[arin-ppml] ARIN-prop-133: No Volunteer Services on Behalf of Unaffiliated Address Blocks

[Benson Schliesser]

On Feb 14, 2011, at 9:41 PM, Gary Buhrmaster wrote:

> On Mon, Feb 14, 2011 at 13:48, Benson Schliesser <bensons at queuefull.net> wrote:
>> 
>> On Feb 14, 2011, at 3:41 PM, Keith W. Hare wrote:
>> 
>>> My interest in maintaining the status quo for legacy resources is that the system currently works. Unless a change is going to be a large benefit to the internet community, what is the point of changing?
>> 
>> IPv4 exhaustion is a change that won't benefit the Internet community*, but it's happening.  We really can't expect the status quo to be maintained forever.
>> 
>> Cheers,
>> -Benson
>> 
>> * - Well, not in the short-term at least.  Once we have critical mass of IPv6, I think it will benefit us all... But that's a different discussion.
> 
> While I am not as optimistic about the dates of critical mass
> of IPv6 adoption as some in the community, I see little reason
> to invest efforts in ARIN IPv4 policy at this point unless there
> is a clear benefit to the community in the migration to IPv6.
> This policy proposal would seem to result in a distraction for
> the community and ARIN in dealing with legacy holders of
> IPv4 space, would seem to have operationally negative
> consequences, and does nothing to move forward with IPv6.
> Opposed.

Thanks for this feedback.  It is a sentiment I've heard from others, over the past year or two.  And it deserves some credit for emphasizing and promoting IPv6.

However, I disagree; I think such a view ignores the present reality, in which IPv4 is the dominant network protocol.  I would be shocked to learn of anybody, who is responsible for a production Internet-connected network, that is willing to return all of their IPv4 addresses and rely completely on IPv6 at this time.  I hope we get to that point in the near future.  But between now and then, we have a need for both protocols.

Given that the need currently exists, and given that the RIR supply is almost exhausted, it is my opinion that we should implement policies to more effectively manage IPv4 numbers during the transition period.  If we don't, then new networks may find it impossible to acquire needed IPv4 resources, while existing networks hold idle resources with no incentive to share them.  Further, the gap between those with excess supply and those with unfulfilled need is ripe for exploitation, and change is happening whether we prepare for it or not.

In other words, IPv6 transition will happen for its own reasons; we don't need to further break IPv4 in order to encourage it.

Cheers,
-Benson