[ppml] Policy to help the little guys
owen at delong.com
Wed Mar 19 16:18:47 EDT 2008
On Mar 20, 2008, at 3:41 AM, Scott Leibrand wrote:
> Randy Bush wrote:
>> David Williamson wrote:
>>> If you can get a PA /24 for multi-homing, and anyone can, why can't
>>> you get PI?
>> as a gedanken experiment, and maintaining my reputation as a
>> trouble-maker, why not longer than /24? e.g. take a /16 and create a
>> well-known micro-swamp of say /29s for multi-homed content sites.
>> i am not necessarily advocating this. but i think the exercise of
>> discussing the pros and cons might be good for us.
> I can think of a few reasons to stop at /24 for now, but I think the
> most important is that IMO policy shouldn't force ISPs to liberalize
> filter policy, but rather should respond to it. To date, most ISPs
> accept PA /24's, so allowing PI blocks down to /24 won't change things
> much. But giving out PI blocks smaller than /24 would create a lot of
> additional pressure on ISPs to accept the smaller blocks.
ARIN policy cannot force ISPs to do anything with regard to
routing, so, this argument is specious at best.
ISPs can filter on whatever boundary they was, and, if you
don't believe that, I point you back to your statement that "MOST"
ISPs accept PA /24s. If ARIN policy were "forcing", then, the ARIN
policy allowing a PA /24 strictly on the basis of being multi-homed
would require that ALL ISPs accepted PA /24s.
I don't pretend to know what the best stopping point is, but,
I will say that the disconnect in prefix size between 18.104.22.168 and
22.214.171.124 strikes me as nothing short of hypocrisy.
> I would rather focus on matching PI policy to what ISPs have already
> demonstrated they're ok with for PA space, rather than having PI lead
> the way. If ISPs decide in a year or two to start relaxing the /24
> boundary, then I think it would be appropriate to adjust the PI
> as well.
I would rather focus on having policies that meet the needs of
legitimate multi-homed end-users and let ISPs manage their
More information about the ARIN-PPML