ietf
[Top] [All Lists]

RE: Nomcom is responsible for IESG qualifications

2013-03-08 15:19:50
At 03:05 PM 3/8/2013, Eric Gray wrote:
Actually, Joel is not to blame for my understanding.  Sorry to say, I did not 
read his report.

My understanding comes ? at least in part ? from the oxymoronic "oral 
Traditions" written in an 
Appendix in RFC 3777.

--
E

I've included that section in its entirety below.  Where in that section does 
it say that the confirming body must confirm or reject a slate?

Or if you mean "balance" - where in that section does it say the IESG balance 
is even desirable?  (Yup - the below specifically says "IAB"  balance because 
the members of the IAB are at-large members rather than assigned a specific 
role link the IESG members).

Or was there another section of 3777 that you might have gotten your impression 
from?

Mike







Appendix A.  Oral Tradition





  Over the years various nominating committees have learned through
  oral tradition passed on by liaisons that there are certain
  consistencies in the process and information considered during
  deliberations.  Some items from that oral tradition are collected
  here to facilitate its consideration by future nominating committees.

  1.  It has been found that experience as an IETF Working Group Chair
      or an IRTF Research Group Chair is helpful in giving a nominee
      experience of what the job of an Area Director involves.  It also
      helps a nominating committee judge the technical, people, and
      process management skills of the nominee.

  2.  No person should serve both on the IAB and as an Area Director,
      except the IETF Chair whose roles as an IAB member and Area
      Director of the General Area are set out elsewhere.

  3.  The strength of the IAB is found in part in the balance of the
      demographics of its members (e.g., national distribution, years
      of experience, gender, etc.), the combined skill set of its
      members, and the combined sectors (e.g., industry, academia,
      etc.) represented by its members.

  4.  There are no term limits explicitly because the issue of
      continuity versus turnover should be evaluated each year
      according to the expectations of the IETF community, as it is
      understood by each nominating committee.

  5.  The number of nominating committee members with the same primary
      affiliation is limited in order to avoid the appearance of
      improper bias in choosing the leadership of the IETF.  Rather
      than defining precise rules for how to define "affiliation", the
      IETF community depends on the honor and integrity of the
      participants to make the process work.



From: Michael StJohns [mailto:mstjohns(_at_)comcast(_dot_)net] 
Sent: Friday, March 08, 2013 2:57 PM
To: Eric Gray
Cc: ietf(_at_)ietf(_dot_)org
Subject: RE: Nomcom is responsible for IESG qualifications
Importance: High

At 02:15 PM 3/8/2013, Eric Gray wrote:

Mike,

               Notwithstanding your greater direct NomCom experience, it 
seems clear that our understanding
of both RFC 3777 and actual practice differs.

Yup.  And I think I found someone to blame.  Joel (in his Nomcom report) 
mentions a negotiation and agreement with the IAB to confirm or reject the 
slate rather than individuals.  My guess is that the oral history from that - 
wrong - agreement has continued down the line  (via the IAB and Past Chairs) 
to the current day on both sides of the aisle.

Please go back and review the bidding - especially the report done by Dondeti 
for his Nomcom.  At the end of it, look at the Issue 5 discussion.  It is 
clear that the "confirm the slate" interpretation and possible change to 3777 
was considered and rejected.

The Nomcom has repeatedly fallen into this fallacy and has been aided and 
abetted by the IAB.  It needs to stop as it makes the actual process of 
filling positions many times harder.  I've been told privately that this isn't 
a proximate cause of the current set of Transport AD issues, but I've also 
been told privately that this shared delusion has caused much additional work 
with out much additional benefit.


Mike




--
Eric

From: Michael StJohns [ mailto:mstjohns(_at_)comcast(_dot_)net] 
Sent: Thursday, March 07, 2013 7:06 PM
To: Eric Gray; <mailto:ietf(_at_)ietf(_dot_)org>ietf(_at_)ietf(_dot_)org
Subject: RE: Nomcom is responsible for IESG qualifications
Importance: High

At 05:27 PM 3/7/2013, Eric Gray wrote:

In addition to trying to guess what the "talent-set" requirement is for a 
complete slate, the NomCom
also has to try to figure out balance on a lot of different dimensions.  
Company-mix, representation
by regions, extra skills and/or tools each AD might bring to the table, etc.

In fact, having worked through this, the single biggest dread a NomCom might 
face is the potential
that the IAB may decide to exercise a line-item veto on nominated candidates - 
either forcing the
NomCom to effectively start over, or giving the NomCom a clear indication that 
their effort to come 
up with a balanced slate was a complete waste of time.


I'm still trying to figure out where this "requirement" came from.  It seems 
to pop up in each and every nomcom, but is no where in RFC3777.

The phrase in 3777 that is appropriate is:




The confirming body
may 

reject individual candidates, in
which 

        case the
nominating 

committee must select alternate
candidates 

        for the
rejected 

candidates.
Please note NOT an alternate slate, but an alternate candidate or candidates.  
Confirmation is done per nominee, NOT per slate.   The Nomcom MAY NOT start 
over if one of its candidates is rejected.  It MAY NOT pull the slate back.

And this "line item veto" is the only veto available to the confirming body.

Unfortunately, this phrase follows one of the more useless sections which has 
(my opinion) caused no end of harm:




If some or none
of 

the candidates submitted to a
confirming 

        body are confirmed,
the 

confirming body should communicate
with 

        the nominating
committee 

both to explain the reason why all
the 

        candidates were
not 

confirmed and to understand the
nominating 

        committee's
rationale 

for its
candidates.

The confirming body does not have a reason or reasons for why a candidate is 
rejected, it has a vote or result rejecting that candidate.  Individual 
members of the confirming body have reasons, some or all of which they may or 
may not care to state.  The only thing the Nomcom should infer is that the 
confirming body (or a sufficient portion thereof) did not agree with the 
nomcom as to the suitability of that specific candidate for that specific 
position and it should then try again.  To put it succinctly, it's not the 
process it's the person - the nomcom didn't do anything wrong, they just came 
to a conclusion that the confirming body couldn't support and the Nomcom 
should just move on to the next fully qualified candidate for that position.

The nominations and confirmation process is not and should not be a 
negotiation between the Nomcom and a confirming body.

The Nomcom shouldn't spend it's time trying to craft the perfect slate of 
candidates.  It needs to put good people in each of the spots, and if you need 
to sacrifice balance to attain that, then you sacrifice balance and move on.  
In fact balance should only come into play when you have two fully qualified 
people for a slot where it may make sense to take the lesser (but still fully 
qualified) candidate to "balance" the slate.

The Nomcom has a hard job - but it needs to do the job one position at a time 
and not make its job harder.  Pick the best qualified people and move on.  

I say this as a former Nomcom member, former Nomcom Chair and former Nomcom 
past-chair.

Mike

ps -

The difference between a engineer and a bureaucrat is that the engineer takes 
large insolvable problems and breaks them down into solvable pieces, while the 
bureaucrat takes solvable problems and combines them into large insolvable 
masses.  Each and every position to be filled is in someways a solvable 
problem, but trying to find the absolute perfect combination of people (for 
some value of perfect) is possibly close to intractable.  



<Prev in Thread] Current Thread [Next in Thread>