[arin-ppml] Draft Policy ARIN-2013-3: Tiny IPv6 Allocations for ISPs

[Michael Richardson]

>>>>> "Steven" == Steven Noble <snoble at sonn.com> writes:
    Steven> That is exactly my point, if ARIN says that someone
    Steven> requesting IPv6 will not have higher fees, then how does
    Steven> that work with a legacy holder?  Do we want people to adopt
    Steven> IPv6 or not?  A policy that makes it the same cost to
    Steven> request and hold IPv4 and IPv6 works both ways.  If I am
    Steven> charged the same as someone who has both IPv4 and IPv6
    Steven> resources, why would I not request IPv4 resources too? 

+1

And, given that I can get IPv4 for the same price, why go to the hassle
of doing anything with IPv6?   The business case for doing new things
with IPv6 should include "and the address space is essentially free,so
we wilkl do the network architecture correctly, rather than creating a
hack that supported IPv4 NAT/bridging"

Do you know how many layer-2 **hack** have been created because IPv4
subnets are a fixed quantity?  (particularly in the pre-CIDR days)

IPv6 allocations and especially end-user assignments, need to be
essentially free.  In the IPv4 time scale, this is 1989, when
enterprises just came to the table for address space, with *no
intention* of advertising that space.

-- 
]               Never tell me the odds!                 | ipv6 mesh networks [ 
]   Michael Richardson, Sandelman Software Works        | network architect  [ 
]     mcr at sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [