IPv6 multihoming question

[Nick Hilliard]

> I can see a *lot* persons in the IPv4 world that are really unhappy
> with "how IPv4 works".

My comment about ipv4 not being broke was more to do with end-user
experience rather than the opinions of network engineers having to deal
with the sort of crap that they have to deal with on a daily basis.  We
know that there are serious problems which would disappear in an ipv6
world (or at least be replaced by a different set of perhaps more
tractable ones).

The truth is that the internet is made sufficiently simple these days so
that people can just plug in whatever trashy equipment they own, and it
will probably work reasonably well.  They don't give a rat's ass about
ipv4 or ipv6, just whether they can do google searches and read their
email.  And they are certainly not going to implement a v6 capable CPE
access device which costs €5 more than the competitor's v4-only system.

Corporate deployment policy mirrors this.  If end-users do not deploy
ipv6, why should they spend time and effort fixing a problem which
doen't exist?  If it ain't broke... 

I look forward to the day that Microsoft enables ipv6 by default on
NT-du-jour.  This is one of the few things which may actually make a
serious difference to v6 deployment, because only will there be a large
v6 install base.  Other than that, I can't really see any further
serious drivers for v6 deployment other than imminent v4 address space
starvation, government mandates notwithstanding.  It's a bit pathetic,
but there you go.

Nick