IPv6

Mailman List Mirror (Read Only)
1

Andy Dills wrote:

Am I the only one that thinks IPv6 is a minimum of ten years out
before you see actual non-geek demand?

It will probably happen before that. The Japanese government
requirement that all businesses be fully ipv6 compliant before 2005 is
certainly going to have a major impact on vendor ipv6 implementations,
from the core to the desktop So while you may not get 30Mpps on your
backbone router, you're probably not going to be stuck with a white
elephant rate-limited to 200Kpps either. One day, you may even be able
to run an ipv6-only desktop from vendor M, who knows?

From one perspective (and not necessarily the best, or even a remotely

accurate one), all it's going to take is for Microsoft and a bunch of
NAS & ADSL equipment vendors to implement stable ipv6 edge connectivity
and to prefer AAAA over A records. Once this happens, there will be
demand from customers by default, and this may create enough of a
business case to justify more infrastructural spending on ipv6. This
would help those providers who have a partial ipv6 deployment in their
core, and may bootstrap the uptake process for those who haven't
bothered looking at it yet.

I'm a bit more optimistic about its take-up these days, mainly because
support from the desktop and the network edge is going to be the main
driving factor, and because this is probably going to become much more
widespread from now on.

Nick

2

Nick et al,

and to prefer AAAA over A records. Once this happens, there will be

Mind that some of the major content so(u)rcerers will have to adopt
their Bind 4.x hacks from the last century to make their DNS respond
to 4A queries instead of just timing out as they do today :wink:

cheers,

Sven

3

Mind that some of the major content so(u)rcerers will have to adopt
their Bind 4.x hacks from the last century to make their DNS respond
to 4A queries instead of just timing out as they do today :wink:

To be fair, this is much less of a problem now than before. Setting a
preference for quad-A is now generally feasible, whereas a year ago, it
involved daily teeth-gnashing and hair removal.

There are a pile of things to fix, including this, and ospfv3 and
core-router packet switching rates and end-user dns requests over ipv6
and stable end-user ipv6 stacks which don't bsod or panic all over the
place, and so on and so forth. This is why the Japanese government is
so important for the uptake of ipv6 globally: it's going to force a
population of 130 million highly-wired people to use ipv6 for everyday
network connectivity, which is going to 1) wring most of these problems
out and 2) cause large vendor software systems to be made ipv6-aware.
These are good things.

Nick