[arin-ppml] Policy Proposal 122: Reserved Pool for Future Policy Development

[Owen DeLong]

On Nov 23, 2010, at 6:29 AM, Leo Bicknell wrote:

> In a message written on Mon, Nov 22, 2010 at 10:36:42PM -0800, Owen DeLong wrote:
>> On Nov 22, 2010, at 8:25 PM, Scott Leibrand wrote:
>>> Q2: Does NRPM 4.10 need to be updated at some point?
>>> Again, I think the answer is yes, and it seems we have quite a bit of
>>> community support for some kind of revisions.
>>> 
>> Yes, but, I think finding consensus on what kind of revisions will be quite
>> hard.
> 
> I don't, I just think we're trying at the wrong time.
> 
> It will only take weeks, or perhaps months past _ARIN_ runout to
> collect a lot of information about what the world really looks like
> on the ground when you can't get IPv4 space.  With that information
> I believe consensus will develop quite quickly around the best thing
> we can do.
> 
I think that is true in any particular subset of the user community. However,
ARIN is quite diverse and I think there will be many factions all vying for
the biggest piece of the remaining pie they can get. In the past, the
community has been particularly bad at swallowing hard and accepting
a policy that makes everyone equally unhappy. What makes you think
this time will be substantially different?

> I realize folks want to have this define in advance, but I just
> don't see how to do that in this case.  If we're going to modify
> 4.10 in any way I think it should be to not give out any space (even
> for "transition technologies").  We need a period of several months
> where folks are forced to do without IPv4 to figure out what really
> is and isn't the most useful way to use the last /10.
> 
I think that's a very poor choice. That basically says that all growth
on the internet during that time is users who have no access to the
remaining IPv4 content and services that have not adapted to IPv6
yet.  This is an untenable situation for providers that are completely
out of IPv4 space and a major competitive advantage for their
competitors that still happen to have space for whatever reason
or are able to obtain it quickly through the transfer market.

As such, I think holding back addresses in this way is not only
monumentally poor stewardship, but, also dramatically disproportionate
in the effect it will have on organizations, providing minimum
disadvantage to the largest consumers of address space.

If you want to conduct an experiment in this manner, I suggest
leaving 4.10 alone and setting aside an additional /10 from the
final /8 to be subject to this future policy effort, possibly encompassing
the 4.10 space as well if there is consensus at that time.

> Part of the reason there are so many views now is that each person
> has a slightly different prediction for what the world looks like
> after run-out.  What you're trying to get consensus on isn't a
> policy, it's on a view of the future.  If we just wait for it to
> happen there will be no more disagreement on what it looks like.
> If we end up with a month or two where this sits unable to be given
> out, oh well.
> 
I think another part of the problem is that people are coming from
different perspectives on how the inability to get new IPv4
addresses for expansion will affect their business and the
criticality of their particular need vs. that of the rest of the
community.

Owen