on ships.

Stating absolutely that UTC is not broken ignores these other users.

UTC is not broken.  We may agree or disagree on whether it meets
various civil or technical timekeeping requirements - but broken
would imply that it fails to meet its *own* requirements.  UTC is
eminently capable of continuing unchanged for many centuries - and
for millennia more with only slight changes.  After that, nothing yet
proposed (except for those danged rubber seconds) is any better (see

It would be the abandonment of leap seconds that would break UTC.
Lobbying to base civil time on some underlying timescale distinct
from UTC would be one thing.  Conspiring to emasculate UTC is quite

GLASNOS is a backup system to GPS that is not subject to DoD's
selective denial of signal.

Glasnost was Mikhail Gorbachev's policy of encouraging open public
debate, particularly in support of perestroika - restructuring - of
the Soviet economy.  On the other hand, GLONASS is the Russian Global
Navigation Satelllite System :-)  In any event, one suspects that the
Russians (or the FSU, even more so) would object to its being
characterized as a GPS backup.

Rob Seaman

Re: Risks of change to UTC

2006-01-21 Thread Rob Seaman

On Jan 21, 2006, at 10:11 AM, M. Warner Losh wrote:

Over generations, the problems with noon drifting to 1pm can
trivially be solved by moving the timezones that civilian time uses.

Neither trivial or a solution - quadratic disaster still looms.

Keeping universal time synchronized to an arbitrary meridian is
already arbitrary.

The prime meridian is conventional.  It is not arbitrary, rather
the choice was responsive to any number of political, social,
historical, etc. issues.

Implementing leap seconds in software is hard to get pedantically

Pedantically right is an interesting phrase.  A software design is
either right or it isn't.

Even many ntp servers on the net got the leap second wrong

And many got it right.  The world did not end.

astronomers and celestial navigators are being selfish

Pointing out pedantic facts of nature is unremarkable behavior for
either scientists or sailors.  Whether we're also selfish is immaterial.

Rob Seaman

Re: Internet-Draft on UTC-SLS

2006-01-19 Thread Rob Seaman
 bits to an enum containing the
timescale in question.

Like most strawmen, this won't survive through to the end of the
discussion, but it serves the purpose of priming the pump.  (I can
mix metaphors with the best.)

Assign different timescales very different
numeric epochs:
TAI:1972-01-01 00:00:00 UTC
UTC:MJD's epoch.
UT1:Flamsteads birthday ?
NTP:defined in RFC1305

Moving in the right direction.


Sufficient resolution to represent any likely physical
measurement or realizable frequency for the forseeable
future (13.8e-18 seconds resolution).

Any guess at likely physical measurements is going to fall short
for some purposes.  For one thing, one might want to represent
theoretical values in addition to experimental.  That said, you are
likely correct for our purposes.

Extracting the whole second part can be done by accessing
only the top 64 bits (which are enough to contain all
of history and then some).

I like this feature.

Conversion to/from NTP timestamps is trivial.

Conversion to time_t is a matter of addition and extraction
of the relevant 32 bits.

The binary format makes for fast and efficient arithmetic.

By assigning the UTC timescale an identifier of zero,
the majority of implementations can disrecard the
multiple timescale aspect in total.

Small platform implementations can use a smaller width,
for instance 64 bits split 48/16 and easily transform
to standard format by zero extension.

High quality implementations will check the bottom 8 bits
for identity and fail operations that mix but don't match

Different epochs will make it painfully obvious when people
mix but don't match timescales in low quality implementations.

These are all interesting goals that might be polished into
functional requirements.

Now, please show some backbone and help solve the problem rather
than add to the general kludgyness of computers.

Do you find this tone of voice productive when collaborating?  :-)

It seems to me that we're discussing apples and oranges again.
Whatever the representation of time values - whatever the underlying
standards - there have to be mechanisms for implementing same on
particular platforms.  I haven't heard any claim that UTC-SLS is
intended to serve all needs.  Rather, we've heard the opposite.
Suspect I'm not alone in being suspicious of any overreaching
solution proffered for all timekeeping situations - sounds like the
definition of a kludge.

Rob Seaman

Re: Internet-Draft on UTC-SLS

2006-01-19 Thread Rob Seaman

On Jan 19, 2006, at 10:02 AM, Rob Seaman wrote:

How delightful!  A discussion about the design merits of actual
competing technical proposals!

