[address-policy-wg] agreement

> On 9 May 2016, at 14:25, Arash Naderpour <arash_mpc at parsun.com> wrote:
> 
> In the real world, even when a customer needs for example an /24, they need to become an LIR (and get the /22 from the last /8) as their old LIR cannot provide them with additional blocks. That also speed up the depletion of last /8. have you considered these when you made your objection?

Well I know I did.

Setting up an LIR just to get a /22 is a possibility. The NCC’s checks on new LIRs should be robust enough to identify this sort of behaviour. Questions would be asked if there was an unusual increase in the rate of membership applications and related anomalies.

While it’s possible, it seems unlikely that someone would become an LIR just to get a /24 or even a /22 from the NCC. That would cost them a minimum of EUR3400 in NCC fees: an initial sign-up of EUR2000 and annual membership of EUR1400. Plus some overheads for paperwork, legal fees and so on. That’s roughly $5 per IPv4 address, assuming they got a /22. Which is reasonably close to the price in the secondary market IIUC. The addresses available there generally don’t come with the same overheads that come with NCC membership, so they may well be considered more valuable/attractive.

If people set up LIRs just to get a /22, increasing the NCC (sign-up?) fees would be a quick and effective way to stop that behaviour. ie Once the cost of IPv4 space on the secondary market is less than the NCC fees, why would someone come to the NCC and deplete its pool?

> This policy is not increasing the demand for IPv4, It creates a possibility for small LIRs to receive additional blocks (not from last /8) based on some conditions,

While I oppose 2015-5 in principle, I’m curious why you think the proposal has to favour small LIRs at the expense of other LIRs. And disadvantage future small LIRs once there’s no IPv4 left at the NCC.

That doesn’t seem fair. Unless "something which benefits me and/or my customers at someone else’s expense” is the definition of fair.

BTW, the current policy applies to ALL IPv4 allocations by the NCC, not just those from 185/8. It’s loosely called the last /8 policy. Which is a bit confusing since it applies to more than the NCC’s allocations out of that last /8. 

> so no change in depletion rate from my point of view.

You're mistaken. Please read 2015-05 again. I suggest you pay closer attention to this bit:

b. Arguments opposing the proposal

	Further allocations will speed up the depletion of the free pool.