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

[Owen DeLong]

>>> ARIN policy on IPv6 assignment presents far fewer barriers than it did
>>> four years ago, but reasonable analysis grades any resource request
>>> from ARIN as "difficult."
>> 
>> I disagree. It took  me less than an hour to prepare HE's subsequent
>> allocation request for ARIN and submit it. It took less than 48 hours
>> for ARIN to issue the requested /24.
> 
> You're a subject matter expert, as am I. Try it as someone who is
> still learning IPv6 and has no knowledge of ARIN's deep mysteries.
> 

Sorry, I'm not willing to buy this argument when discussing an organization
that already has an IPv4 assignment from ARIN. Somewhere in the
organization's past, someone already went through the IPv4 process.
If they're going back for an additional assignment (as was described
in the proposal), then, someone is going to have to learn all of that to
navigate the IPv4 process anyway.

IPv6 is the exact same process except that the template is much simpler
and your justification section can be as simple as:

	I have N buildings on O campuses. The largest campus has Q
	buildings. Please give me R /48s. (it's not particularly complicated
	to compute the values of N, O, Q, and R for any organization).

> Folks who don't have an IP address management specialist on staff
> often contract someone each time they interact with ARIN or get one of
> their ISP's specialists to help. Like hiring a lawyer to represent you
> in court. You don't do that until your goal becomes a priority.

When any of my customers engage me to help them get an IPv4 assignment,
I urge them to consider also doing an IPv6 assignment at the same time. I am
often able to do this without adding hours to the contract as the data required
is already in the IPv4 process in most cases. This often leads to additional
hours of IPv6 consulting for said customer, so, any consultant that is doing
IPv4 applications for end-users and not selling IPv6 as an add-on is
missing business opportunities.

> For the vast majority of organizations, IPv6 deployment is not a high
> priority and it's not likely to become a high priority any time soon.
> You mean if I and everybody else rush to deploy IPv6 then I might not
> have to worry about IP addresses ten years from now? Hrm.

Sure, but, those organizations that might benefit from this proposed
auto-assignment of IPv6 are coming back for additional IPv4 anyway,
so, any rational consultant would be looking to add IPv6 as an
inexpensive upsell that may leverage additional consulting business
later.

> Folks will fiddle with IPv6 out of curiosity and as part of the
> routine effort to keep one's skills current. But such idle priority
> tasks don't get high priority resources. Which means no consultants
> and no major fees. Which means no address request to ARIN. Which means
> zero idle-priority deployment of IPv6 for multihomed orgs. Which in
> most cases is the only form of IPv6 deployment that can reasonably be
> expected at this stage.

Sure, but, if they're doing IPv4 anyway, they can get an IPv6 added without
paying much more, if any, to the consultant.

Owen