[arin-ppml] Discussion Petition of ARIN-prop-125 Efficient Utilization of IPv4 Requires Dual-Stack

[Owen DeLong]

On Dec 28, 2010, at 11:10 AM, William Herrin wrote:

> On Mon, Dec 27, 2010 at 2:46 PM, Owen DeLong <owen at delong.com> wrote:
>> On Dec 27, 2010, at 8:35 AM, William Herrin wrote:
>>> 1. The organization has previously received an IPv6 allocation or
>>> assignment from ARIN or another RIR, or
>> 
>> Delete paragraph 1. What is the point of allowing one to bypass the
>> requirement by merely acquiring a block of IPv6 addresses?
> 
> Hi Owen,
> 
> IIRC, ARIN staff presented some numbers in Atlanta to the effect that
> only about a quarter of the folks using IPv4 addresses under an ARIN
> RSA had actually requested IPv6 addresses. I can think of a number of
> possible explanations but they largely imply the same thing: most of
> the key orgs on the Internet haven't taken the first step in their
> respective IPv6 deployments.
> 
> The point of paragraph 1 is to get folks started.  If folks get
> started and then choose to stop, that's another problem. But my
> thought is: let's try a solution to the problem we know for sure we
> have before we worry about the downstream problems we might have.
> 
I don't believe merely requiring them to request IPv6 addresses will
in any way motivate them to actually use them. I don't like these kinds
of indirect attempts in law and I certainly don't want to start down that
road in ARIN policy.

If we want to require them to use IPv6, then, let's require them to use
IPv6.

> 
>> However, good leadership requires sifting through and moving beyond
>> those factors to what is actually good policy. Limiting the last bread
>> crumbs of IPv4 to those who have implemented (for some arbitrary
>> definition of implemented) IPv6 is the policy equivalent of rearranging
>> the deck chairs. It doesn't really do anything to help the community
>> but it might keep some people occupied for a while so that they don't
>> notice the water temperature until it reaches their necks.
> 
> That's really a fundamental question here, isn't it? Do we want ARIN
> to lead or follow?
> 
I think ARIN has done a pretty good job of leading so far. However,
perhaps it is important to consider the difference between leadership
and management. ARIN is leading. ARIN should not be attempting
to manage. Leadership is bringing the horse to water, which ARIN
has most certainly done. Management is forcing him to drink, which
generally tends not to end well. Proposal 125 sought to manage.

> If we want ARIN to -lead- towards the IPv6 solution that we, as
> engineers, consider technically superior to carrier NAT then it's past
> time to start gently nudging folks in that direction with our policy
> choices. It's no different than when the folks of their day said hey,
> CIDR's the new ball game, get used to it.
> 
I believe ARIN has been and continues to lead towards IPv6.

> If we want ARIN to -follow- then we should, as Robert says, let "the
> economics of the transfer market vs. the equipment market make the
> proper course of action crystal clear."
> 
First, I don't buy into your concept that "lead or follow" is somehow
a binary choice.

Good leadership operates at the consent of the people lead. I see
nothing to indicate consent or consensus for ARIN to start trying
to strong-arm IPv6. I believe external forces will drive that soon
enough. People that have been paying attention are already
moving on IPv6. Those that are not will not deploy it just because
we hold the last few IPv4 addresses hostage for them obtaining
an IPv6 allocation/assignment. That's like those "did you
really mean to..." dialogues everyone just dismisses without
reading on their computers.

Owen