Apologies for failing to credit the quotes from Poul-Henning Kamp.

Risks of change to UTC

2006-01-16 Thread Rob Seaman

On Jan 15, 2006, at 1:07 PM, John Cowan wrote:

There are a lot of systems, it seems, that assume DUT1 is bounded
by either 0.9s or 1s.  If leap seconds are turned off, then I'd
expect that these will break and be replaced by systems that assume
DUT1 is unbounded.

Ah.  I see.  You are focusing here on explicitly Y2K-like risks.
Those are indeed an issue, but I agree not a large one for most
constituencies.  Only a minority (small minority, one would think) of
systems currently include any DUT1 correction at all - although these
will perhaps tend to be the most safety-critical applications.  As
with Y2K, all systems have to be inventoried for potential risks, not
just ones you know about in advance.

That is, of course, one of the major issues for astronomers - we rely
on UTC providing a 0.9s approximation to UT1 and most of our systems
don't use DUT1.  Even our high precision applications (in either
interval or universal time) don't tend to require conversions other
than as a preprocessing step.  Remediating our systems for such a
fundamental change to UTC would involve much larger changes than Y2K
did - algorithms and data structures would have to change, not just
the width of some string fields and sprinkling some 1900's around.
(I know that oversimplifies Y2K - suspect virtually everybody on this
list was intimately involved with their organization's Y2K
remediation effort.)  Also, standalone applications would have to
become network aware to have access to externally derived tables of

Astronomers might be unusual in needing to introduce DUT1 into our
systems (on a short schedule for a large expense) should Sauron win
and the nature of UTC change, but we wouldn't be alone.  And as clock
time diverges further and further from solar time, more systems in
more communities (transportation, GIS, innumerable scientific
disciplines, what have you) would be revealed to need remediation.
That's where I was coming from.

Another issue is interoperability.  There is the thought, perhaps,
that merely by deciding to cease issuing leap seconds that all clocks
on the planet will automatically converge on TAI (+ inane constant +
local offset).  Some systems (like those in astronomy) will converge
on universal time.  Would think it unwise to simply assume that no
significant risks will be revealed as the dual timescales diverge.
We all can likely agree that many deployed systems (of whatever
nature) are naively configured.  Is this likely to change overnight?

Rob Seaman

Re: Problems with GLONASS Raw Receiver Data at Start of New Year

2006-01-15 Thread Rob Seaman

On Jan 14, 2006, at 8:59 AM, Richard Langley wrote:

The problem existed for only 2-1/2 minutes, not hours.

Thanks for the clarification.

Might be coincidental with the leap second but I've not noticed
this problem at other times.

Would be a significant coincidence.  Any simple explanation for the
90 second lag in the issue being triggered?  The latency associated
with emergent behavior is of interest in itself.

Stations were not running during the previous leap second.

Right - that's a central fact influencing policy.  That different
groups take this to imply that different choices should be made adds
a little spice to the discussion :-)

UNB1 Web page is here:
IGS Central Bureau Web page is here:

Thanks for the pointers.

Rob Seaman

Re: Monsters from the id

2006-01-14 Thread Rob Seaman
On Jan 13, 2006, at 12:46 AM, John Cowan wrote: In the end, it will be impossible to maintain the notion that a solarday is 24h of 60m of 60s each: we wind up, IIRC, with the solar dayand lunar month both at about 47 current solar days. There's a lot of difference between what happens over a billion yearsand a million years.  Length of day increases only about 20s per millionyears.  Should we be here to care in a million years, only a 1/4 of 1/10 ofone percent tweak to the length of the "civil second" would suffice to allowour Babylonian clock paradigm to continue in use.  Alternately, we mightdecide to add one second to just one minute out of each hour.I won't claim one of these would be the choice.  There are manifoldoptions for representing time.  But I do assert that our descendants - foras long as they may be regarded as human - will desire to have somecommon way to represent fractions of a day.  And no matter whatrepresentation they choose, they will still face the quadraticaccumulation of leap seconds or their equivalent.Far from being a motivating factor for deprecating leap seconds, thequadratic clock lag resulting from the roughly linear tidal slowing ofthe Earth is precisely the strongest argument for preserving meansolar time as our common basis for timekeeping now and forever.Besides, it is simply a charming fact of life in the solar system thatour Moon is receding while the Earth spins down.  Apollo era laserretro-reflectors show that for each second our day lengthens, theMoon's orbital radius grows by a mile or so.Time is a fundamental element of all that we do.  Surely publicpolicy should not be governed by a drab and dystopian vision ofa fragmented planet scrabbling randomly to keep our disjointclocks aligned.The simplest - nay, the only - way to keep our clocks synchronizedone to the other is to keep them all tied to Mother Earth. "You think the Earth people think we're strange you think."Rob SeamanNOAO

