Re: ASN draft

[Sean Doran <[e-mail]>]

| Then one has to ask why BGP keeps mutating... :) Yakov?

No, not Yakov.  Sprintlink and ICM are really nasty
environments filled with natural selection pressures.
This encourages the Darwinization of BGP4.

Other natural selection pressures exist in the land of MCI,
Alternet, ANS and elsewhere.

| > Some organizations have lots of ASes.  Some organizations
| > have no AS.
| 
| Yup. So why are we trying to indicate that an ASN is used strictly for
| routing policy?

Because organizations and autonomous systems are not the same.
Administrative policy and routing policy are not the same.

Organizations and ASes may once used to have been the same
(that's debatable), but they simply aren't now.  Insisting
that there's a one-to-one correlation between organizations
and ASes is fantasy.

If you want to uniquely identify organizations with some
handy tag, great.  I think it's a good idea.  However, if
you think that you can do it with ASNs, you're dreaming in
technicolour.

Sorry, but the real world no longer conforms to what you
consider the original model of ASes or ASNs.  The draft
should reflect reality.  So should the RADB.

Adapt, RA, adapt! 

	Sean.