[arin-ppml] New Entrants shut out? (Was: ARIN-2011-5: ... - Last Call
owen at delong.com
Mon May 2 19:05:40 EDT 2011
> I get the feeling that you see IPv4 as a lame horse that needs to be
> taken out back. While that's certainly one approach, my logic has
> always been to allow the free exchange of IPv4 until such a time that
> it becomes naturally cost prohibitive. CxO's talk $, and letting the
> market make IPv4 costly is a quick and dirty way to convey the
> benefits of IPv6.
Not an entirely accurate characterization of my view.
I see IPv4 insufficiency as an inevitable consequence of continued
growth of the internet. To claim otherwise is to assume that one
of the following sentences must be true:
1. At some point in the very near future, the internet will stop
growing and demand for more public addresses will cease.
2. There exists a technology which will provide access to all
required internet services without requiring additional address
resources to meet the demands of continued growth.
3. There will exist some combinations of limited growth and
technologies which will provide a globally acceptable
subset of current internet services so as to remove the
need to adapt to a protocol which provides an address
larger than 32 bits.
I support the exchange of IPv4 until such a time that it becomes
naturally cost prohibitive so long as those on the receiving end
of that IPv4 exchange meet the same requirements that would
exist for them to get space from the RIRs while there is still a
I do not support the creation of market speculation in IPv4 addresses
as a way to inflict pain on CxOs as a motivation to deploy IPv6.
While I agree it would likely be effective and, indeed, I could probably
benefit greatly from such a move, I think it is contrary to the interests
of the wider community and would be an unnecessarily disruptive
approach to solving the problem.
> At the risk of getting too far off track, I give you this scenario.
> There may be major ISP's who were early entrants and managed to obtain
> more IP's than they ever really needed. Perhaps they have austerity
I would say that there definitely are. However, what they ever really
needed and what they will need in the coming future are two different
> plans that allow them to keep chugging on IPv4 for years. Perhaps
> their fellow competitors will do the same and IPv6 will fail to gain
> immediate traction. It seems that only small to medium sized
I tend to doubt this. The number of things that are already moving
to IPv6, the increase in the IPv6 growth curve since February, and
the rate at which IPv6 is being deployed all point to the fallacy of
such an argument.
Finally, there is not a sufficient critical mass of those organizations
to govern the actions of the global internet overall. Instead, they may
become islands disconnected from what will rapidly become the
majority of the internet until they migrate their environments to IPv6
> businesses are taking IPv6 seriously. Advertising IPv6 support seems
> like more of a marketing gimmick than a realistic push for support by
> many companies.
Interesting... Care to name some names? (I'm not above a name
and shame of vendors that aren't doing the right things with IPv6
at this point). (Are you paying attention, Linksys?)
I will point to http://www.he.net as a counter-example.
Yes, we're getting lots of marketing mileage out of our IPv6 support,
but, it's most definitely a genuine push for support on our part.
I think you would be hard pressed to claim that our IPv6 support
is in any way incomplete or strictly for purposes of marketing.
I gain no marketing advantage of the fact that my home is fully
dual stacked (except a few devices that can't do IPv6 yet)
other than being able to say so when I give IPv6 talks.
What I gained instead is:
1. Assurance that I won't be left behind as the world moves.
2. Experience and knowledge on IPv6 operations.
3. The ability to talk with authority about real world IPv6
deployment experience. (Yes, this took some additional
experiences outside my home, too, but, those early
experiences dual-stacking my house helped a lot).
More information about the ARIN-PPML