On Mon, 2012-06-04 at 11:56 -0400, Bruce Momjian wrote:
> I get your point about COUNT(*) really counting rows, not values, but
> why doesn't GROUP BY then skip nulls?
A while ago, I came to the conclusion that applying logic to extrapolate
the behavior of NULL is a bad idea:
http://thoughts.davisjeff.com/2009/08/02/what-is-the-deal-with-nulls/
Jose was not wrong about the inconsistency between UNIQUE and GROUP BY.
But the answer is that "we do it that way because the standard says so".
And that's a good reason.
Regards,
Jeff Davis