[arin-ppml] CGN multiplier was: RE: Input on an article by Geoff Huston (potentially/myopically off-topic addendum)

Owen DeLong owen at delong.com
Wed Sep 14 22:05:13 EDT 2011

On Sep 14, 2011, at 8:53 AM, Chris Engel wrote:

> Ultimately organizations are going to do what feels works best for them, both in the short term and in the long term. Some, perhaps many, organizations may be mistaken about what really is in their best interest...but isn't that their right as free entities? What makes the internet great is it's diversity. It encompasses a huge swath of organizations all with their own individual needs, priorities and interests. I have no doubt that individual organizations will choose to adapt to the looming IPv4 runout in their own individual ways... and I wouldn't have it any other way. Would you really rather have some sort of hive setup where everything was dictated from the top down by some sort of king or queen bee? Yes that might be more efficient in some ways, and it would certainly be alot easier to get dramatic systemic changes like IPv6 enacted...but it would also come with HUGE downsides in many areas. You can't really have it both ways.
> There may, indeed, be some or even many individual organizations that choose to adopt CGN as long term alternative solutions to IPv6. However, if they can make it work well for them and their community of users....what's so terribly wrong with that?  It may not have been the choice you preferred...but that's what it means to live in a free society...you don't get to make other peoples choices for them.

In places where consumers have choice, perhaps nothing is particularly wrong with that. However, there are a great many places where the consumers choice is limited to whatever service they can get from monopoly A or no internet access at all. In such cases, the definition of "works well for their community of users" becomes a very difficult metric to uphold. In even more places, the choice is between duopoly provider A and duopoly provider B where the bandwidth which is available from A and B is substantially different meaning that B effectively has a monopoly on meaningful broadband. This situation is even more insidious because the appearance of choice from a regulatory perspective is present where no meaningful choice actually exists.

> If you want IPv6 to be the new standard then your best bet is to actually make it attractive for people to use, not try to shove it down peoples throat. Aside from


> costs, one of my big stumbling blocks to considering IPv6 for my organization is a lack of support for the sort of functionality NAT/PAT gave me in IPv4. What do I

If you can spell out that functionality, I'll be happy to show you how to do that in IPv6. It may require a certain amount of rethinking your methodology, but, I have yet to encounter an application where you could not achieve just as good a result without NAT using better methods available in IPv6.

> find almost every strong proponent of IPv6's response to that? Is it "Yeah, some-one should really define a standard for that?" or "Hey take a look at what this vendor is working on, it might fit your needs?".  No,  instead I get, "You're wrong for wanting that. That should never be supported by IPv6."  Well guess where I'm going to look to address IPv4 runout if that's the case?  Really, sometimes I think people have no clue how free market systems work. You don't win support for something by telling people their needs/wants/concerns are not legitimate. 

I'm not saying you're wrong for wanting it, but, I am saying that IPv6 probably provides a better method for achieving the same overall result without NAT.

I will say that in general, Address multiplexing NAT is bad, large scale NAT is worse, and double (or more) NAT is even worse than that.


More information about the ARIN-PPML mailing list