On 11 dec 2007, at 21:20, Joel M. Halpern wrote:
I think the answer, to a significant degree, depends upon exactly what
you mean by "IPv4 related work."
For example, the ForCES work may or may not be useful. It certainly
needs to work with and over IPv6. But if we did not make sure it
worked
with and over IPv4, then no one would use it, as a router that doesn't
handle IPv4 is not acceptable.
Right, "first do no harm".
But under a "no new IPv4 work" doctrine, certainly all NAT traversal
work would be out?
Even the RRG work is taking the view that any proposed new
architecture
has to provide value to and connectivity with folks who have IPv4,
if it
is going to be adopted.
"Has"? If we stumble upon the perfect solution that works with IPv6
but not with IPv4 I don't think that solution will be ignored. But on
the other hand, I don't think everyone will call it a day and go home
at that point, either.
Even the plenaries last week noticed that a plan based on co-existence
makes a lot more sense than an assumption of transition. IPv4 is
going
to be used for a LONG time. And it will be the bulk of the usage for
quite a while. Pretending otherwise is just not effective.
If that happens because people can keep running IPv4 without trouble
for years to come, I have no problem with that. But I wouldn't want to
accept this prediction as a design constraint.
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
