[address-policy-wg] 2015-01 Draft Document and Impact Analysis Published (Alignment of Transfer Requirements for IPv4 Allocations)

Guten Tag,
> Hi Garry,
>
> It's simple math. Any new LIR would pay 2000 EUR besides the yearly fee.
> I think it can be considered a "hoarding tax" which at this moment seems
> quite considerable when compared to the profit of the "hoarder". We all
> benefit from that money. RIPE needs to keep a stable income therefore
> the membership fee is lowered when more new LIRs are established.

> I was part of the team that had the largest IPv6 deployment in the
> world, long time before the "exhaustion". It's not that easy to
> achieve full IPv6 deployment and I'm sure that most of the buyers of
> IPv4 resources can't deploy IPv6 and even if they do, they can't give
> up on IPv4 yet. Dual stack is the only real solution and it doesn't
> exclude the need for IPv4. If you were at the last RIPE meeting in
> Amsterdam maybe you have heared about a few cases of IPv6 deployments
> and their problems. 
.. just as there have been problems for early ISPs on IPv4 ... what's
the relevance of that in this context?  "It's hard to deploy v6, so we
need to stick to v4!" ???
>>>  - help the last /8 pool become even larger
>> Measures for IP space conservation have ensured availability of
>> addresses over the last ~10 years - if sensible decisions about policies
>> cause push the frame further than previous measures have, I'd say: Job
>> well done! Hopefully, by the time the Internet disables IPv4 there are
>> still IPv4 addresses available for assignment by RIRs ...
> Here I can't agree but I also can't contradict you. There are opinions
> that say if the perspective that IPv4 will really be exhausted it will
> push ISPs to deploy IPv6 sooner. If you tell them that there will be
> IPv4 resources for RIPE to give even in 10-20 years, then probably many
> will say let's see if we live to that time and then we'll make a decision.
OK, maybe we are getting somewhere: Apart from you contradicting
yourself in part, you would consider IPv4 shortage to push v6
deployment. Good. So what do you believe would happen if all RIRs
dropped IPv4 conservation policies tomorrow. Let's say the impeding doom
of no IPv4 addresses available would push everybody to ask for
additional addresses, causing all addresses being used up by December
31st. Do you believe that all ISPs _AND_USERS_ would be v6-ready by that
date? Or Dec 31st 2016? What about 2017? Personally, I reckon if we all
(all ISPs, all users, all IoT devices) made a migration by 2020 I'd be
really surprised ... Sure, impending doom (IPv4 runout) might speed
migration up to a certain degree, but corporations move slow. Heck, we
still have analog modem dial-ups in certain (many?) parts of the world.
Do you really believe the Internet can get around without using v4 any
time soon? And that's the whole point of the policy - ensuring that new
entries to the market - be it ISPs or companies - are still able to
receive at least a basic set of v4 addresses for foreseeable time,
otherwise they will need to find someone willing to sell them addresses,
most likely at some inflated prices ... RIR's policies are the one thing
that keeps prices DOWN, because legitimate use has a calculable pricetag
and does not rely on "free market" ...
> Let's not help the prices raise then. The demand for IPs is supported
> by real needs as otherwise nobody would pay so much money for them. In
> a free economy when you shorten the supply, prices will grow. If there
> would have been a policy that would say let's get back the IPs from
> those who don't use them, that would really help. 
But we have a limited supply - if RIRs didn't put policies in place to
reduce IP use, we would have already run out quite some time ago. Just
by ignoring the fact that there is an IP shortage doesn't make it go away.

-garry