The Kobe "revolution"

Dave Crocker dhc at dcrocker.net
Sat Mar 22 12:59:27 CET 2003


Folks,

The changes made to the IETF in 1992 are getting cited quite a bit.

It is probably worth reviewing what did -- and did not -- happen then.

As revolutions go, the time of Kobe was impressively mild.

Strategic. Essential. But mild.

Let's review:

We had a structure.  We had a process.  We had been operating in that
mode for 3 years, or so.  Standards were developed in a certain way
and approved in a certain way.

So what changed?

1.  Ultimate approval for standards moved "down" the EXISTING
hierarchy one level, from the IAB to the IESG.

    The IESG had already been providing process management, and
    technical facilitation. What it had *not* had was what one might
    call "authority". With the change, it got a veto, though this was
    cast as the power to require further review, discussion and
    refinement.

    
2.  Those holding an AD or IAB position became subject to regular
review and replacement.

    Previously, positions were held until the occupant tired of the
    job.  With Kobe, we put in the 2-year review, renewal/replacement.

    
3.  Nomcom was invented to effect the renewal/replacement process.

    Nocmom is peculiar to the IETF. Its design represents a very
    careful attempt to add a review body, but not alter basic
    characteristics of the IETF. Given that we have no membership, we
    could not "elect" our leadership through a more typical mechanism.

That's all folks. Nothing else was done.

It is worth noting that Dave Clark's famous IETF tag line:

   We reject kings, presidents and voting.
   We believe in rough consensus and running code.

Did not declare something new. Rather he was documenting existing IETF
culture and practise. He was reminding us of core IETF principles.

Some of the running code that the IETF was careful to use and respect,
during the Kobe change, was the design and operation of the IETF. We
only changed what was essential, without losing what was essential.

d/
--
 Dave Crocker <mailto:dcrocker at brandenburg.com>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 Sunnyvale, CA  USA <tel:+1.408.246.8253>, <fax:+1.866.358.5301>



More information about the Problem-statement mailing list