Why use zone names at all?

Markus Kuhn mskuhn at cip.informatik.uni-erlangen.de
Tue Jun 18 09:25:46 UTC 1996

> Actually, in a worldwide database I would avoid using zone names altogether
> and use numbers only; whatever MET or CET is right, they're both GMT+1.

What we refer to as "timezones" are often strictly speaking two different

  (A) A fixed offset relative to UTC (e.g., in Germany MEZ = +0100
      and MESZ = +0200)

  (B) A rule that determines the offset to UTC (e.g., in Germany
      MEZ is +0100 in winter and +0200 between the last Sunday of
      March and October, with a switch at 02:00 or 03:00).

For (A), I agree that a numeric designation a la ISO 8601 is
clearly the best solution. However, you will still need names in
order to identify (B).

What is the exact terminology with regard to (A) and (B). So far,
I have heard the word "timezone" being used inconsistently for both.


Markus Kuhn, Computer Science student -- University of Erlangen,
Internet Mail: <mskuhn at cip.informatik.uni-erlangen.de> - Germany
WWW Home: <http://wwwcip.informatik.uni-erlangen.de/user/mskuhn>

More information about the tz mailing list