Re: Precision of data types and functions

On Fri, 2006-09-01 at 13:33, Brandon Aiken wrote:
> Of course the year exists. The date itself is nonsensical, however.
>
> 'January 3, 648' does reference a valid day, but the date itself has no
> meaning at that time in the world, so there is very little meaning in
> using Gregorian dates except to give us a relativistic idea of when it
> occurred. Nevertheless, you can never say with specificity anything
> that occurred on any given date prior to the inception of the Gregorian
> calendar without doing conversions to a calendar no longer in use while
> taking into account the numerous error corrections that have been made
> to various calendars. Knowing the year and season something happened is
> about the best that can be expected.

Actually, it's far worse than that. I was reading up on the adoption
rate of the gregorian calendar, and because of delays in its adoption by
various folks over the centuries since it was invented (and the fact
that some religions STILL don't recognize it) there are many dates that
are recorded differently throughout history since it was created.

I read up a bit on it at http://en.wikipedia.org/wiki/Gregorian_calendar
and it's enough to make your head spin... The trivia section near the
bottom is particularly interesting.