Re: Report of Leap Second Problem with GPS Data

2006-01-14 Thread Rob Seaman

On Jan 13, 2006, at 6:26 AM, Richard Langley wrote:


Thanks!  Actual reports from the field, how novel!


IGS Station Mail  12 Jan 14:59:42 PST 2006  Message
Number 760

Author: Michael Moore

  Geoscience Australia
Australian Regional GPS Network
  Geodetic Operation


High rate data, 1Hz 15 minute files, from the ARGN suffered a software
problem due to the recently introduced UTC leap-second. Data from
DOY 001 to
DOY 009 is 1s off in the timestamps reported n the RINEX files.
This problem
only applies to the 1Hz 15minute files submitted from the ARGN. The
problem has been fixed, and all files from DOY 010 is reporting the

RINEX headers for DARR from DOY 009, was incorrectly reporting an
height of 0.000. The headers have now been fixed to report the correct
antenna height of 0.0025, and the data from DOY 009 has been
resubmitted with
the correct header information.

I won't claim to know the intrinsic importance attached to this.
Critical systems may depend on the information.  But is it fair to
sum up the situation by saying that a leap second triggered a couple
of bugs (or perhaps one common bug), they were detected, have been
fixed, and affected data products have been remediated?  Also, it
appears that some other data products were unaffected?

So, the issue has been resolved - would likely have been resolved
sooner if a leap second had occurred earlier - and is no longer
directly pertinent to a discussion of future leap seconds?

Well done, Geoscience Australia!

Rob Seaman

Re: Problems with GLONASS Raw Receiver Data at Start of New Year

2006-01-14 Thread Rob Seaman

On Jan 13, 2006, at 7:51 AM, Richard Langley wrote:

The International GNSS Service (IGS) includes a sub-network of
operating GLONASS monitor stations (about 50) including one at the
of New Brunswick (UNB1).  At UNB1 we lost C1 (coarse code on L1
P1 (precision code on L1), and P2 (precision code on L2)
observations on the 5
GLONASS satellites we were tracking at 00:01:30 GPS Time on 1
January 2006
along with phase jumps in L1 (carrier phase on L1) and L2 (carrier
phase on

Perhaps you can expand on the meaning of all this.  Presumably this
would represent an infrequent occurrence?  What are the implications
for downstream systems?  For that matter, what systems lie downstream?

Code measurements were back at 00:04:00.

So the problem extended for 2.5 hours from 00:01:30 - 00:04:00 GPS
Time?  Were there repercussions that have persisted after this?

I have just learned from one of the IGS analysis centres that all
January 1
IGS GLONASS observation files that they checked show a similar

The leap second has not been mentioned, but presumably we are to
infer that it triggered this behavior?  Would be absolutely delighted
to learn more about the IGS, both in general and to provide context
for interpreting this report.

As with the previous mail, I won't claim to be able to attach an
estimation of the importance of the events described.  We obviously
all believe leap seconds are worthy of discussion or we wouldn't be
here.  I presume many of us read RISKS Digest and can dream up scary
scenarios.  But there are also risks associated with *not* having
leap seconds, with allowing DUT1 to increase beyond 0.9s, for
instance.  And events triggered by those risks would not draw
worldwide scrutiny - they could occur year-round and the media circus
would have moved on.

Rob Seaman

Re: Report of Leap Second Problem with GPS Data

2006-01-14 Thread Rob Seaman

This goes counter to my claims so it is of no importance.


This time, there were no reports of death with the leap second,
therefore they can't be too bad... :-)

I invite derision with my flights of rhetoric.  But this is an
internet forum and a little leeway may be warranted.  We all have our
day jobs with more pragmatic requirements.  For whatever reason, UTC
is of importance to each of us - both the immediate day-to-day issues
as well as the long term philosophical issues.

