ARIN Justified...

Clayton Lambert Clay at exodus.net
Wed Jan 10 19:50:55 EST 2001


I think there is some virtue here:

-----Original Message-----
From: Bill Van Emburg [mailto:bve at quadrix.com]
Sent: Wednesday, January 10, 2001 3:54 PM
To: Clayton Lambert
Subject: Re: ARIN Justified...


Clayton Lambert wrote:
>
> IPv6 is not the panacea you seem to think it is...
>
> With a mentality like that, we'd burn thru IPv6 in 10 years or less...
>

I think I missed something here....

Wasn't Allen proposing that the /20 minimum allocation encourages
organizations to attempt to justify larger space than they really need?
I believe he was suggesting that allocations like /22 or /24 would
discourage IP waste, since people could get their direct allocation,
bypassing some of the issues he speaks of with an upstream provider.

....I think /20s are better than /19's for minimums...And as far as process
overhead ....(due to larger routing tables and such)

I don't believe that this is feasible, given the nature of routing and
current HW capabilities (not to mention the capacity of the ARIN
organization), but I really wouldn't know the answer to that
question....

....I think this brings up a good point:  The IP policy should differ
allocation requests to the maintainer, and escalations should go to ARIN, as
in the IP policy recommendation I posted earlier.

Do you really think there is danger of blowing through the IPv6
allocation?  Even the IETF proposal, although I think it went a bit too
far, would have a hard time pressing through the entire IPv6 space, and
would have plenty of room left if a policy change were enacted after
initial allocations were made....

....If we take the attitude of "hey, we could never possibly burn up all the
IPv6 space" then...Yup, I bet we burn thru it in less than 25 years.
I still think IPv6 is a bad idea anyway...I think we proved the
inept-ability (yes, I invented that word, but it fits) of Classfull
addressing in the early days of the net...didn't we?
...Of course, thats a different topic  ;-)

-Clay





More information about the Vwp mailing list