[ppml] IPv6 flawed?

[Paul Vixie]

> Yep, that's right. I really don't do enough meaningful networking to  
> speak up here. I should have kept my mouth shut.

i disagree on both counts.  your example was good, since it showed what
can be done if the original network was provisioned "lightly".  and as
for keeping your mouth shut, i'm glad you didn't, since it means folks
got to hear somebody say "ipv6 enables different provisioning philosophies
than ipv4, and renumbering can be made less painful in v6 than in v4."

was anybody else here alive at the time microsoft decided to get into IP?
i myself was a beame and whiteside fan, though i had friends at ftp
software and other ip-on-windows companies.  the common belief was that
with microsoft coming 20 years late to the party, they would be an also-ran.
(i think similar thoughts were spoken aloud when internet explored came
out in a mostly-mosaic web community.)  hopefully a lesson was learned?

that lesson is, the installed base is meaningless, and how we did it before
is meaningless, all that matters is getting growth right.  with all due
respect to owen, the number of other people's firewalls that had our old
ip addresses in it is not going to matter in the long run, and the long run
is what we've got to plan for and worry about.

that said, ipv6 doesn't have the same inevitability than, for example,
internet explorer had in a mostly-mosaic web, since it's a new technology
that's not interoperable with anything that existed before.  so, i'm in
total sympathy for those who say ipv6 might die before deployment, or that
we might all die of old age before deployment, and so on.  i am particularly
upset that ipv6 failed to address the failed ipv4 routing paradigm, which
was (and which remains) the biggest reason ipv4 had to be replaced.

but it's not about the number of vpn's landing on current networks.