Reports of significant misbehavior triggered by the leap second are
to be expected.  Honestly, I am surprised that there have been so few
so far - but perhaps two weeks is about the right time for data to be
gathered and turned into a report.  I won't belabor the notion that
the solutions to any problems revealed in these reports might indeed
be expected to be a little more subtle than never issue another leap

But let's imagine we were to identify a consensus vision for the path
forward.  (Seems a bit unlikely at the moment :-)  So all the
interested parties would be in agreement on the changes to be made to
UTC (and/or TAI and/or whatever else) - and in agreement that any
changes were needed at all.  Again - just for the sake of argument.

There would still need to be an implementation plan.  That plan would
need to analyze risks (and benefits) and costs.  It would need to
reveal a schedule, likely in stages over many years.  If you believe
there are significant risks associated with the complex system
involved with the issuance of leap seconds, are significant risks not
to be expected with making changes to that system?

We're all concerned about risks.  Unplanned changes to deployed
systems are among the most dangerous.


Re: Problems with GLONASS Raw Receiver Data at Start of New Year

2006-01-14 Thread Rob Seaman

On Jan 14, 2006, at 11:20 AM, John Cowan wrote:

I'd expect to see a wave of breakage as DUT1 exceeded 0.9s for the
first time, and a second wave as it exceeded 1s for the first
time.  After that, of course, the problems would no longer be
relevant.  :-)

I haven't been able to decipher what the humor is meant to be here -
will gladly admit that this is likely a failure on my part.  I won't
ask you to explain the joke, but rather I suspect you had a more
basic point you were seeking to make.  Is there some reason that
risks resulting from a diverging DUT1 can be expected to be mitigated
(even in part) as it grows past 1s?

If we discount the melodramatics of this list (will gladly admit to
being one of the offenders), we are left with a rather interesting
technical discussion of how best to deliver both interval time and
earth orientation information to a variety of classes of users.
Whatever the context of that discussion and of our own personal
points of view, one large element of confusion is precisely that TAI
mimics UTC mimics GPS mimics...  We might more easily recognize the
distinct character of each time-scale if neither their values or the
representations of those values permitted them to masquerade one for
the other.

Atomic Time is naturally represented as an unending count of
seconds.  Universal Time is naturally represented as a fraction of a
day (equivalent to an angle).  It is that naive heuristics exist that
claim to convert Atomic Time into fractional days - or alternately,
to convert Universal Time into a count of seconds - that creates
confusion between the two.

Rob Seaman

Re: The real problem with leap seconds

2006-01-13 Thread Rob Seaman

I'm glad to see such active traffic on the list - particularly
discussions such as this that are wrestling with fundamental concepts.

   On 2006-01-13, Mark Calabretta wrote:

 The point is that UTC is simply a representation of TAI.

On Jan 13, 2006, at 4:17 AM, Michael Deckers wrote:

I believe I'm now grasping what you mean:

Have spent many hours wrestling with standards documents written by
M. Calabretta.  The key to understanding what they mean is to
carefully read what is on the page.  Simply a representation is not
an editorial comment like, say, just a representation might be
taken to be.  Representations are important, too.  It is - simply -
not accurate to suggest that UTC is discontinuous at a leap second.
DST is indeed discontinuous twice a year, but the underlying standard
time never is.

Astronomers who write UTC as a real (eg, in JD or MJD notation)
want an approximation of UT1 to point their telescopes, they do
_not_ want TAI.

Astronomers want various flavors of time for various purposes.  It is
indeed exceptionally useful to be able to rely on simple closed form
calculations relating various quantities to universal time, where
UTC is often a sufficiently accurate approximation.  It happens that
we also use TAI and other interval time scales for many things.  And
it turns out that our images and other data products often benefit
from the use of civil time stamps for which good old reliably
continuous UTC serves admirably.  Ultimately it is the risk to this
last category of use cases that concerns me most about the notion of
leap hours.

That said, I suspect M. Deckers and M. Calabretta could productively
collaborate on a document synthesizing the fundamental issues into a
common vision.  Or perhaps somebody is aware of such a document that
already exists?


Re: The real problem with leap seconds

2006-01-13 Thread Rob Seaman

On Jan 13, 2006, at 8:05 AM, Ed Davies wrote:

MJD 27123.5 means 12:00:00 on day 27123 if it's not a leap second
day, but what does it mean on a day with a positive leap second?

And we're back to the point in question.  The precise issue is the
definition of the concept of a day.  A leap second itself is a
clearly defined moment in time.

