[address-policy-wg] [policy-announce] 2013-03 New Draft Document and Impact Analysis Published (No Need - Post-Depletion Reality Adjustment and Cleanup)

Hi Filiz,

One question for clarification:

> And to better address the need based concerns objecting your proposal, I
> think you could consider taking the "intent" you mentioned above one step
> further and have it explained to the RIPE NCC.
>
> Accordingly, I think following will be a more appropriate wording:
>
> 3. LIR must demonstrate its need for the IPv4 address space and must
> confirm it will make assignment(s) from the allocation.
>
> replacing what you proposed:
> 3. The LIR must confirm it will make assignment(s) from the allocation

What is your motivation for adding the 'LIR must demonstrate its need for
the IPv4 address space' part?

As the RIPE NCC can currently only allocate /22's the demonstrated need
will have no impact on the allocation. Those that demonstrate a need of 1
address will get the same /22 as those that demonstrate a need of a
million addresses.

Your suggested text doesn't seem to have an impact on _what_ the NCC will
allocate. It does have an impact on _when_ the NCC will allocate though.
LIRs with existing allocations won't need the new /22 allocation until
they used most of their existing ones. This only seems to affect the
runout speed of the remaining /22's.

Looking at that: the RIPE NCC currently
(http://www.ripe.net/internet-coordination/ipv4-exhaustion/ipv4-available-pool-graph)
has more then 14000 /22's left (not including quarantine and reserved).
There are less than 9000 LIRs that can have allocations from before the
runout. Some of them already have their /22. The remaining ones might be
able to get their /22 sooner with the current policy text. More than 5000
/22's will remain even if they do.

With my chair hat on: I have no opinion on your suggested change, I'm just
interested in what effect you want to achieve with it.

Thanks,
Sander