On 9/30/2022 5:01 PM, Brian E Carpenter wrote:
On 01-Oct-22 07:10, John C Klensin wrote:
--On Friday, September 30, 2022 13:26 -0400 Michael StJohns
<mstjohns@xxxxxxxxxxx> wrote:
...
(Side comment - why can
the various boards have various levels of executive session,
but this wasn't contemplated for the Nomcom?)
All NomCom meetings are executive sessions, since they are
bound by the rule of strict confidentiality.
You're being disingenuous and pedantic, but perhaps I should have used
"closed session" instead of executive session even though both of those
sometimes mean the same thing. Consider for example the LLC Board. A
given meeting has an open portion, a closed session with LLC members,
Exec Dir and Staff only, a closed with LLC Members and Exec dir and
finally, a closed/executive session with only the LLC Board members.
Each of those levels of exclusion have a purpose.
A meeting that excluded the liaisons would be unacceptable,
IMHO, since one of the liaisons' duties is to witness the
process. A session from which the liaisons were excluded
would raise an intense suspicion of horse-trading.
At one point there were only two liaisions (the ISOC rarely appointed a
liaison), plus the chair and the past chair as non-voting attendees.
We're now up to a chair, past chair, 5 liaisons, and 3 advisors
(according to this years list). I'm sure that the chair and past chair
- keeping them as mandatory for a closed executive session - could be
trusted to manage the "intense suspicion of horse trading" (seriously -
not sure where you get this), especially if most of the sessions were
open. Or maybe add the ISOC rep (who has no dog in the fight) as a
third mandatory. Seriously, just how many people does it take to satisfy
"qui custodiet ipsos custodes"? Should we assign a commisar to each
voting member and be done with it? (Ok - now I'm getting hyperbolic,
but really - 10 non-voting participants is way too many relative to the
size of the voting membership).
For curiosity - how do you prevent two Nomcom voting participants from
having a beer and saying "I'll support your guy if you support mine"?
And how does having Liaisons in all the sessions prevent that?
I find the idea that *any* of the confidential material
should be concealed from the liaisons quite disturbing, for
the same reason. How can a liaison audit the process and
assert that it was followed correctly if they have not seen
all the material?
Again, just how many liaisons does it take to screw in an audit? I would
say that the Chair and Past Chair are in a good position to keep
everyone on the straight and narrow.
In any event, anecdotal evidence says that the Nomcom leaks. Having
fewer people with access to information might make the leakers think
twice as it might be easier to track a leak back to a source.
If a candidate is concerned that one of the liaisons might
be prejudiced against them or that there is a conflict of
interest (e.g. the candidate and the liaison have the same
employer), is this worse than the same situation between a
candidate and a voting NomCom member or the NomCom chair?
We rely on personal integrity for such cases, because
we have no choice. (RFC 8713 doesn't seem to mention recusal
for such cases.)
I think here you're comparing apples and oranges? The issue with the
voting members is usually not a conflict of interest as it is an
inability to act as an individual against their employer's desires and
in the best interest of the IETF and the community as a whole. Yes,
there is no mechanism for recusal, but, at least the last time I served
on a Nomcom, where the candidate and a member were from the same
organization or had other conflation of interests, the group tended to
bring that forward, discuss the possible bias, and use that as part of
the deliberation. E.g. we didn't a priori trust personal integrity, we
made sure the possible issues or biases got as much sunlight as possible
and required more than just "I like this person" as an argument for why
they should be selected.
Later, Mike
Brian
