[ppml] Policy Proposal: Changes to IPv6 policy - removal of "multiple /48" justification

Stephen Sprunk stephen at sprunk.org
Sat Jan 27 22:37:29 EST 2007

Thus spake "JORDI PALET MARTINEZ" <jordi.palet at consulintel.es>
> In my opinion, policies are not made to be a show stopper, but to
> regulate the situation. If the policy is broken (incomplete, because
> can't be applied due to the lack of an objective criteria), there is a
> problem.
> The solution is not to ignore it until someone suffer the problem. 
> That's
> not fair.

ARIN's implementation of the intentionally-vague policy is to approve 
_all_ requests.  Unless that changes, there is no chance that _anyone_ 
will suffer from this policy by being denied what they need.  The only 
possible suffering is by the community as a whole because we're giving 
out _too many_ addresses without cause, but based on the trivial number 
of reassignments to date, that's not happening yet.  When it does, we 
can come back and look at making things tougher -- not easier.

> A business can't wait for 1 year (my guess about minimum
> time required to succeed with a policy proposal) or even have the risk
> that there is never an agreement in the definition of an objective
> criteria and wait forever.

"Never" and "forever" are strong words.  If there's a problem, we'll fix 
it within a year or two.  In the meantime we're erring on the side of 
being too permissive, and the BOT has the authority to take emergency 
action if we're proven wrong in a way that doesn't allow the normal 
policy process to fix things in time.

I still have not seen any substantive claims that the current policy is 
broken in a way that merits fixing, nor any proposals to "fix" the 
policy that are backed up by real data.  Nobody has even presented a 
case where there's a problem, other than your own hypothetical (as far 
as we know) example of a new ISP that in five years only intends to 
serve <200 v6 customers who are not large enough to get PI space yet 
large enough PA space is not technically feasible.  I'm sorry, but 
that's such a strange example that I have to think you've deliberately 
fashioned it to show a "weakness" in our policies, not that you are 
imagining anyone would actually fit it.

The simple matter is that it's far, far easier to get IPv6 space -- many 
would say too easy -- than IPv4 space under existing policy and people 
still don't want IPv6 yet.  Our efforts should be spent on stimulating 
demand, not spinning our wheels making policy changes to lower the bar 
even further.


Stephen Sprunk         "God does not play dice."  --Albert Einstein
CCIE #3723         "God is an inveterate gambler, and He throws the
K5SSS        dice at every possible opportunity." --Stephen Hawking 

More information about the ARIN-PPML mailing list