[rfc-i] insufficient engagement on draft-iab-rfc-editor-model-00

Brian E Carpenter brian.e.carpenter at gmail.com
Mon Oct 20 14:07:36 PDT 2008

Aaron, I've read your comments, and John Klensin's,
and I agree that poking a stick into a well-functioning
clock is generally a bad idea. But is that really what
the IAB draft does? I view it more as setting boundary
conditions within which the IAOC can protect both the
IETF's financial situation *and* the smooth functioning
of the RFC system. Maybe the boundary conditions are
over-specified; I can certainly see scope for allowing
the IAOC more discretion about some aspects. So if you
want my bottom line, it's probably that, rather than
agreeing with:

On 2008-10-20 11:28, Aaron Falk wrote:
> I've been watching this thread with interest and concern.  I'm in  
> general agreement with John Klensin's critiques and am developing the  
> opinion that a) this new RFC Editor model is too vaguely defined,  
> possibly with fatal flaws; 

I think it's really not vague enough, in that it doesn't
explicitly allow the IAOC to vary the model in the light of
experience. My bet is that as an RFI/RFP process takes place,
the IAOC will definitely need to adjust the model.

> b) the community (including the IAB other  
> than Olaf) is not engaged in serious consideration of implications of  
> the changes on the table and the possible damage that can be done; and 

I'd guess that most people in the community (excluding those who
are on the rfc-interest list) just want the RFC system to keep
on working, and trust the IAB, IESG and IAOC to make that happen.
>From all my bruises in IETF administrative and process discussions,
I suspect that the readers of this thread add up to 99% of the
people who care about the details.

> c) that the timeline to get to contracting this year will force action  
> before a) and b) can be fixed.

My view is that the IAOC needs the authority to fix a) and that b)
is unfixable. But the IAOC's authority should include the ability
to extend the deadline if necessary; rushing an RFP is never good.

> But my biggest concern is that I sense insufficient engagement.   
> Perhaps, as someone mentioned to me recently, the community is still  
> burned out from AdminRest.  Nevertheless, if this were a working group  
> polling for consensus on a major decision and I was the chair, I'd say  
> there was not enough active engagement to assert there was an informed  
> consensus.  This initiative should not move forward until that exists.

See my comment on b) above. I just don't think you're ever going
to get more than a small subset interested, and they're here.

> --aaron
> ps. While the contract duration hasn't been discussed yet, I don't  
> know if or when there will be an opportunity to provide input.  So,  
> let me add that IMO re-competing the contract every two years is just  
> insane. I think it will take at least that long for any new performer  
> to learn how to do the job well.  So, we should be prepared to live  
> with the results of this process for some time.

That's a good point, but of course if you do switch to a new
contractor, you *must* have an early way out in case of disaster.
A two year contract with straightforward renewal is a painless way
to construct that emergency exit.


More information about the rfc-interest mailing list