On 5/21/22 11:52, John R Levine wrote:
My contention is that we (the ietf) should have done exactly what p=reject
said. They don't want their mail forwarded, we shouldn't forward it.
Which is probably true if IETF were a mail forwarding service, like iki.fi.
But it would be such fun to reprise our success at getting the world not to use NAT, and at the same time make it impossible to get work done.
The sad truth is, IETF didn't really even try to get the world to not use NAT, and the minimal efforts that were made in this direction were far too late.
Beyond that, we already have the revsrsible address rewriting hack I invented for the mailing lists which turns steve@xxxxxxx into steve=40aol.com@xxxxxxxxxxxxxx. It's ugly but it works and it is simple enough that you can undo it in procmail as you deliver your own mail.
Every deployed hack (including NAT) is ugly but "works" in isolation, provided you only consider the use cases you care about. It's when multiple hacks (each with limited applicability) are layered that the problems crop up. And yet, quite often the proposed solutions are to add more ugly hacks that are themselves of limited applicability.
Keith