Re: Monsters from the id

2006-01-12 Thread Rob Seaman
On Jan 12, 2006, at 12:36 AM, John Cowan wrote:No one, at least not on this list, is arguing for an alignment of theabsurd leap hour proposal (henceforth ALHP) with DST changes.I went rummaging through the ITU proposal and back as far as Torino.  Found this comment from a LEAPSECS thread on 28 July 2003: At Torino the proponents of omitting leap seconds supposed that the governments of the world might handle this situation using leap hours introduced into civil time by occasionally omitting the annual ``spring forward'' change to jump to summer/daylight time.This particular quote originated with Steve Allen's excellent page: couldn't find any explicit mention of this in the discussions at Torino, but Steve must have gotten it somewhere - and as you say, not from the list.  It may be an opportune time for folks to reread the presentations from Torino: example, I found this interesting tidbit from the Russians: "This is to inform you that to our opinion it is necessary to preserve the status-quo of the UTC time scale."Considering GLONASS is always trotted out as the only explicit example of a system that fails to handle leap seconds, this seems significant somehow.More-or-less the entire text of the proposed change to ITU-R TF.460-6 is expressed here:Operational rules(after  UTC 21 December of the transition year)1   ToleranceThe difference of UT1 from UTC should not exceed ±1h.2   Adjustments to UTC2.1Adjustments to the UTC time-scale should be made as determined by the IERS to ensure that the time-scale remains within the specified tolerances.2.2The IERS should announce the introduction of an adjustment to the UTC time-scale at least five years in advance. At the time of the announcement the IERS should provide directions regarding the details of the implementation of the adjustment.2.3All operational rules and nomenclature prior to  UTC 21 December of the transition year given above no longer apply.NOTE 1 – The broadcast of DUT1 will be discontinued.NOTE 2 – Predictions of the Earth’s rotation currently indicate that such an adjustment would not be required for thousands of years.Note the inaccurate and self-serving "thousands of years" that is corrected to 500 years in the draft.  There isn't the slightest specification (or analysis) of how a leap hour might be implemented - just an assumption that the IERS will persist indefinitely.  We're certainly aware that "all operational rules" are to be changed - but what about the nomenclature?  Imagine changing an ISO or SI standard - preserving a trail of coherent nomenclature would be half the document.  And then, of course, the amazing fact that the document simultaneously increases the importance of DUT1 by orders of magnitude, while discontinuing its issuance.  This "proposal" is not only ill considered, it is simply - well - lazy and arrogant.We already have that repeated time sequence and gap in much of the world,and live with it.  These repetitions would be no better and no worse;when a gap is present, the local sovereignty can omit the gap, but thisis not a necessary feature of the proposal.The point I was trying to make is that you can't simultaneously omit the overlaps/gaps and preserve anything even vaguely resembling the familiar relationship between our clocks and the solar day.  It doesn't matter whether we continue an international civil time system or abandon it for local anarchy - people everywhere in the world would  have to deal with the repercussions.  That the situation will degrade slowly over a few hundred years before collapsing catastrophically doesn't really seem to recommend the plan.It may not sound like it, but I am willing to be convinced otherwise - but you'll have to do a lot better than rivaling the scant length of the ITU proposal.  How about a detailed scenario of exactly how you see this working for a couple of neighboring but distinct local timezones?  What is the precise mechanism that might be used?The subtext of both your position and the "absurd leap hour proposal" is that civil timekeeping is so trivial that everybody from barbers to burghermeisters should be encouraged to make public policy - after all, these aren't "important" scientific and technical issues.  Rather, civilian users deserve as good or better a timescale as the technical users (who ultimately can take care of themselves). Historians already deal with the discontinuity between Julian andGregorian calendars, which was similarly conducted in a decentralizedfashion between 1582 and 1924.That there was a global mess several hundred years in the past is not a particularly good reason to generate another global mess several hundred years in the future.Aliens?  Us?  Is this one of your Earth jokes?Rob SeamanNOAO

War of the Worlds

2006-01-11 Thread Rob Seaman

I see Steve Allen has already supplied a thorough answer.  Interested
individuals might also scrounge through the list archives (http:// since the topic has come up
before.  In fact, Demetrios Matsakis speculated on solar system wide
timescales even before this list was started.  My own skepticism over
more extreme flights of fancy is expressed under Future Directions

On Jan 11, 2006, at 7:01 AM, Daniel R. Tobias wrote:

If, however, this Martian second is actually defined as a
particular multiple of the SI second, then the use of leap seconds
on Mars would ultimately be necessary to account for any future
changes in the length of the Martian day.

