[arin-ppml] Encouraging IPv6 Transition (was: Clarify /29 assignment identification requirement)

[William Herrin]

On 5/14/12, Drake Pallister <drake.pallister at duraserver.com> wrote:
> I don't have the answer, just more questions and some thoughts to ponder.
> How related do you think the widespread dragging of the feet about IPv6 is
> to the current economy?

Hi Drake,

I think the major reasons for slow IPv6 deployment have nothing to do
with the RIRs whatsoever and rest in areas over which the RIRs can
exert no control.

The economy is one drag for sure. Tight budgets mean no money for
anything that isn't needed last week.

Another major drag, is that the programmer's sockets API provides no
clean, simple way to "connect to the first connectable of this list of
IP addresses." As a result, only those applications whose programmers
have gone to extraordinary measures will promptly fail over to the
second IP address when the first is inaccessible. This isn't a problem
in the normal state of IPv4 operations: most names map to only one
address. Add IPv6 and that's no longer true: most names map to at
least two addresses, one IPv6 and one IPv4. So, deploy IPv6 and
suddenly your customers report problems reaching your server even
though the IPv4 path is unchanged.

These are not things ARIN can affect. What can ARIN do to encourage
IPv6 deployment? Make the process of getting addresses cost and hassle
free to the absolute maximum extent practical.

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