[arin-ppml] support for 2014-1 (out of region use)

[David Farmer]

On 2/10/14, 12:16 , Andrew Sullivan wrote:
> On Mon, Feb 10, 2014 at 05:55:18PM +0000, Milton L Mueller wrote:
>>
>> Let me repeat Scott's question in a different way: David Farmer says "we need to clarify that our of region use was always intended to be allowed by policy" and you agreed very strongly with him, and 2014-1 does that. Are you still maintaining that you are opposed to 2014-1?
>>
>
> I should have trimmed that last sentence, actually.  (I apologise for
> being so careless.  I'm unusually obtuse today, which is saying
> something).  The main thing that I wanted to agree with was that the
> historic policies were ambiguous and that anyone who thinks they're
> going to fix that with a lot of tight controls is mistaken, so we have
> to live with the consequences.
>
> To me, this new proposal is worse than doing nothing.  I prefer it
> over previously-floated alternatives (which seemed to me to be
> attempting to create a new policy that was never there).  So, if a
> policy is going to be adopted, I can stand this, but I don't think any
> policy is actually needed here.

Andrew,

If we were only considering IPv4, I would very much agree with you, 
"move along, there's no problem worth solving here".

However, we also have IPv6 to consider as well.  I feel there actually 
is a problem worth solving here for IPv6.  Hopefully, everyone is busy, 
or soon will be, deploying their IPv6 networks, some of those networks 
will have a global reach.  Many people are under, what I think is, a 
misconception that you MUST get IPv6 resources from all five RIRs to 
deploy a global IPv6 network.  While I don't want to tell anyone they 
CAN NOT get resources from all five RIRs, if they wish.  I very much 
don't want them to do that just because of a misconception that they CAN 
NOT get there global resources from one Primary RIR, if they wish.

For an example of what I'm talking about, see the following, jump to 
time stamp 1:34:20 for the specific conversation;
http://new.livestream.com/internetsociety/INETDenver2013

So, I believe this is a problem we can and should solve now, before the 
vast majority of enterprises deploy their IPv6 networks, especially 
large international enterprises.  Preventing unnecessary bloat of the 
IPv6 routing table, solely because of a policy misconception, seems 
worthy of action in my opinion.  This is something we can fix now and if 
we wait until later it may not be, like it mostly is mostly too late for 
IPv4 now.

My personal motivations for working on this issue has been IPv6.  The 
problem is as soon as you touch this issue everybody assumes its all 
about IPv4.  However, for me this is much more about the future of IPv6, 
which is something I hope everyone agrees is worth doing something 
about.  I believe if we really think out of region use is permitted we 
need to clear up this ambiguity and clearly permit out region use at 
least for IPv6.  Otherwise, we need to be clear it is not permitted at 
least for IPv6.  Leaving this issue ambiguous for another generation of 
technology, is just asking for similar problem with IPv6 in a decade or two.

Now, if you think a simpler out of region policy would be better, I'd be 
fine with that.  Much of the complexity comes from the assumption this 
is about IPv4.  Also, if you think it would be better to deal with this 
only for IPv6, I'd be fine with that too.  However, if we fix the issue 
for IPv6 only, I'm concerned in contrast it only exacerbates the issues 
for IPv4.

Recommendations anyone?

Thanks.

-- 
================================================
David Farmer               Email: farmer at umn.edu
Office of Information Technology
University of Minnesota
2218 University Ave SE     Phone: 1-612-626-0815
Minneapolis, MN 55414-3029  Cell: 1-612-812-9952
================================================