Re: [Eligibility-discuss] Size of nomcom (Re: [Gendispatch] New Version Notification for draft-knodel-nomcom-gender-representation-00.txt)

[Michael StJohns <[e-mail]>]

Hi Harald  -

My thoughts below.

On 11/25/2023 7:08 AM, Harald Alvestrand wrote:
> (This is a distraction from the gender topic, but tangentially 
> relevant since it touches on one of the remedies proposed in 
> discussion. So it gets its own subject line.)
>
> Having served in many forms of committees:
>
> Making a decision in a group of 15 is significantly more work than 
> making a decision in a group of 10.

My experience is that making one large decision is harder in a larger 
group as you suggest.  My counter experience is that making a lot of 
small decisions that require specific knowledge and where some members 
or advisors have more relevant knowledge than others for a given 
question can go quite quickly.  I will also note that standing 
committees tend to take a lot more time to deliberate than task 
committees (such as the Nomcom).

I'd also suggest that larger committees also tend to make use of the 
subcommittee structure.  Allocating a group of members to vet IAB, 
another to vet IESG, the third for the rest and then bringing only the 
best candidates back to the full committee for approval is also a 
strategy.  If you appoint subcommittee chairs, this is also a way to 
possibly train a next or future Nomcom chair.

>
> If we want to minimize the work that the nomcom has to do, we should 
> make it smaller, not bigger.
We can't do that without a pretty huge restructuring of the nomcom 
selection rules - consider both the current discussion and the current 
impact of the max 2 rule

So let's do the experiment.  Two nomcoms, ideally separated by 2 years 
so they don't have intentional people overlap of size 15 or 16.   Have 
the Past Chair be the one responsible for comparing and contrasting 
between the Nomcom they chaired and the larger one. Empirical evidence 
rules the IETF or should.

>
> (This is about the work of coming to a conclusion, not the work of 
> interviewing candidates. But even for interviews, the work of 
> communicating what has been learned from the interview grows with the 
> number of committee members.)

Not really - below the line candidates tend to get weeded out fairly 
early, both from their questionnaires and from the interviews.  Usually 
the first question asked of the interviewers after an interview is "So, 
what do you think, should we spend any time on this one?"  Having say 
three interviewers will tend to minimize any individual bias in that 
process. It's only when you get near the end where you're trying to 
decide between a few very good candidates does the discussion get more 
involved.

Later, Mike



>
> Harald
>
> On 11/24/23 04:42, Christian Huitema wrote:
>> If on the contrary we want to guarantee that every year sees some 
>> non-zero representation of various minorities, we have two options. 
>> We could do a variation of what Mallory proposes, or we could just 
>> increase the size of the nomcom to 15 or 16.
>