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

* Hans Petter Holen

> for me the following are open issues for me

> -1 how it will affect inter region transfers - if this is important 
> to the community 1 - presently it will not be simple to establish an
>  inter-RIR transfer policy with Arin. Personally - based on previous
>  discussion at Ripe meeting and other places - I do not think the 
> Arin communicty will change opinion on this, at least until they run 
> out.

To answer the "is it important to the community" you don't have to look
further than to proposal 2012-02, which, if accepted, would allow for
inter-region transfers with ARIN (and APNIC). However, this proposal has
gone nowhere; there appears to be practically no interest in it.

Sandra Brown, the author of 2012-02, even stated in an earlier message
that the proposal had been withdrawn. The www.ripe.net PDP pages does
not (yet?) reflect this, but nevertheless the proposal is beyond any
doubt "on ice".

So the answer seems to be a resounding "NO".

> This could be an issue as the "hidden reserves" of "legacy" 
> assignments are in the Arin region.

Up until the moment of depletion, APNIC was using IPv4 faster than any
other RIR by a huge margin. I think it reasonable to conclude that by
now there must be a huge unmet need for IPv4 space in the APNIC region.

APNIC and ARIN happens to share compatible inter-region transfer
policies. Yet, the number of inter-region transfers going from the ARIN
region to the APNIC region is negligible - at RIPE66 the total amount of
transfers was reported to be 11 (5 per annum, if extrapolating
linearly). So the notion that there is some big hidden reserve sitting
around in the ARIN region, ripe for the taking and ready to transform
our current our state of scarcity into one of abundance (if we only pass
an inter-region transfer policy!), seems extremely implausible to me. At
least it has not worked out that way for the APNIC region.

That said, if having an inter-region transfer policy that is compatible
with ARIN's becomes important to us at a later stage, it is quite
possible to add back whatever "need basis" is necessary to placate ARIN
then. APNIC prop-096 is a demonstration of exactly this being done.

> -2 how will it affect reclamation of address space by RIRs - and is 
> this important 2 - I think reclamation by the RIPE NCC will decay 
> over time.

While I concur that the NCC's reclamation rate will likely decay over
time, I doubt 2013-03 will make any difference whatsoever in this regard.

There is by now enough unmet demand in our region that all addresses
that are likely to be offered on the market would have no problem
finding a new owner ("need" requirement notwithstanding). Or to put it
another way, the notion that an LIR that wants to sell an allocation
would be unable to find a *single* willing buyer strikes me as
completely unrealistic.

So whatever the motivation an LIR might have for voluntarily
surrendering an unused allocation to the NCC instead of selling it might
be - I don't see how the removal of the need requirement on the buyer's
part could possibly have anything to it.

> -3 how wil it affet the critics of the RIRs 3 - the traders will be 
> happy - he ITUs and critical governments will probably make more 
> noice. how much of an issue is this to us?

I don't doubt the ITU and "critical governments" will make whatever
noise they can, on whatever grounds they can, regardless of the validity
of the arguments they may muster. This is nothing new, though, it's an
ongoing struggle that we'll have to deal with anyway.

Put it another way, I highly doubt that *not* passing 2013-03 is likely
to make the ITU and critical government cease to be a thorn in our side.
(They might as well take any failure of 2013-03 to form another attack:
«Your IPv4 policies are clearly out of touch with reality and your
community has proven itself unable to fix them. We'll take it from here,
thankyouverymuch...»)

Furthermore, there are plenty of other policies, services, and events
that these critical entities could use to mount attacks in some way or
another. Some examples: 2006-02, 2007-08, 2009-06, 2010-02, 2012-02, the
IPv4 Transfer Listing Service, the Recognised IPv4 Transfer Brokers
listing, Resource Certification, the IPv4 depletion event itself, and
indeed 2013-03. I'm not claiming that any attacks against the NCC or the
community based on argumentation stemming from any of the above would be
at all valid - just that 2013-03 is not unique in this regard.

In the end I don't think we should make (or not make) policies in order
to kowtow to the ITU or any government entity, in the hope that they
will not try to "take away our toys". If they want our toys, they'll
come after them anyway, and we'll have to stand up to them. What gives
us legitimacy as a community (and by extension the NCC), is that the
policies we make are for the benefit of the Internet, and not in the
pursuit of any political power struggle goals.

Best regards,
Tore Anderson