[ppml] Policy Proposal 2005-1: Provider-independent IPv6 Assignments for End Sites - Last Call

[F. David Sinn]

On Apr 17, 2006, at 2:17 PM, Iljitsch van Beijnum wrote:

> On 17-apr-2006, at 22:41, David Williamson wrote:
>
>> Oh, and in case it isn't obvious, I'm very much in favor of  
>> 2005-1.  I
>> think PI space until such time as it isn't needed (if ever...) is far
>> preferable to nothing deployed and running out of v4 space.  Yes,
>> there's a ways to go before that event, but it's coming soon  
>> enough to
>> already look to me like a train wreck.
>
> Question for those of you in favor of the policy proposal:

So is this to mean that anyone that reply's is implicitly in favor of  
2005-1?  :)

Officially, I'm in support of it.

>
> If you're not ISP / IP carrier:
>
> Are you planning to obtain an IPv6 PI prefix as per the proposed
> policy and then deploy, say, 25% or more of the services you now
> provide over IPv4 over IPv6 using those addresses? And if so, how
> soon would this happen?
>
> If you are an ISP / IP carrier:
>
> How many people that are going to do the above are you going to
> provide IP transit for? And how soon do you expect these people to
> start doing this?

Being that my day job is in R&E regional connector, I would say that  
all of my customers currently running IPv6 would either apply for  
2005-1 or just apply for a PI since they are institutes of higher  
education (if they haven't already).  All are already multi-homed for  
v4 and will likely want to do the same once they are reliant upon v6.

Beyond that, I would guess a good half of my customers will also  
follow suit and want some form of PI space, again either under 2005-1  
or under the fairly easy process as a research institution.  As to  
their timeline, that's really the whole question of all of these  
conversations.  Today they have no real need since there are no real  
services (you can only watch a swimming turtle or listen to Virgin  
Radio for so long), so most don't have it on their radar.  If there  
were meaningful services for them to utilize v6 then the story would  
be different.  Which is in part why helping the multihoming concern  
forward with 2005-1 is important, since that stumbling block is one  
of the detractors from v6 adoption.

David


> _______________________________________________
> PPML mailing list
> PPML at arin.net
> http://lists.arin.net/mailman/listinfo/ppml