> From: "Stephen D. Strowes" <[email protected]>
> a larger layer-3 address space renders NAT redundant
Ah, no. If it were just space, why is there, e.g. NAT66?
For _one_ example of a non-extra-space capability that NAT provides, two
words: provider independence. Others may have different causes to mention...
> Some recent efforts in the IETF ... at least acknowledge NAT is
> probably here to stay
Only insofar as the architecture does not provide people the capabilities
they need (see preceeding comment). Had we an alternative which actually met
peoples' needs, over time it might recede (although no faster than IPv4,
where the need for NAT is manifest).
> In a sense, the "least evil" future of NAT.
To me, that's like saying that burning someone to death is less evil than
hanging them. Well, maybe, but the difference is 'in the noise'...
>From an architectural point of view, NAT is truly ugly, and not just because
it fiddles with the packet contents. There's also the issue that it's putting
state in the network (and 'engineered' NAT might make that slightly less
ugly, for example by allowing replication, etc, so it's not quite so
brittle), etc, etc.
Noel
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
