[arin-ppml] Policy Proposal 2008-6: Emergency Transfer Policy for IPv4 Addresses - Last Call

[James Hess]

[...]
> The lightweightness (?) of the proposal will probably make it the most likely
> to succeed of all the proposals thus far and we should not have an arbitrary
> end date for a successful policy.

It should be trivial to remove an end date on a successful policy
after a policy has shown to be a good one.
Temporarily applying a policy for trial purposes seems to make sense,
and is safer
than adopting a major change of policy outright in some cases.

Without an end date, a policy that turns out to have major problems in
practice that has
not yet been superceded may remain in place for a long time,  with no
good successor, and not enough consensus to throw it away.

On the other hand,  if the policy proves to just be a bad thing,  a
sunset timer all
but forces the bad policy to be re-examined and adjusted.

The downside of a timer is the time  ARIN staff and members will
consume 3 years from now.


If too many policies have such a timer, and turn out to be successful,
a lot of time would be wasted
making proposals and revising policies to remove  3-year-timer clauses.

But I think allowing user-to-user transfers is a huge enough that the
timer is actually warranted in
this case.

 --
 -J