This bunkum has been debunked extensively by many smarter people than me.
[everything else snipped]
Dave, all this ex cathedra stuff on your part has got very wearing, I don't
see any point in continuing.
it isn't clear to me that the language above is particularly helpful.
i do not single out this note as being unhelpful, as their have been
others in a similar vein; accordingly, i ask that participants kindly
re-read:
http://www.imc.org/ietf-mxcomp/mail-archive/msg00358.html
regardless, certainly the current smtp architecture is based on a
store-and-forward model, and while many sites are configured such that
domains are reachable in one hop, other sites are legitimately
configured otherwise.
setting aside the architectural issues, it is clearly worthwhile to
discuss whether disenfranchising these sites is a possibility, and, if
so, whether it is worthwhile. (thoughtful readers will note that that
the preceeding sentence doesn't preclude making this change, but rather
advocates a careful discussion of the benefits and costs.)
/mtr