[arin-ppml] Consensus development for policy (was: Re: ARIN-2012-3: ASN Transfers - Last Call)

[John Curran]

On May 7, 2012, at 11:40 PM, William Herrin wrote:

> Come to think of it, I wonder if this doesn't reflect a deficiency in
> the PDP itself.
> 
> We talk and talk about the wonder of community consensus policy, but
> consensus is a progression of dissonant ideas into a coherent plan
> with mass appeal. No such progression is present in the PDP... from
> first to last it's a majority vote.

The intent is that views which are not in the majority are at least fully
discussed and documented so that any insights or perspectives added can be 
considered by everyone and potentially be incorporated into the overall
consensus position if well-supported.  This is fairly important aspect
of building consensus and the proposed revision to the PDP (sent out for
community consultation in September 2011) call this out more explicitly 
<http://lists.arin.net/pipermail/arin-consult/2011-September/000191.html> -

 "A strong level of community support for a policy change does not mean 
  unanimous; it may be supported by only a subset of the community, as 
  long as the policy change enjoys substantially more support than 
  opposition in the community active in the discussion.  Furthermore, any 
  specific concerns expressed by a significant portion of the community 
  must have been explicitly considered by the ARIN AC in their assessment 
  of the policy change."

When such explicit recognition of minority positions is combined with active 
agreement seeking, consensus results are more likely.

> Perhaps it shouldn't be. Maybe instead:
> 
> 4 affirmatives accept a proposal onto the docket and move it to formal
> discussion
> 8 affirmatives move a discussed proposal to last call
> 12 affirmatives send a called proposal to the board.
> 
> We can talk the talk about consensus, but such with such a progression
> it'd be hard to avoid walking the walk.

It is important that the decision rule to formally recommend policy needs
to have a higher threshold than simple majority, or you can end up with 
oscillations between closely contested positions based on small variations
in participation.  Having several tiers of thresholds based on maturity is 
also an option, and should be discussed more by this community.  As I noted 
at the Vancouver meeting, we have finished incorporating feedback from the
PDP consultation into a revised PDP Update and will be sending those documents
out for community consultation shortly.

Thanks for raising this important issue!
/John

John Curran
President and CEO
ARIN