> It might be more useful after we get the infrastructure that Kevin's been
> working on to allow collecting all the updates into a tuplestore that
> could be passed to a statement-level trigger. Right now I tend to agree
> that there's little point.
Maybe, this can be used to re-implement FOREIGN KEYs. Never-ending
bulk DELETEs caused by lack of indexes on foreign key columns are
biting novice users quite often in my experience.
