[ppml] Policy Proposal 2005-8: Proposal to amend ARIN IPv6 assignment and utilisation requirement - Last Call

[Lea Roberts]

Randy -

On Sat, 15 Apr 2006, Randy Bush wrote:

> >>> the /64 boundary is hardwired in a lot of places and so that is
> >>> contraindicated.
> >> it was specified NOT to be hardwired.
> > I haven't tried it...  I know in the discussions leading up to 2005-8,
> > that there are alot of mechanisms, some previously specified by Jordi
>
> uh, as much as i like jordi, he does not specify ip architecture

thanks for pointing out the errors of my poor word choices.  I will
certainly try to consider those choices more carefully in the future so as
to not unintentionally convey inaccurate implications.

> > that have been specified assuming the /64 boundary, so trying to move that
> > would have had a high pain ratio.
>
> a - it is specifically NOT specified in architecture.  architecture
>     specifically says (or used to say) that it should not be a hard
>     boundary in implementations.
>
> b - many of us use p2p links of /126 today

I readily acknowledge it's possible if one is willing to do without those
mechanisms which have enumerated elsewhere (thanks, Owen :-).  I stand by
my assertion that obtaining more address space should never be difficult
or expensive, so that getting an additional or expanded assignment would
be the normal way for dealing with the need for more subnets.

one question for you, Randy:

is 2005-8 at least moving in the correct direction for IPv6 address
assignment policy or do you think the current RFC3177-based assignment
policies should continue?			thanks, 	/Lea