[LEAPSECS] Cheating means more planning, not less

[Tony Finch]

On Sat, 27 Dec 2008, Rob Seaman wrote:

>

> Rather, a clock can be deposited at any meridian on any planet, set to any

> time, running at any rate.  The question is whether a particular choice of

> parameters is useful and sustainable.


Really what it boils down to is a question of how frequently and by how
much we reset our clocks so that civil time (the time used for every-day
purposes) has a useful relation to Earth orientation.


> Additionally if a planet has populations scattered at wide longitudes,

> the more basic requirement is to organize a coherent system to manage

> the whole.


That's just a matter of quantizing timezone offsets with a reasonably
large granularity, as was established in the 19th century.


> > > Only one - standard time based on mean solar time - has ever been shown to

> > > be *practically* workable.

> >

> > Two: standard time plus daylight saving time is the other

>

> DST is a trivial gimmick layered on standard time.  Standard time is a

> global system layered on the mean solar day.


I don't think DST is trivial, nor is it a gimmick.

It's very tempting to dismiss DST as a stupidity, especially if you try to
understand it without regard to the sociology of time. (I used to think so
myself.) There is a very good reason for DST and it is entirely to do with
the way people collectively relate to clocks. The most obvious artefacts
of this relationship are timetables. Ben Franklin's article was satire,
but like all good satire it was pointing out an important failure of civic
life: that we are too strictly governed by clocks, regardless of how well
we use natural daylight.

Back when Franklin was writing, relatively few people lived in cities, and
timetables were rare - even ideas like opening hours or working hours. But
urban living naturally generates timetables: the timing of deliveries of
goods from the counryside determines the opening times of the markets,
which determines shopping hours, etc. There is an inevitable coupling of
the timing of activities between different walks of life. This became
painfully obvious in the USA when (over a period of decades) various
jurisdictions disagreed about when and where DST should be applied. The
worst chaos occurred when physically overlapping jurisdictions used
different rules, so different offices in the same city (e.g. government
vs. private) would be operating on different time. The chaos only subsided
when the US federal government set rules for how DST should be applied,
thereby establishing a consensus.

What DST teaches us is that the the most important property that civil
time must have is consensus over large areas, whether or not this makes
sense w.r.t. natural philosophy. DST is a violation of the concept of
standard time, but it is more useful to society than strict adherence to a
fixed offset from mean solar time. In fact the same lesson is taught by
the establishment of GMT as legal time across Britain. Legal precedent
said GMT is a violation of local time - despite the fact that nobody in
practice behaved as the law said they should, between the establishment of
railway time and the endorsement of GMT as legal time.

So how is this relevant to leap seconds?

Firstly: The history shows that almost any violence can be done to civil
time so long as everyone agrees to it. DST shows that sociology can trump
astronomy. (Standard time shows the same thing, but DST's arbitrariness
makes this fact much more starkly clear.) Broad agreement and consensus is
the foundation of civil time. The way that leap seconds work clearly does
not have enough consensus, in that people still produce software and
standards and specifications that are incompatible with leap seconds. That
fact is probably enough to doom them, just like Britain's local time law
was doomed to lose to GMT in the 19th century.

Secondly: The commonality between DST and UTC is that both of them require
us to reset our clocks occasionally to keep them in a convenient relation
to the position of the Earth. DST's resets exist because the way we
coordinate our activities using timetables (aligned to mean solar time)
doesn't work well with the hours of daylight. UTC's resets exist to keep
our clocks matched to mean solar time.

Can we not accomplish both goals with the same occasional resets, instead
of having two independent reset schedules? Would this also solve the
problem of leap seconds disagreeing with the practical consensus model of
time?

Is there any other way of cleaning up this mess?

My "sunrise time" suggestion is not entirely a joke. Originally its
serious purpose was to point out that there is a rational basis for DST:
our timetables work better if they are anchored to sunrise than to midday.
The point of my recent post is that if you are resetting clocks frequently
to make gross adjustments, then you do not also need a second reset
schedule to make fine adjustments.

Sunrise time has layers of satire and absurdity, as well as of astronomy
and sociology, but to explain them all would spoil the fun and keep me up
too late.

Tony.
-- 
f.anthony.n.finch  <dot at dotat.at>  http://dotat.at/
IRISH SEA: EAST 4 OR 5, INCREASING 6 AT TIMES. SLIGHT, OCCASIONALLY MODERATE.
FAIR. GOOD.