That is indeed the issue.  Is there anybody from the geophysics side
who can comment on long term trends in Martian length of day?

I don't have an envelope large enough, but there are various issues
to consider.  The Hurtling Moons of Barsoom are much smaller than our
own and should have a negligible tidal breaking effect.  (See http://, for instance, for their
interesting history.)  And do the Earth's oceans mediate our Moon's
breaking or is that a crustal phenomenon?  (The Earth-Moon system
should better be regarded as a double planet, than planet and
satellite.)  On the other hand, Mars passes much closer to Jupiter,
the 800 pound gorilla of the solar system, but then it is further
from King Kong - the Sun, that is - and tides are an inverse cube
effect.  But Mars is much smaller and has a smaller moment of inertia
in the first place - but then Mars is much smaller and the lever
arm to grapple with it is less pronounced.

Taken all together, one suspects that LOD(Mars) is many orders of
magnitude more constant than LOD(Earth).  One would not be
flabbergasted to be utterly wrong, however.

Hopeful news for John Carter fans:


Monsters from the id

2006-01-11 Thread Rob Seaman
What now, Dr. Moebius?                      Prepare your minds for a new scale...                    of physical scientific values, gentlemen.Mark Calabretta takes the lazy man's way out and appeals to facts: Here in a topology-free way is what the axis labels of my graph looklike during the said leap second insertion:            UTC axis                    TAI axis                 DTAI       2005/12/31 23:59:58         2006/01/01 00:00:30            32       2005/12/31 23:59:59         2006/01/01 00:00:31            32       2005/12/31 23:59:60         2006/01/01 00:00:32            32                        60.9                        32.9          32                        60.99                       32.99         32                        60.999...                   32.999...     32       2006/01/01 00:00:00         2006/01/01 00:00:33            33       2006/01/01 00:00:01         2006/01/01 00:00:34            33The seconds keep step and the graph has no gaps, jumps or kinks.Now let's look at a leap hour introduced as an extra "fall back" hour:  UTCTAI  2600-12-31T23:59:58   2601-01-01T00:00:31  33  2600-12-31T23:59:59 2601-01-01T00:00:32  33  2600-12-31T23:00:00 2601-01-01T00:00:33  33  2600-12-31T23:00:01 2601-01-01T00:00:34  33 (?)  ... ... 2600-12-31T23:59:58  2601-01-01T01:00:31 33 (?)  2600-12-31T23:59:59 2601-01-01T01:00:32  33 (?)  2601-01-01T00:00:00 2601-01-01T01:00:33  3633I chose to introduce the leap hour on December 31 - I don't believe the proposal indicates the date for doing so.  Folks have been tossing around the notion of aligning this with daylight saving time - but DST in what locality?  Does anyone really believe that a leap hour would be introduced on different calendar dates worldwide?  (It seems to me that the one time it is guaranteed NOT to occur is during a daylight saving transition.)Not satisfied with the ITU position that UTC should merely be emasculated to correspond to TAI - 33s - Nx3600s (which, of course, really has the effect of ensuring that TAI itself will remain a completely irrelevant mystery to the public), some would completely eliminate UTC from the equation (or is it that they would eliminate TAI?) Something like: GMT   TAI  2600-12-31T23:59:58   2601-01-01T00:00:31   2600-12-31T23:59:59  2601-01-01T00:00:32   2600-12-31T23:00:00  2601-01-01T00:00:33   2600-12-31T23:00:01  2601-01-01T00:00:34 ... ...  2600-12-31T23:59:58   2601-01-01T01:00:31   2600-12-31T23:59:59  2601-01-01T01:00:32   2601-01-01T00:00:00  2601-01-01T01:00:33  But we're to believe that this would be implemented as an omitted "spring forward" hour - ignoring the fact that many localities don't currently have this option because they don't use DST at all - can't omit what you don't have in the first place.  Well - fine, a "spring forward" event might look like: GMT   TAI  2600-12-31T23:59:58   2601-01-01T00:00:31   2600-12-31T23:59:59  2601-01-01T00:00:32   2601-01-01T01:00:002601-01-01T00:00:33   2601-01-01T01:00:01  2601-01-01T00:00:34   2601-01-01T01:00:02   2601-01-01T00:00:35 But under this interpretation we're to believe that the very notion of international civil time is anathema (except perhaps for TAI with some oddball persistent 33s offset and either a one hour gap or one hour repetition every few hundred years).  What this means is that *local* civil/business/legal time contains this gap or this repetition.  I suspect we can agree that the civilians/businesspersons/lawyers won't care whether the issue is local or not, all they are going to see is a repeated time sequence or a gap - and with no possibility of appeal to standard time, because standard time as we know it simply won't exist anymore.And historical time?  Well, historians will simply have to get with the program.  Suck it up.  Perhaps loudspeakers will announce the arrival of the leap hour (or leap timezone migration event) with the admonition to refrain from historically significant activity for the space of one hour.  (This announcement would be unnecessary in the Washington, D.C. city limits, of course.)And more to the point, since international time is a fiction, this gap/overlap in civil/business/legal/historical time would occur twice 

