[arin-ppml] Draft Policy 2009-1: Transfer Policy - Revised andforwarded to the Board

[William Herrin]

On Thu, May 7, 2009 at 10:37 AM, Lee Howard <spiffnolee at yahoo.com> wrote:
> Which of these statements better reflects your position?
> If an organization can use NAT44, they SHOULD.
> If an organization can use NAT44, they MUST.

Hi Lee,

I'd say: none of the above.

I think wading into a redefinition of "need" prior to actual free pool
exhaustion would be begging for trouble if it was doable at all.
Doubly so without the other RIRs on board. But let's be eyes-wide-open
about it and not pretend that the orgs abusing this and similar
loopholes aren't actually hoarding. And let's start thinking about how
we're going to redress this in a practical, moving-forward manner once
the free pool is gone.


On Thu, May 7, 2009 at 12:50 PM, Milton L Mueller <mueller at syr.edu> wrote:
> Hoarding, no. You are misusing the term for emotional purposes.

Milton,

Hoarding is when, upon observing the scarcity of a desired resource,
you acquire and store a quantity of the resource against a future
need, thereby depriving others. It's a form of waste.

Let me put it another way: scarcity + waste + retention = hoarding. If
you waste a scarce resource in a way that allows you to keep the
scarce resource, that's hoarding.

Still another way to define hoarding: knowing a resource is scarce,
you choose to acquire far more of it than what you could have gotten
by with.

What my favorite vendor has done is akin to emptying the local stores
of flashlighs before a hurricane and then saying, "well, it's not
really hoarding because every flashlight we bought we actually used,
turning it on and off at least once."

Actually, that last paragraph isn't really fair. My favorite vendor
hasn't stepped forward and defended its choice; the defense comes
primarily from 3rd-party apologists.


On Thu, May 7, 2009 at 3:30 PM, George, Wes E [NTK]
<Wesley.E.George at sprint.com> wrote:
> This is not about profit, it's about breaking even on what
> would otherwise be expensive, but *optional* work for
> many organizations to reclaim IP space for the good of
> the Internet community.

Hi Wes,

With respect to folks who got their IP addresses before scarcity was a
looming emergency, I wholeheartedly agree. And I've said so serveral
times.

For the IPs that went out in the last year or two, well, not to put
too fine a point on it but they should have known better. If you're
trying to get a /22, well okay. That's our fault for not allowing /24
assignments for multihomers. But a /9 that didn't need to be?


> So...using all of the available private space, I get a grand total of 17.1M addresses.
>
> VZW subscriber count = 86.6M
> Adding, so that we don't keep singling out VZ:
> ATT subscriber count = 78.2M (6M of which are iPhones)
> Sprint subscriber count = 49.3M
>
> At best, we're talking about nearly 3:1 oversub on addresses, at worst, 5:1.

Oversubscription is the wrong model here. A reasonable network
architecture at the current time wouldn't put the 1M addresses
available just in 172.16/12 through a single stateful firewall, NAT or
otherwise. The technology to make that practical is still a ways off.

That means that in order to support a stateful firewall, you're going
to segment the network into smaller groupings, each of which could
reasonably resuse the same RFC1918 IPs.

Yes, there are non-trivial architectural differences between a network
that is generally segmented and one which must be completely
segmented. You don't need to run down the list. Nevertheless, the step
between modest packet filtering and a stateful firewall is a whole lot
larger than the step between a stateful firewall and a stateful NAT
firewall.

And for the record, I support releasing the class-E space as
additional private-IP space in order to make large-scale NAT easier.
Legacy system issues make class-E just about worthless for use on the
public Internet but they could well be valuable inside the controlled
environment of a NAT interior.

Regards,
Bill Herrin



-- 
William D. Herrin ................ herrin at dirtside.com  bill at herrin.us
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004