[ppml] PIv6 for legacy holders (/w RSA + efficient use)

[William Herrin]

On 7/30/07, Paul Vixie <paul at vix.com> wrote:
> i'd like to know what makes this alternative the most likely.  if someone
> who holds the "markets are efficient and inevitable so just relax" view
> can show how that view is falsifiable, i'm listening.

Paul,

If I understand David's argument (and hopefully he'll correct me if I
got it wrong) its that IPv4 depletion is more likely to drive IPv4
address space conservation (and hence an explosion of the DFZ table)
than it is to drive IPv6 adoption. Business will tend towards this
path because the incremental cost of conservation is small and the
benefits are immediate while the cost of IPv6 deployment is large and
the benefits are remote.

The argument would be false if IPv6 could be used as a less-desirable
drop-in replacement for IPv4. For example, if IPv4 was Cola and IPv6
was juice, a restaurant could reasonably serve juice after the cola
ran out. That doesn't appear to be the case. Until IPv6 reaches some
critical mass where the remaining IPv4-only servers can be ignored,
the two protocols will lack sufficient equivalence. It would be more
like serving animal crackers when the steak ran out.

The argument would be false if there was a reason to deploy IPv6
independent of IPv4 depletion. For example, if there was some killer
app for which use of IPv6 was a prerequisite. No such app has emerged
and I can't imagine what requirement would make such an app unable to
use IPv4 instead. Even the peer to peer guys have managed to make
their software work through the NAT firewalls.

The argument would be false if a sufficiently large subset of the IPv4
community could be enticed to begin paying money for IPv6 regardless
of whether they could yet use it. The fact that they're paying for it
anyway would motivate them to deploy IPv6, and put pressure on their
service and content providers to deploy it as well, independent of
IPv4 depletion. For example, if the legacy end-user registrants had a
short, well-publicized opportunity to get an IPv6 PI assignment before
the window permanently closed, for which they had to pay $500 and then
continue paying $100/year...

Regards,
Bill Herrin


-- 
William D. Herrin                  herrin at dirtside.com  bill at herrin.us
3005 Crane Dr.                        Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004