[LEAPSECS] USWP7A docs for 2013 September meetings

[Clive D.W. Feather]

Stephen Colebourne said:

>>> The easiest way to eliminte them is to remove them. However, I differ

>>> from Poul-Henning however in that I don't find it palatable to ignore

>>> the fact that humans define days by the Sun.



>> The result was that most humans live perfectly comfortable lives

>> with the sun being 1-1.5 hours "off" from solar time, due to timezones

>> and daylight savings times.



> "We all know that time doesn't follow that definition exactly, but

> over the long term, the definition has always stood and has real

> meaning to  non-technical humans."

> 

> ie. the actual offset of an individual to the day, and the variance of

> that is clearly a larger factor in peoples lives than leapsecs. But

> when averaged out over the long term, the day is still defined by the

> Sun.

> 

> Its that underlying fact that cannot just be swept under the carpet.

> Any abolition proposal must be crystal clear on how the increasing

> deviation from the underlying day would be handled.


Using the time-zone proposal.

UTC ceases to contain new leap seconds. This means that it will gradually
drift away from mean solar time at Greenwich. For a country whose local
time ("LST") is currently UTC+N, this means that over the years/centuries
mean noon will gradually get further and further away from 12:00 LST.

In any given country, eventually this discrepency will approach the
"unacceptable" point. When this happens, the relevant authorities will
issue an ukase changing LST from UTC+N to UTC+(N+1) at some moment. This
will work in exactly the same way as the present summer-time changes do
(for example, those that move my LST from UTC+0 to UTC+1 and back [1]).
Indeed, the easiest way to do this is to omit one of these, so that it
goes:

    March      UTC+0  ->  UTC+1
    October    UTC+1  ->  UTC+0
    March      UTC+0  ->  UTC+1
    October    [nothing happens]
    March      UTC+1  ->  UTC+2
    October    UTC+2  ->  UTC+1
    March      UTC+1  ->  UTC+2
    etc.

It wouldn't be suprising to find that the entire EU makes this (non-)change
on the same date.

Result:
- UTC doesn't need leap seconds
- local time remains very roughly aligned with the sun
- local time is atomic based
- only familiar processes are used
All wins.

The only downsides are:
- The UK eventually ceases to be on UTC+0, raising arguments as to what
  "GMT" actually means (but, on the upside, the Daily Mail will be so
  confused it might spontaneously self-destruct).
- We start to see zones like UTC+15, UTC+16 to add to the present UTC+14,
  and *eventually* will see UTC+24, UTC+25. In the same way that we
  presently have two places in the world with the same 24-hour clock time
  but on different dates, we will eventually have the UTC date be different
  to the local date everywhere all the time. But that is so many thousand
  years away that I'll let my descendants - if any - worry about how
  they're 're going to reform the calendar. I suspect by then we'll have
  moved the Earth away from the sun a bit to counter global warming, so it
  won't matter.

[1] I ignore the fact that my local legal time is defined as GMT+0 or
GMT+1, rather than UTC.

-- 
Clive D.W. Feather          | If you lie to the compiler,
Email: clive at davros.org     | it will get its revenge.
Web: http://www.davros.org  |   - Henry Spencer
Mobile: +44 7973 377646