# Re: Calendars (was: Re: leap second)

```In <m0wfr3z-000CMdC@rulcmc.leidenuniv.nl> joost@rulcmc.leidenuniv.nl (joost witteveen) writes:
>
>                       Now, we know the length of a year/day better, and
> only 1 in for of those turn-of-century years are leap years. Maybe that
> will change again. And about the seconds: we (currently, prossibly always)
> simply cannot calulate the length of a day accurately enough to know
> well in advance when to insert them. But I'd say the two animals are
> at least related, if not mother and daughter.

It is my understanding that:

Leap days are used to keep the calendar in sync with the season.  That
is, you don't want winter to be in August (in the northern
hemisphere.)

Leap seconds are used to keep the time of day in sync with the sun
rise.  That is, you don't want the sun to be rising at midnight (out
side the article circles.)

The length of the year is almost constant over time periods of
thousands of years.  It does vary due to gravitational interactions
with other planets, but it only makes a "significant" difference
(10%?) when dealing with time periods of around 100,000 years.
These gravitational interactions are predictable, so if you really
wanted to, you can calculate the exact length of the year 1million
years ago.  (It is my understanding that there are people who do
this.)

The length of the day is not quite as constant.  It depends on how
quickly the earth rotates which depends on things like how much snow
has fallen on mountain peaks and how much water is in man made
reservoirs.  I kid you not, these things are significant enough to
change things on the order of a second or two per year.  Neither the
weather nor people's water usage/reservoir building is very predictable.
This makes predicting leap seconds futile.

-wayne

--
Wayne Schlitt can not assert the truth of all statements in this
article and still be consistent.

--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org .
Trouble?  e-mail to templin@bucknell.edu .

```