It looks like NAT's are a fact of life, and we
just need to figure out how to deal with them.
well, that's the question after all - how best to deal with them?
I claim that NATs are architecturally bankrupt and we should therefore
devote as little energy as possible toward legitimizing NATs and/or
trying to make them any more complex than they already are.
let's not pretend that they don't exist, or that they don't have some
valid uses. but let's not pretend that they are viable in the long
term either.
Keith