pgsql(at)mohawksoft(dot)com writes:
> On a very basic level, why bother sampling the whole table at all? Why not
> check one block and infer all information from that? Because we know that
> isn't enough data. In a table of 4.6 million rows, can you say with any
> mathmatical certainty that a sample of 100 points can be, in any way,
> representative?
This is a statistical argument, not a rhetorical one, and I'm not going
to bother answering handwaving. Show me some mathematical arguments for
a specific sampling rule and I'll listen.
regards, tom lane