[LEAPSECS] The relation between Easter and leap econds.

[Peter Bunclark]

I deliberately chose words that didn't refer to being right here in the
early 21st century;

On Mon, 10 Nov 2008, Tony Finch wrote:


> On Sun, 9 Nov 2008, Peter Bunclark wrote:

> >

> > So a User requirement might be:

> > The rythms of life, including the orbits of the earth and the moon,

> > the rotation of the earth, and convenient sub-divisions of the rotation

> > down to nearly the limit of human perception, shall be expressed in

> > a single monotonic calendar.

>

> I guess you mean the limits of human perception as assisted by suitable

> tools...


No. Tools evolve a lot more quickly than we do.  I had around 1s in mind.
My radio-synchronised wristwatch seems to me to be perfectly accurate; but
really only has to achieve sub-second for that. TV frame rate is beyond my
perception, by design (just about).


>

> Most common calendars have given up trying to track both the moon and

> the sun. Most of them prefer to track the sun; the Muslims track the moon;

> and the Jews track both and the result is impressively complex.


We have months, which are lunar-ish.


>

> Your specification doesn't capture how to measure the rotation of the

> Earth. We used to use apparent solar time, then we switched to local

> mean solar time, then we switched to standard time, then we added

> arbitrary unpredictable seasonal offsets.


That's because it's not a specification, it's a requirement. It should not
pre-empt a solution.  What you might be on to, is that "earth rotation"
should actually be "daily solar cycle". See how hard it is; even in a
quick attempt I let my astronomical knowledge get in the way. The Sun
comes up in the morning, doesn't it, the world doesn't sink; duh!


>

> My point is that user requirements change. The relation between natural

> phenomena and our consensus idea of time depends on how accurately and

> quickly we can measure those phenomena, and which phenomena we care about

> tracking.

>

Again, I tried to be as timeless as possible.

Note, this is "*A* user requirement" not "the ..."; another might be:

2. For the purposes of temporal metrology, clocks must be built to the
best accuracy of the technology of the period.  Clock keepers must combine
output to ensure a single consensus of current synchronous time.
This coordinated time must be available as widely, efficiently and freely
as possible to all users.

well I'm on a roll now:

3. The relationship between the Calendar(1) and Synchronous Time(2) must
be deterministic, such that an instant represented in one system may be
uniquely identified in the other.

and now to the rub:

4. The fundamental unit of both systems shall be the SI second.

Hm we need a snappy title...
"Requirements for improving the accuracy of Civil Time while
 reconciling the Calendar with Precision Timekeeping". Ug...

When the requirements are agreed, tenders are in and you've won the
contract to produce a preliminary design, section headings in the
Preliminary Design Document might include

Measurement of the rotation of the Earth

Propogation of Atomic Time across the Internet

Algorithm for conversion of Atomic Time to Calendar Time.

Pete.