[ppml] Policy to help the little guys

[David Williamson]

On Wed, Mar 19, 2008 at 09:03:35AM -0700, Stacy Taylor wrote:
> B) If
> 'helping the little guys" means routing table bloat we need to be against
> that.
> and C) I will always oppose "I want a pony" microallocation proposals in the
> v4 space.
> Stacy

I'm also against bloat and ponies, but I am in favor of efficient
utilization of the remaining IPv4 space.

To be clear, any proposal that I'm likely to support would require the
recipient of a PI /24 to be multihomed.  Heck, I'd be happy if there
was no non-multihomed PI available.  I don't quite get why large
singlehomed PI space is permitted under the current policy.  If you
don't need to be multihomed, get space from an ISP.

Perhaps I'm overly simplifying the world, but I see organizations in
one of three roles:

* ISPs
* end sites that are multihomed
* end sites that are singlehomed

The first two have much in common, networking wise.  The last of the
three is what the first would call a "customer".  The only difference
between the first two is that the second one doesn't have downstreams
in the third class.

In theory, under current policy, an ISP could deaggregate their entire
space down to /24s, and hand them out to downstream multihomed
customers as part of their service bundle.  I don't see how that's
any worse for the routing table than permitting PI /24s.

-David