I have a question about schema usage in postgres. Not how to use
schemas- I've read the friendly manual (and in this case it is most
definately friendly manual- helpfull, well written, lots of examples,
truely a friendly manual, kudos to the authors) so I know how to use
them. My question is over when and why to use them. For example, is it
better to have a small number of schemata with lots of tables in each,
or a larger number of schemata with fewer tables in each? This is
probably a coke vr.s pepsi (or emacs vr.s vi) sort of debate, but I'd
like to know what the arguments on both sides are.
Is there a performance difference between the two, vr.s having
everything in public (and just real long table names)?
What are the rules of thumbs/patterns people use for what belongs in a
given schemata, vr.s some other schemata, vr.s it's own schemata? Or do
you just throw stuff where ever and refactor the database design when it
becomes unmanageable?
Thanks.
Brian