To be honest, RIRs policy proposals process in the end do not look too
much different than RFCs. It is just another part of the Internet.
This document is not just technical, it also covers matters of resource
allocation, in this case numbers. In that sense I see a value to inspect
current mechanisms in the Internet to do that task (and make them work)
instead off reinventing the wheel.
Also, who is going to use these resources (locators)?
Have you asked them how they want to get them?
Regards,
as
On 04/03/2013 08:15, Roger Jørgensen wrote:
> On Mon, Mar 4, 2013 at 11:04 AM, Jeff Wheeler <[email protected]> wrote:
>> On Sun, Mar 3, 2013 at 10:04 PM, David Conrad <[email protected]> wrote:
>>> I agree that costs shouldn't be dictated. More specifically, I don't think
>>> the draft should get into how the service is funded other than to say the
>>> allocation service can't charge more than the cost to provide the service.
>>
>> Do you know if any RIRs want to provide registration service for LISP
>> EID space? Have you asked any RIR for a cost estimate?
>>
>> As you may know, my opinion is the LISP mapping service is far from
>> ready for Internet-scale use.
>>
>> It would require RIR policy proposal to cause them to begin allocating
>> address space for LISP. The RIRs may guess how much money they would
>> spend on doing it, and you would be amazed how much time and money the
>> RIRs can spend on something that seems to be simple (and probably
>> should be.)
>>
>> I think many members would be inclined to vote AGAINST spending
>> resources on LISP registration services at this time.
>
> You just mention my biggest problem with using the RIR system as it
> is - policy proposal.
>
> As said earlier, this draft mention RIR because that's a starting point, it
> goes
> back to the suggestion I gave on draft-lisp-eid-block-03 [1*]
> We could have written more in the draft on how we should do this but I think
> both you and David Conrad has mention plenty of reason why we should
> consider our options way before we go into the question of who should
> do the job involved.
> If you read my suggestion close you'll see I suggested the RIR as _one_
> option.
>
>
>
>
> [1*] http://www.ietf.org/mail-archive/web/lisp/current/msg04155.html
>
>
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp
