[arin-ppml] ARIN-PPML 2014-7

[William Herrin]

On Fri, Feb 7, 2014 at 8:13 PM, Jimmy Hess <mysidia at gmail.com> wrote:
> Agreed.   Also agree that IXPs with only a handful of participants are a
> very easy low-cost renumbering scenario.
> Why should the bar be as low as two or 3 participants?
>
> Why not make the required number  at least 9 or 10 participants minimum,
> with actual documentation for 4 or 5,  before a whole /24 is warranted?

Hi Jimmy,

Personally, I like the number 5. Here's why:

A) I've participated in a couple of IXPs that were more wishful
thinking than reality. By the time an IXP has 5 participants, it's no
longer wishful thinking. I'm no longer concerned that it may fail to
grow, stranding a bunch of otherwise usable addresses.

B) 5 participants plus the IXP's route server fits just so into a /29

C) When it comes time to renumber, the more participants involved, the
more of a PITA it becomes. A handful is not too bad, but coordinating
the action of a dozen folks starts to get messy.

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