[ppml] IPv4 Soft Landing - Discussion and Support/Non-Support Requested

David Williamson dlw+arin at tellme.com
Fri Oct 5 10:54:15 EDT 2007

On Thu, Oct 04, 2007 at 03:27:40PM -0700, David Conrad wrote:
> David,
> On Oct 2, 2007, at 3:08 PM, David Williamson wrote:
> >Does it matter if that happens slower as a result of policy changes?
> The point of a "soft landing" proposal is to extend the runway for  
> transition, but do so in a way as to discourage procrastination.  The  
> answer to whether it matters is likely a subject opinion.  Clearly I  
> think it does (or I wouldn't have bother to make the proposal), but  
> opinions vary.

That explains how your proposal is intended to work, and also explains
why you think it is important, but I don't think it answers my
question.  You note that my question is a matter of opinion...that's
almost certainly true, given that predicting the behavior of groups
of humans is fundamentally hard, even at the scale we're discussing.

Perhaps I'm pessimistic, but I think the extra time achieved from your
proposal is probably just more time to avoid dealing with IPv6.  I
personally really dislike v6 - it's just going to be a PITA with little
real value besides the fact that there's more of it than v4.  That
said, I'm the primary cheerleader with my organization because I
understand the consequences of not pushing the transition *now*.
Organizaions that lack someone pushing them through the transition are
unlikely, in my opinion, to benefit from the delay.

As someone else noted, there's nothing mechanically wrong with the
proposal, but that's hardly a vote of support.

Bill's original question required an answer in the form of a definitive
statement of support or non-support, which I have thus far failed to
provide.  I think I can resolve my fence sitting: "do no harm" is a
nice idea, but "if it ain't broke, don't fix it" is a stronger
criterion.  I don't think this will do much good, so current policy
isn't broken.  That implies there's no need to change anything, which
is effectively a lack of support for this.  It's the best of the lot,
but the problem it tries to solve likely cannot be solved in policy.


More information about the ARIN-PPML mailing list