[arin-ppml] Discussion Petition of ARIN-prop-125 Efficient Utilization of IPv4 Requires Dual-Stack

Matthew Petach mpetach at netflight.com
Tue Dec 28 20:45:50 EST 2010


On Wed, Dec 22, 2010 at 12:10 PM, Chris Grundemann
<cgrundemann at gmail.com> wrote:
> The AC should not have abandoned ARIN-prop-125 Efficient Utilization
> of IPv4 Requires Dual-Stack. I petition to move the following text
> forward for discussion on the list and at the next Public Policy
> Meeting. Please support moving this proposal forward now by posting
> statements in support of the petition to this list.

I do NOT support the proposal as worded; I do not think it is appropriate
for the word MUST to be used in a policy such as this.  You state that
the intention of the policy is to "encourage IPv6 deployment" -- yet this
is not encouragement, this is enforced compliance; this is not a guideline,
this is an edict.

I strongly recommend considering carefully the use of words such as
"SHOULD" when writing proposals designed to encourage a behaviour,
and reserve the use of "MUST" for cases only where absolutely needed;
by using the word "MUST", you preclude any option for the ARIN staff
to exercise judgment about the particulars of any request, and encourage
entities to take half-baked measures to meet the rules, rather than taken
a more measured and well-considered approach to deploying IPv6.

I would rather an organization take an extra six months to roll v6 out
correctly, than have them feel coerced to just throw a half-backed
implementation up to get just enough ports pingable via v6 to meet
the requirements which will then take them six more months to back
out before they can move along towards a more reasonable
implementation; ultimately, if the ISP has a bad experience trying
to rush v6 out to meet a requirement like this, they're *less* likely
to be willing to embrace it wholeheartedly in the future.

Encouraging is good; dictating should be reserved only for those
situations in which it is truly justified.

Matt



More information about the ARIN-PPML mailing list