[arin-ppml] Policy Proposal 93: Predicable IPv4 Run Out b Prefix Size - Revised
Joe Maimon
jmaimon at chl.com
Mon Jun 22 17:43:09 EDT 2009
Bill Manning wrote regarding utilization:
The transfer system is intended to wring inefficiencies out of previous
utilization measures. It will probably succeed to the extent that our
routing system can accommodate it.
However, I believe complete reliance on a transfer market driven
situation is not advantageous or necessarily equitable to all players.
We need to hedge our bets.
Martin Hannigan wrote:
> On Mon, Jun 22, 2009 at 3:24 PM, Joe Maimon<jmaimon at chl.com> wrote:
>
>> Martin Hannigan wrote:
>>
>>>> This proposal is better than nothing and might win some consensus this
>>>> time
>>>> around.
>>>>
>>>>
>>> I think that's unlikely myself.
>>>
>>>
>> Rather sad commentary, since the only thing shining through is the attitude
>> "if I cant have it no one can".
>>
>
>
> Quite the opposite. My comments and suggested alternatives are
> generally considerate of a) the region b) continuing to find some
> method of providing v4 space to everyone in this region until final
> exhaustion and c) the rest of the regions. After regional exhaustion,
> markets will undoubtedly take over entirely with regards to v4. This
> is likely to occur in all RIR regions.
>
I dont think complete reliance on a transfer system that could as easily
lock out small players as it could larger players is a defensible
position in light of ARIN's goals.
>>> We could instead simply raise the minimum allocation unit and raise
>>> the bar of entry _for PI_, eliminate end user allocations altogether,
>>> and start pushing people towards PA.
>>>
>>> Wouldn't that equalize the market for new entrants?
>>>
>> No, since there isnt any more space available after 5 to 10 of the 80 large
>> organizations per year get their requests fulfilled.
>>
>
> Their requests would include the needs of their customers.
>
Only by revocation of PI availability. This sounds like pouring salt
into the wounds of the small organizations.Which is all moot at the rate
it is going anyways.
>
>> Making it easier (or harder) to get requests approved when there is no space
>> to fulfill them is silly and pointless.
>>
>>
>>> They should be
>>> able to operate an ISP as long as they can get numbers. Period.
>>>
>>>
>> Exactly, and we should ensure that they can continue to do so.
>>
>
>
> Getting space from their upstream does not accomplish this?
>
No, there is a reason PI exists.
> -M<
>
>
Joe
More information about the ARIN-PPML
mailing list