[arin-ppml] 2014-2 8.4 Anti-flip Language

Andrew Sullivan ajs at anvilwalrusden.com
Tue Feb 25 20:40:55 EST 2014

On Tue, Feb 25, 2014 at 07:09:42PM -0600, Jimmy Hess wrote:

> Have you tried renumbering an IP network, after IP addresses have been
> assigned?

Yes, but not (admittedly) as an ISP, where I can certainly appreciate
the issues.  All analogies are, in some ways, false, or they wouldn't
be analogies.  They'd be identities.

> every IP address is always located at exactly the same position
> within the IPv4 address space.

That's just equivocation on "move", and I don't see how it's relevant.

> The alternatives aren't just imperfect;  they are deficient, in the sense,
>  that many of the  alternatives  don't work today,  cause major problems,
> or can't work.

If you're arguing that IPv6 just isn't an alternative to IPv4 and
can't be used that way, however imperfect it is (e.g. admitting that
there are still lots of v4-only parts of the Internet), then you're
going to need to show not only that v6 is doomed in that way but also
that there is any hope whatsoever in trying to extend the v4 life
meaningfully and that the proposed policy will actually be effective.
If v4 is both as valuable as you suggest and as scarce as we think,
then I think tight regulation is likely to result in a black market
(and I suggest that economic history supports that opinion).  There
are reasons to suppose this is sort of happening sometimes today, and
I think the burden is on those who'd promote tight regulation to
address those possible negative consequences.
> So massive speculation is apparently equivalent to "damage" or "harm".

It's a kind of harm.  There's a trade-off.  Speculation in land and
houses is doubtless a consequence of a market in property.  The
question is whether a particular result is worth the costs imposed by
attempting to address it, particularly since nearly every policy
action has externalized consequences.  One way to prevent land
speculation is serfdom, for instance.  Do we think that would be
better?  We appear to have decided not.

> On the flip side...  ARIN doesn't need to facilitate anyone's attempt to
> create a market in the first place.

The idea that, in a monetized world such as ours, it will be possible
to prevent such a market for real is, I would like to suggest, a
little optimistic.  It seems to me that the question is much more
whether transfers of this sort will happen in public (i.e. registered
with ARIN) or not (e.g. by address holders effectively transferring
without actually doing so).  I would prefer the former result.  I
think that's better stewardship.

Best regards,


Andrew Sullivan
ajs at anvilwalrusden.com

More information about the ARIN-PPML mailing list