[arin-ppml] ARIN-prop-208 Reduce All Minimum Allocation/Assignment Units to /24

[Owen DeLong]

On May 5, 2014, at 23:49 , William Herrin <bill at herrin.us> wrote:

> On Tue, May 6, 2014 at 2:10 AM, Kevin Blumberg <kevinb at thewire.ca> wrote:
>> Do you support the substantive changes in this policy?
> 
> I support Owen's original policy with the minor tweaks to deal with
> the couple of things he missed.
> 
> I do not support the policy as rewritten. The rewrite is, I believe,
> egregious and harms prospects for further policy development based on
> the distinction between multihomed and singlehomed.

It really doesn't. The distinction could be easily resurrected if necessary (though I think that is actually unlikely to become necessary).

However, having it in the NRPM as a distinction without a difference (as would have been the case in the original policy) would be confusing, unnecessarily complex, redundant.

>> Are there any suggestions you might have to fix the issue that doesn't leave duplicate text in the NRPM?
> 
> Intentionally leave the duplicate text in the NRPM and pursue further
> policy development from there. If there isn't any further policy
> development you can come back and collapse it in a cleanup next year.

We looked at this, and, frankly, leaving it in required more futzing with it than taking it out.

> Eliminating the distinction between singlehomed and multihomed
> entities was not the purpose of Owen's policy proposal and is, in my
> opinion, a bad idea.  Leave the distinction in place so that when we
> re-examine multihomed versus singlehomed in light of the new minimums
> we don't have to re-create it from the whole cloth with all the
> attendant trouble that will cause.

The existing text is archived and can easily be included as a re-insertion with
modifications for any new policy development. None of the section numbers being
retired are being reused at this time. It's actually easier to resurrect it from the archives
and modify to fit whatever future policy development may require it than it would be to
make enough changes to have it make sense in the context of this proposal.

> Seriously, look beyond the immediate policy proposal before you
> consider ripping out huge chunks of the NRPM. It took a long time and
> a lot of debate to craft that language. Don't throw it away until
> we're sure we won't need it again.

The use of the term "huge chunks" is a bit odd to my thinking here.

Ripped out:
	2 sentences of 4.2.1.5
	1 phrase in 4.2.2.1
	2 sentences in 4.2.2.1.1 (unrelated to the single/multihome issue)
	Section 4.2.2.2 (which would need a major rewrite if it were to stay in unless you actually wanted
	the policy to make it only possible to get a /24 if you were single-homed, but you could get a /22 as
	multihomed, which I don't believe was anyone's intent, certainly not mine)
	Title 4.3.2.1 (whois text is moved to 4.3.2)
	Section 4.3.2.2 (which would have to have been almost entirely rewritten if it remained in)
	Sections 4.9 and 4.9.1 (unrelated to the single/multihome issue)

I was on the call with the shepherds and others when these changes were discussed and most of them are actually my own recommendations to prevent the policy from making the NRPM nonsensical. All of the text that is removed can easily be put back by any future policy proposal if it becomes useful to do so, so there would not be any need to recreate said text from whole cloth for any future policy work.

Owen