Re: MJD and leap seconds

2006-01-10 Thread Rob Seaman
On Jan 10, 2006, at 9:17 AM, Peter Bunclark wrote:On Tue, 10 Jan 2006, Poul-Henning Kamp wrote:In message [EMAIL PROTECTED], Peter Bunclark writes: Good grief.  MJD is used widely in astronomy, for example in variablility studies where you want a real number to represent time rather than deal with the complications of parsing a date.So how do you deal with fractional days in that format ? with decimals.I'm not one to shy away from irony (see!  just proved it again...), but I do think there is a real issue here.  Was interested to read the pages Tom pointed us to.  Both the IAU position and McCarthy's exposition of same are curiously silent about the issue of resolving ambiguities resulting from non-denumerable SI intervals and solar days.The IAU tells us:1. Julian day number (JDN)The Julian day number associated with the solar day is the number assigned to a day in a continuous count of days beginning with the Julian day number 0 assigned to the day starting at Greenwich mean noon on 1 January 4713 BC, Julian proleptic calendar -4712.2. Julian Date (JD)The Julian Date (JD) of any instant is the Julian day number for the preceding noon plus the fraction of the day since that instant. A Julian Date begins at 12h 0m 0s and is composed of 86400 seconds. To determine time intervals in a uniform time system it is necessary to express the JD in a uniform time scale. For that purpose it is recommended that JD be specified as SI seconds in Terrestrial Time (TT) where the length of day is 86,400 SI seconds.Which is to say that day number is (always) a solar unit and fraction of day (sometimes) an SI unit.In "practical" terms, a JD(TT) _expression_ would simply be calculated by running a count of TT seconds since some epoch through the obvious conversion mill, but we're then returned to the central issue of reconciling such a JD(TT) with a JD(UT1).  A calculation would simply show a growing fractional difference between the two, of course.  At issue is the unit jump in JDN.  Which day is it?  This ambiguity only holds for a bit over a minute a "day" in the current epoch.  (UTC = TAI - 33s, TT = TAI + 32.184s) The ambiguity is growing.Perhaps the SI unit should have been called the "essen", rather than the "second", as Steve Allen has said.  But whatever it is called, it has a clear definition.  But what is the definition of a day?  Am convinced we need to reach a consensus on this before leaping (irony again) into any changes to the current rules of civil/business/international/legal/historical date and timekeeping.You'll note that I omitted "technical" and "scientific" from that list.  This is not now and has never been a discussion about resolving purely technical issues, although some of the implications strongly affect technical people.Rob

Re: interoperability

2006-01-09 Thread Rob Seaman

On Jan 9, 2006, at 12:06 AM, Poul-Henning Kamp wrote:

You yourself defined stage one as TAI with some constant offset
yourself, you can't change definition in the middle of the discussion.

I was attempting to describe your position.  In point of fact, I
agree with Tom Van Baak:

You cannot divide timekeeping, time dissemination, into neat stages.

What we can do, however, is layer our standards upon a coherent
vision of the requirements placed on timekeeping by the wide range of
activities engaged in by humanity.  Talking to other humans aside
from the 114 members of this list might be a good first step.

I've never been in favour of the leap-hour proposal as other than a
political instrument to be abandonned well before the clock strikes.

Just wanted to re-emphasize your position.  Considering that you and
I,  the polar extremes of this issue, both reject the notion of leap
hours, perhaps we can find something else to talk about?

