Brian: (1) I can't resist invoking the e2e argument and say you shouldn't depend on intermediaries for reliability. (2) It's a good thing that this list is really about NAT66, not those stateful edge NATs.
Brian E Carpenter allegedly wrote on 10/28/2010 17:37 EDT: > Roger, > > On 2010-10-28 17:29, Roger Marquis wrote: > ... >> Home and business users require NAT for 1) layer abstraction (no >> different, fundamentally, from the abstraction of Ethernet's MAC layer), >> 2) flexibility (unlinking internal from external and multi-homed >> topologies), and 3) security. > > For home users and the vast majority of business users, these three > arguments are well known to be spurious. I won't waste bits on that. > > On the other hand, who's willing to refund me the $500NZ that I > unintentionally donated to a charity earlier this year, because > a NAT session timeout aborted a "secure" connection before the > ack for the credit card transaction came back over HTTPS, allegedly > a protocol that NATs handle impeccably? > > OK, it was my own stupid fault for repeating the transaction before > checking my credit card account, and it was for a good cause, > but this is the basic reason why NATs are a bad thing for homes > and businesses. They break stuff. > > Brian > _______________________________________________ > nat66 mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/nat66 _______________________________________________ nat66 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nat66
