Re: NomCom procedures revision

[Tony Hansen <[e-mail]>]

On 8/28/15 7:13 AM, Harald Alvestrand wrote:
> Den 27. aug. 2015 19:54, skrev Murray S. Kucherawy:
>> Have a look at Appendix F. I plan to fill out Section 10 once we know
>> for sure which changes have consensus rather than a few points that
>> might still be fluid.
> Some notes (this is NOT a detailed review):
>
> Having just gone through the exercise of establishing a voting mechanism
> for my nomcom, I think the following needs to go:
>
> 5.5.  Voting Mechanism
>
>    The Chair must establish a voting mechanism.  The mechanism by which
>    this is accomplished is left to the discretion of the Chair, but must
>    be accepted by at least 75% of the selecting volunteers before the
>    work of the committee can begin.  Once established, this procedure
>    cannot be altered until the current nominating committee is
>    dissolved.
>
> A) It turns out that voting mechanisms are *tricky* beasts. The idea
> that a nomcom will make them 100% right on the first try is a Bad Idea.
>
>
> I suggest changing this to "Once established, the minimum threshold for
> changing the procedure is 75% of the selecting volunteers".

I agree completely that the decision on voting mechanism needs to be a
bit more flexible than "once and done". For example, there are some
voting mechanisms that can yield effective ties. If you don't decide on
how to deal with that up front, are you prevented from adding in a tie
breaker? Another example: it might be that you decide initially on an
overly complicated weighting algorithm, then figure out after using it
that a much simpler weighting algorithm would work just as well. Or you
discover some other facet that "just does not work", because when you
first chose the mechanism to begin with, you didn't have enough
experience to understand how the voting mechanism would be used.

> Note: The 75% acceptance criterion is new. 75% of the selecting
> volunteers is 8 people. That means that 3 people can a) block the
> acceptance of any procedure and b) (with my suggested change) prevent
> any change to the procedure. We should make sure that's what we want.

Do we need to be precisely 3/4? I'm pretty sure that a nomcom I was on a
few years ago used 2/3 majority of all voting members as the criteria.

>
> B) A strict interpretation of "before the work can begin" would have the
> members accept the voting mechanism before they start discussing the
> voting mechanism. I don't think that is a good idea.)
>
> I suggest "before the first vote on candidates is taken" (which is very
> late, but at least it's an explicit point in time).

ack. There's lots more work that the nomcom must do instead of just
"voting on candidates". In particular, the voting mechanism seems like
it definitely does NOT need to be a first-order-of-business topic.

>
> Also, is it an obvious consensus that the chair and the liaisons should
> have no input into the decision to accept the voting mechanism?

I would certainly expect that the chair and liaisons are there to
provide guidance on what they've seen has worked well and not well in
the past for many topics, including the voting mechanism.

>
>
> 5.6.  Voting Quorum
>
>    At least a quorum of committee members must participate in a vote.
>
>    Only selecting volunteers vote on a candidate selection.  For a
>    candidate selection vote, a quorum is comprised of at least two
>    thirds of the selecting volunteers.
>
>    At all other times, a quorum is present if at least 75% of the
>    nominating committee members are participating.
>
> It is not at all clear what a "quorum" does. In our procedures work, we
> found that separating out the idea of "meeting quorum" from "voting
> quorum" made a lot of sense, espcially since we chose to do secret
> ballots only - which means we can't do it in a meeting anyway, and there
> was actually no requirement for the members of the quorum to be present
> at the same time.
>
> It also seemed unusual to have a meeting quorum without either the chair
> or the prior year's chair; we put that into the procedures too.
>
> Note that 4.15, 5.7 and 7.1 still uses "voting member" where it should
> have "selecting member".

There is also a need to be specific about "voting" vs "selecting a
candidate". There are many decisions that the nomcom might vote on
during the course of their business -- the same rules need not apply to
all cases.

With eballots, you can actually require a higher percentage to vote on a
candidate, not tied to any meeting.

I agree that a chair must be present to hold a meeting.



A general comment: I have a problem with all of these numbers being cast
in stone. I can see recommendations being put in with specific numbers,
but not fixing them. Also, I think 2/3 is a more realistic number than
3/4 for most of these values.

    Tony Hansen

>
> Apologies if this reiterates previous discussions; I must admit that I
> seem to have skipped those debates when they happened.
>
> Harald
>
>