[ppml] Policy Proposal: IPv4 Transfer Policy Proposal

[Scott Leibrand]

michael.dillon at bt.com wrote:
>> Yes.  There is a large installed base for whom migrating to 
>> IPv6 may be painful (expensive).  Reducing the community's 
>> total expense for operating their IP networks is a benefit to 
>> the Internet community.
>>     
>
> RED HERRING!
>
> Given the fact that ISPs who implement IPv6 transit service
> will also have to implement transition mechanisms such as
> Teredo, 6to4, etc., there is no imperative for any part of
> the installed base to migrate to IPv6 before they are ready.
>
> The imperative today is for those organizations with steadily
> growing networks at the heart of their business model (ISPs) 
> to begin transitioning. Whether it is painful or not, they must
> do it or die because network growth is fundamental to their
> being.
>   

I think you make an unjustified assumption that ISPs will be able to 
deploy IPv6-only service to new customers and operate transition 
mechanisms such as Teredo, 6to4, etc.

An alternative, and IMO more likely, transition mechanism is to use two 
much more familiar technologies: dual stack and NAT.  Many consumer ISPs 
will be able to start providing IPv6 support, provide the ability for 
their customers to dual-stack, and then start providing NAT'd RFC1918 
addresses to their DHCP customers instead of public IPs.  This will be 
fairly easy for some ISPs, and more difficult for others (such as 
business ISPs, with more static-IP customers and smaller DHCP pools).  
In addition, there will be a great diversity of IPv6 support in 
middleboxes (home routers, enterprise firewalls, etc.)

Given this diversity in cost of migrating to IPv6 (dual-stack) and 
reducing IPv4 demand, there is an opportunity to allow organizations for 
whom the migration cost is higher to delay migration until IPv6 
technology is better developed/deployed, and in the mean time get IPv4 
addresses from other organizations for whom the migration cost is 
lower.  Additionally, such a transfer policy would provide an incentive 
to encourage organizations to migrate their installed base to some form 
of IPv6 where it's easier to do so, rather than requiring growing 
networks to do so if it's more expensive for them.

> If an ISP decided to try and avoid implementing IPv6 by getting
> IPv4 addresses from other sources, they are simply backing
> themselves into a corner and relying on their competitors to
> operate transition mechanisms for them. This is a risky strategy
> since the market segment who are willing to buy IPv4 network
> access will be steadily shrinking. In addition, their existing
> customers will begin to move away because the ISP is perceived
> as being incompetent and at risk of hitting a brick wall.
>   

I suspect that after ARIN free pool exhaustion, all ISPs will offer some 
form of IPv6 service.  To do so successfully, and support dual-stack, 
however, there will be a continued need for IPv4 addresses.  In some 
cases, a network may be able to free up enough IPv4 addresses to allow 
them to transfer them to other organizations.  In other cases, they'll 
be able to free up some, but not all, of their existing IPv4 addresses 
and use the remainder for growth.  In other cases, growth may overtake 
the ability of a network to cost-effectively reclaim IPv4 addresses 
(possibly due to a small install base), so continued availability of 
addresses from other organizations will be essential to avoiding much 
higher transition costs than are necessary.

>   
>>  This policy 
>> proposal allows organizations to choose what's best for them, 
>> rather than forcing a one-size-fits-all solution.
>>     
>
> I disagree that this policy does what you say. In fact this policy
> is trying to set up a market for buying and selling IP addresses.
>   

I don't think we're in disagreement here.  This policy proposal allows 
organizations to choose what's best for them by setting up a market for 
transferring IP addresses.  The addresses aren't property, so no one 
will be buying and selling the addresses themselves, but the underlying 
market economics will be similar.
> Under current policy, an ISP who migrates infrastructure to IPv6
> can return their IPv4 addresses to ARIN. And an ISP who is not
> migrating can continue to apply to ARIN for more IPv4 addresses 
> and receive the returned ones.
>
> This is clear and simple and easy to understand. It is the way
> that IPv4 allocation has always been done and is fully understood
> by everybody who deals with IP networking. Any new policy like
> the one proposed, simply muddies the waters and creates confusion.
>   

Yes, it is clear and simple, but it is not sufficient.  If a network has 
no incentive to go to the trouble of renumbering out of IPv4 addresses, 
they won't return them, and there won't be enough IPv4 addresses to meet 
the needs of those who need more addresses.  However, if the transfer 
process can cover the cost of renumbering, many more organizations will 
choose to do so.  To put it in basic economic terms, if we fix the price 
of scarce IPv4 addresses at zero (by leaving policy unchanged), supply 
will be insufficient to meet demand after the free pool is exhausted.  
If we allow price to be set by a market, the price will rise to the 
point where it increases supply and reduces demand enough to get them 
into balance.

-Scott