[time-nuts] The forbidden question

[Kevin Birth]

When a clock is represented as only losing a second in billions of years
that is a statement packaged in a rhetorical fashion to impress readers.
Another way to think about the statement of such long-term accuracy is
that it is a improvement in reducing uncertainty about accuracy over time,
and that includes uncertainty in the short term at high levels of
precision.

Consider that the SI second is 9,192,631,770 transitions of cesium, but
that when the number of cesium transitions per second was originally
measured, there was a plus or minus of 20 transitions.  That means in
first generation cesium clocks there was a lurking uncertainty in accuracy
of at least plus or minus 20 not taking into account all the other factors
that can influence a clock¹s performance.  Now if a clock¹s time was
uncertain by just a few transitions, then that could produce a 1 second
loss of accuracy over billions of years.

So one way to look at the claim that a clock is accurate for billions of
years (although NIST-F2 is claimed for hundreds of millions of years) is
that it is a clock that has reduced the uncertainty in the short run,
making it not only precise, but accurate at high levels of precision over
short periods of time.

Now that we live in a world where big data analysis (including data feeds
on Wall St) have ns levels of precision, reducing uncertainty in accuracy
in primary standards is highly valued.

Or, at least that is how I understand things.

Best,

Kevin  


-- 
Kevin K. Birth, Professor
Department of Anthropology
Queens College, City University of New York
65-30 Kissena Boulevard
Flushing, NY 11367
telephone: 718/997-5518

"Tempus est mundi instabilis motus, rerumque labentium cursus." --Hrabanus
Maurus

"We may live longer but we may be subject to peculiar contagion and
spiritual torpor or illiteracies of the imagination" --Wilson Harris




On 6/4/19, 12:43 PM, "time-nuts on behalf of William H. Fite"
<time-nuts-bounces at lists.febo.com on behalf of omniryx at gmail.com> wrote:

>EXTERNAL EMAIL: please report suspicious content to the ITS Help Desk.
>
>
>Warning: Potentially heretical material below
>
>Let me begin by saying I am neither an engineer nor a time expert. My PhD
>is in statistics and my spouse's PhD is in theoretical computer science,
>working on quantum computer algorithms. Neither of us claims any special
>expertise when it comes to time and frequency measurement. I am a radio
>amateur and I came to this group following a recommendation from John
>Ackermann, who very kindly answered some questions for me regarding the
>amateur radio frequency measurement test. I thoroughly enjoy the dialogue
>here and I think that I have learned a bit about the subject though, by
>any
>standard of this group, I am the rankest newbie.
>
>My question is a serious one. I am not trolling, nor am I trying to begin
>an argument, nor am I implying criticism of anyone or any endeavor, here
>or
>elsewhere.
>
>What useful purpose, if any, is served by the continuing evolution of
>clocks like NIST-F2 that now achieve accuracy along the lines of one
>second
>per many billions of years? Are there tangible benefits to be had? I
>consulted an astronomer friend who advised that the current generation of
>clocks would allow a suitable space vehicle to plant a probe squarely in
>the middle of Alpha Centauri, if rocket technology existed to do so. We
>have many friends in the academic computer science community who say that
>neither conventional nor quantum computers that exist at present or in the
>projectable future require anything like this kind of accuracy.
>
>By no means am I questioning the value of new knowledge qua knowledge. For
>theoreticians like the one to whom I am wedded, no justification is needed
>beyond the words of mountaineer George Mallory: "Because it's there." I'm
>sure that engineers and scientists in the field of time and frequency
>measurement feel the same. From that perspective, there need be no
>rationalization beyond the desire to do it just a little better than it
>has
>been done.
>
>Please don't lecture me about the value of science for its own sake. My
>career has largely been built on that principle. I'd like to be informed
>as
>to present or anticipated applications that require such accuracy. Are we
>developing these incredible devices just to push boundaries? Or do they
>have some practical purpose?
>
>I'll appreciate thoughtful answers. Dismissive and/or snarky replies will
>be deleted unread.
>
>Thanks for your help.
>
>
>--
>Homo sum humani a me nihil alienum puto.
>_______________________________________________
>time-nuts mailing list -- time-nuts at lists.febo.com
>To unsubscribe, go to
>http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>and follow the instructions there.