Re: GUC time unit spelling a bit inconsistent

Am Donnerstag, 21. Juni 2007 00:10 schrieb Gregory Stark:
> Afaict nobody has expressed a single downside to accepting other
> abbreviations.

The two downsides I can see are that it would confuse users (even if it
apparently wouldn't confuse *you*), and that there is a chance that the
configuration system would work differently from other PostgreSQL components
or parts. For example modules like earth distance or other astronomy,
physics, or geography modules might all have to create their own sets
of "clearly unambiguous" unit sets for themselves. Few or none of these
types of modules exist yet, of course. I would like to have a units-aware
data type that you can use for storing and computing with measurements, and I
would like to be able to use that same type for dealing with configuration
quantities.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/