In message <6E447820-48E8-11D6-AC4B-00039357A82A@extremenetworks.com>, RJ Atkin
son writes:
>
> On Friday, April 5, 2002, at 10:49 , Curtis Villamizar wrote:
>
> > Ran,
> >
> > I doubt that the IAB was suggesting...
>
> Sure, but not sure how that relates to the note I sent earlier today,
> which was about venue for conversations -- not which conversations
> should be held...
>
> Maybe you can educate me offline ?
>
> Ran
Ran,
It seemed like you were suggesting that to be a "clean slate" the
topic of evolving the network could not be discussed and that topic
should only be discussed within the IETF. Where did I go wrong in
interpreting what you said?
My personal opinion is that an "evolutionary approach" is desireable
as long as it does not constrain the outcome to the point where the
outcome does not meet other stronger requirements. We should be able
to deal with the idea of requirements and desireable characteristics
that are secondary.
Curtis
ps - We're nitpicking in discussing the "clean slate" requirement but
if we wanted to be perfectly clear:
4.10 Clean Slate
For the purposes of development of the architecture, we assume
that there is a 'clean slate'. Unless specified in section 3,
we have no explicit requirements that elements, concepts or
mechanisms of the current routing architecture are carried
forward into the new one.
Frank could (if he wants) add:
Nor do we prohibit concepts or mechanisms of the current routing
architecture being reused simply on the grounds that they were
used in the prior architecture.
Date: Fri, 05 Apr 2002 10:49:35 EST
To: RJ Atkinson <rja@extremenetworks.com>
cc: avri <avri@sm.luth.se>, rrg <irtf-rr@puck.nether.net>
From: Curtis Villamizar <curtis@workhorse.fictitious.org>
Subject: Re: Evolution and the routing architecture
In message <B0F06812-489B-11D6-9A67-00039357A82A@extremenetworks.com>,
RJ Atkinson writes:
>
>
> I have seen several folks (me included) suggest that an evolutionary
> approach was not the kind of thing that belonged in the iRtf, but
> instead that evolutionary work more nearly belongs in the iEtf.
>
> This is very much different than the summary above, because this speaks
> to *where* a given approach should be undertaken, not which is or
> isn't the right answer for the global Internet at a particular point
> in time.
>
> For my own part, I think both approaches should be persued,
> though it seems mighty odd to be trying to undertake both in a single
> Routing Research Group. Maybe the right approach is to split into
> 2 separate RGs, with different leadership (NB: such a decision could only
> be made by the IRTF Chair).
>
> I'm not speaking for the IAB, but my understanding was that
> the IAB thought that evolutionary work would be handled within the
> usual IETF processes and that the role of the IRTF RRG was to undertake
> a more revolutionary ("clean sheet of paper") approach.
>
> Perhaps the RRG Chairs can clarify where which of us are confused
> on scope and charter ?
>
> And maybe the Routing ADs can clarify whether they object to folks
> undertaking normal evolutionary work inside the IETF Routing Area ?
>
> Cheers,
>
> Ran
> rja@extremenetworks.com
Ran,
I doubt that the IAB was suggesting that any considerations of
practicality to assure relevance no matter how secondary are out of
scope for the IRTF and discussion of network evolution entirely
prohibited. I'm quite certain they were not suggesting that any
architecture for which an evolutionary approach is possible should be
eliminated from consideration on that ground.
Curtis
This archive was generated by hypermail 2b29 : Mon Aug 04 2003 - 04:10:04 EDT