So far we've heard from exactly one source that claims this is a real pain
point. Actually, that wasn’t' even the claim; it was merely an observed
possible optimization. That's not enough in my book to support any kind of
formal standardization effort. It might sound like a nifty idea and there
might be myriad implementations around that do stuff like this, but that
doesn't mean there's a large-scale problem with the practice that needs to be
solved in the form of protocol changes.
In light of this thread, I polled several large mailbox providers. None of
them say they do greylisting. One of them (Hotmail) replied that they don't do
greylisting because they just don't see enough benefit from it. None of our
large customers do it either.
To quote Dave from another posting just now: Standards are extremely expensive
to create and deploy. So far I haven't seen anything supporting the idea that
this is worth the expense. Maybe what has to happen is that some small team
needs to go off and try it, and report back results.
Absent that, I suggest this would at best deserve "Experimental" status. I'd
support that sort of thing, though even those aren't cheap in IETF terms.