Am Dienstag, 9. Oktober 2007 schrieb Trevor Talbot:
> I don't think it's wrong, just a particular choice. As an example,
> consider an interval scheduling system that handles everything in
> absolute time (UTC), but uses local time as a convenience.
We are not considering an interval scheduling system, we are considering a
database system. Such a system should have the basic property that if you
store A, it will read out as A. The money type is similarly buggy: if you
change the locale, the semantic value of the data changes. With money type,
the problem is obvious and easy to recognize. With the timestamp with time
zone type, however, the problem is much more subtle and will likely go
unnoticed by many who will then be unpunctual for their appointments.
--
Peter Eisentraut
http://developer.postgresql.org/~petere/
