[arin-ppml] Summary: lowering the ARIN minimum allocation

[Leo Bicknell]

In a message written on Mon, Aug 03, 2009 at 06:49:31PM +0000, bmanning at vacation.karoshi.com wrote:
> 	a) that once the last "greenfield" IPv4 prefix is handed out,
> 	   that any/all policy for IPv4 is dead.

I did not say that.  The policy doesn't magically die.  However,
the discussion here is of a policy that describes a process for
giving things out from the free pool.  If there is no free pool,
or a free pool created only from the voluntary return of address
space then I don't think the policy is going to get used very often,
post-runout.

I also included my own feeling about the future.  IPv4 will end.
IPv6 will replace it.  You can argue if that's in 2, 5, or 50 years,
but no matter the timeframe IPv6 policy matters for a longer period
of time from now than IPv4 policy.

> 	b) there are fundamental differences in how one constructs
> 	   policy for the different address families.

I might remove the word fundametal, but with that removed there are
differences.

Case in point, subnet size does not matter in IPv6.  Under current
policy if you have 1, 100, 100,000, or 100,000,000 boxes in a subnet
you use a /64.  That is not the case in IPv4, those would all get
different sized subnets.

Today many small business get a /27 from their upstream; outgrow
it and add a /27 overlay, outgrow it and renumber into a /24, outgrow
that and renumber into a /23.  In IPv6 they get a /64 (minimum,
maybe a /56, or even a /48), and all those renumbering steps go
away.

So yes, they are different.  Fundamentally different, perhaps not,
but still different.

-- 
       Leo Bicknell - bicknell at ufp.org - CCIE 3440
        PGP keys at http://www.ufp.org/~bicknell/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 825 bytes
Desc: not available
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20090803/f9497d24/attachment.sig>