[ppml] Policy Proposal: IPv6 Assignment Guidelines

[michael.dillon at bt.com]

> I remember when IPv4 was thought to be plenty as well.   I 
> understand the view of conservation with IPv6 being hard to 
> swallow this early, but I'd rather learn from history opposed 
> to repeating it. 

Yes, I too would like to learn from history. With IPv4, the question of
"plenty" was a question of opinion. With IPv6, it is a question of
mathematics. There is no shortage of IPv6 addresses and if the world
ever does run out of IPv6 addresses, we will all be dead when it
happens. I am glad to bequeath a tough technical problem to my
descendants to save them from the boredom of a life in the machine.

The lesson of history is that the introduction of VLSM and CIDR made
life a lot harder for everybody in numerous ways. With IPv6 we have the
possibility to have a simple and clean system where every building or
campus or office connected to the network has the same size address
prefix. Moving a network to another point in the topology only requires
changing the prefix, not redoing subnet architecture and total
readdressing. I can live with a two-tiered scheme in which all consumer
homes have a /56 and all non-consumer sites have a /48 because sites
rarely will switch categories.

--Michael Dillon