Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> Instead, what I propose (and is not really in the patch), as a
> backpatchable item, is an approach in which the functions to compute
> each rel's Browne strength and sort are hooks. Normal behavior is not
> to sort at all, as currently, and sites that have a problem with the
> current random order can install a custom module that provide hooks to
> change ordering as they see fit. So behavior won't change for people
> who have no problem today.
Meh. I'm not really thrilled with adding hooks (that presumably we'd
have to preserve forever) to solve a short-term problem. Nor does this
sound hugely convenient for users with the problem, anyway. Do we even
know for sure that anyone would create such modules?
I think we should just fix it as best we can in HEAD, and then anyone
who thinks the risk/reward ratio is favorable can back-patch that fix
into a private build.
regards, tom lane