[LEAPSECS] QB213 .R4 2013

[Eric Fort]

might there be a bit more simplicity added to this discussion.....  It
would seem to me that what is and is not a "clock" is not and should
not be the question.  A clock tells time, whatever that is.  Planetary
rotation, Planetary orbit, A pendulum, A quartz crystal, or a cesium
beam - none of these are "clocks" they are but frequency sources that
serve up a periodic function from which an interval is derived and
those intervals or in some cases portions thereof may thus be counted,
incremented, and/or displayed.  The question is not what is or is not
a "clock" or at least it shouldn't be as we can deduce an interval
from any periodic oscillator, the mere fact that it's periodic making
it a frequency source and vice versa, but what is time itself?  Is
time an incremental count of intervals based upon some frequency
source which we have thus defined to be exact and unchanging in it's
periodic function and is thus exact by definition or is "time" the
position and orientation of a planetary body such that a given time
indicates things like the season and where the sun or stars present
themselves in the sky at a given moment?  It seems "time" is different
things to different users of it and their use at the moment.  Is this
not why we have different timescales?  Maybe it's time for the minders
of astronomical periodicity and the minders of atomic periodicity to
simply agree to disagree about what "time" is at it's core and simply
use the timescale that is appropriate and useful for their own use.
As for those simply going about their daily lives and wishing to make
a schedule within seconds or even minutes I think the mass public
would probably like to see their wall clocks remain in sync with the
rotation of the planet on which they presently reside.  Due to this
need in some cases for the phase of differing sources of periodicity
to be adjusted to agree between those who track atomic periodicity and
those who track astronomical periodicity a system such as leap seconds
may need be maintained in certain timescales to occasionally
synchronize the phase of these varying oscillatory sources.

I'd be interested in the groups comments.

Eric

On Wed, Jan 15, 2014 at 9:08 PM, Tom Van Baak <tvb at leapsecond.com> wrote:

> Rob,

> Glad you got a chance to read that volume. I thought Steve and I were the only ones who spent time reading the history of atomic timescales over the last century. It's really quite fascinating, if you have the time.

>

>> “Dr. STOYKO commented that even though the atomic standard is not a clock,

>> it can still be used as a time-keeper through the intermediary of quartz clocks.”  […]

>

> Please understand Stoyko's comment. In the early days the cesium beam apparatus was a frequency discriminator and could not, or was not, run continuously. The local time-scale was instead maintained by laboratory-grade quartz oscillator(s) and these drove the clock.

>

> Once in a while, they fired up the cesium beam as a way to measure the absolute frequency offset and drift rate of the quartz oscillator. In the mid to late 1950's this necessity of re-calibration promoted the (valid) argument that atomic frequency standards were not "clocks".

>

> However, by the end of the 50's, and certainly by the early 60's, commercial cesium clocks were available in quantity and were run continuously. You will see references to National Company's Atomichron as early as 1956,57,58.

>

> At this point not only did atomic frequency standards get promoted to rightfully being called clocks, but multiple atomic clocks ran in multiple locations with time transfer by radio. With a critical mass of coordinated atomic clocks in place, there was no turning back. Atomic clocks were so much better than the planet, pendulum, and quartz clocks they replaced.

>

>> The “atomic standard is not a clock”.

>

> This was true for molecular and atomic frequency standards before about 1958.

>

> Steve,

> Thanks for the ADS links.

>

> /tvb

>

> _______________________________________________

> LEAPSECS mailing list

> LEAPSECS at leapsecond.com

> http://six.pairlist.net/mailman/listinfo/leapsecs