[arin-ppml] Draft Policy ARIN-2013-3: Tiny IPv6 Allocations for ISPs
owen at delong.com
Thu Apr 4 17:34:00 EDT 2013
On Apr 4, 2013, at 12:42 PM, Matthew Kaufman <matthew at matthew.at> wrote:
> On 4/4/2013 11:57 AM, David Farmer wrote:
>> It all depends where they started using there blocks. The current use case for this policy is to reduce your holding because of the fee schedule
> Doesn't this just mean that the new fee schedule is broken?
Yes. However, we haven't yet found a good way to fix it, so we're contorting policy a little further to attempt to address the issue instead.
> IPv6 addresses are by no means scarce... why are we treating them that way? If we gave someone a /32 and they aren't using all of it yet, they should just keep the /32.
As was discussed earlier (and believe me, I'm no fan of this, but it does seem the best available outcome at this point), there are too many ISPs in the /32 category to sustain ARIN if /32 falls to $500/year. OTOH, there is a strong desire to provide support for ISPs that are in this XX-Small category in IPv4 (/22 or less) to be able to add IPv6 without having to pay more than the new fee structure for their IPv4 holdings.
>> , but I'd prefer generic rules that allow flexibility for future conditions.
> Agreed. How about "you can't give us back anything but what we gave you" and "we won't have stupid fee schedules that ever encourage you to want to do otherwise".
This is a discussion about the manner in which they can give back part of what we gave them, so I'm not sure how the first part isn't already addressed. As to the latter, if you have a better fee structure proposal, I suggest submitting it to the ACSP and I'm sure the board will consider it.
>> Also, this is the current operational practice we have now, to allow the LIR to select which /36 to retain. If someone started using their /32 in the middle why force them to renumber?
> Why force them to give up the /32 at all? Are we running out of them or something?
No, it's all about the politics of money.
More information about the ARIN-PPML