Not adjusting the clock is less disruptive than doing so, no matter
which half of the year.

Won't repeat my arguments a third time.

They have 600 years to find a solution and an implementation date
for it.

Who is this they you're talking about?  We're discussing changing a
standard that will be in effect now, then, and all times in between.
Our descendants won't be appreciably smarter than we are, and they
won't have access to insights regarding fundamental public
timekeeping issues that we don't have.  If we cannot posit a solution
more creative than forget the whole thing (which I continue to
assert is not an option, outside of extremely dark post-apocalyptic
science fiction tales) then neither will they.


Re: interoperability

2006-01-09 Thread Rob Seaman

On Jan 9, 2006, at 12:23 AM, John Cowan wrote:

This is like the day is light and night is dark statement: there
is, at any given location, one and only one sunrise per (solar)
day, no matter what clocks say.

Communication prospers when people's clear meaning is not subjugated
to petty grammarians.

We are now - and have been - discussing timekeeping changes that call
into question the definition of a day.  Those of us who support
solar time are fundamentally asserting the primacy of the the
standard day over the standard second (for civil timekeeping
purposes).  Those of us who consider solar time to be a curious
anachronism, assert the the SI second over the concept of a day (for
civil timekeeping purposes).

As I've pointed out before, future times in legal documents are
defined as LCT for a particular place, since the future mapping
between LCT and any other time scale is not known.

At the risk of igniting a new round of stage two nonsense, consider
the implications of your statement.  Currently LCT (as you appear to
mean it) is standard time.  Daylight saving (under whatever name) is
merely an overlay on standard time.  Standard time has no jumps
(except for leap seconds).

Under your suggestion, LCT would include the jumps for daylight
saving time (if locally used) as well as the jumps to correct for the
cumulative effect of tidal slowing.  As I hope I have established,
these are fall back discontinuities that would result in the same
hour of LCT occurring twice.  Is this not perceived to be a problem?


Re: interoperability

2006-01-09 Thread Rob Seaman

On Jan 9, 2006, at 1:01 AM, Clive D.W. Feather wrote:

We go through such discontinuities twice a year in most years.

Only the uninteresting daylight saving jumps.  UTC remains without
discontinuities above the level of a leap second.  If UTC weren't
equivalent to what I call civil time, the ITU wouldn't be making a
fuss to change it.

Except that time zone shifting means you don't affect the UTC

Only because you would redefine UTC to be equivalent to TAI.

The proposal is simply to have this jump abolished, so that the UTC
meridian starts drifting around the earth.

Glad to see somebody admit that this is one of the issues in play.
Perhaps we might now bring the cartographic community inside the
firewall and clue them into what is being proposed?  Note again that
the implications of this are not somehow to be embargoed for 600
years, but rather would apply immediately and at all times between.


Re: interoperability

2006-01-09 Thread Rob Seaman

On Jan 9, 2006, at 1:22 AM, Clive D.W. Feather wrote:

At some point, probably around the time that we're seeing an hourly
shift every year, people are going to have to divorce second from
day, or at least re-negotiate the terms of engagement.

By what magic do we believe the issues involved will become more
tractable at some point in the future?

How precisely does one divorce the definition of the day from that of
the second?  What is a clock if not a device to slice days into
seconds?  The fundamental problem is that the second is defined
against one underlying concept of time and the day against another.
As such, there are only three options:

   1) redefine the day
   2) redefine the second
   3) occasionally reset the clock

The only one of these that doesn't beg for a truly vast amount of use
case and requirements analysis is #3, the status quo.  I suspect most
of us would be happy to pursue the research needed by either or both
of the first two options.  How much more interesting than letting our
pasty complected cave dwelling descendants have all the fun!


Re: predicting leap seconds

2006-01-08 Thread Rob Seaman

On Jan 7, 2006, at 11:01 PM, M. Warner Losh wrote:

This would phase in the predictive timeline for leap second
insertions, and would also give the IERS control to end the
experiment if the time horizons exceeded their ability to predict
with confidence.

it would also be completely within the current UTC specification and
practices.  The various bulletins are required to be released with a
minimum look-ahead schedule.  No particular reason they might not
issue a bulletin every six months including both scheduled leap
seconds and unscheduled predictions in a sliding window extending
forward a decade or more.

  1   2   >