Re: SUMMARY: should there be term limits?

[Marshall Rose <mrose+internet.ietf.nomcom@dbc.mtview.ca.us>]

    i apologize for the length of this note.
    
    a few folks seem to be suggesting lots of changes to a system that
    seems to work pretty well.
    
    the fact that the current system doesn't please them, doesn't bother
    me one bit.
    
    in fact, i think that the current system doesn't entirely please
    anyone, but that the vast majority of us can live with it. that
    tells me it's close to optimal.
    
    the purpose of this lengthy note is two-fold: i'm going to try to
    movtivate folks not to change a damn thing; then, i'm going to ask
    for a change.
    
    at the end, if the decision is that it's better not to screw around
    with things, i'm happy to withdraw my proposal.
    
    i urge you to read the following carefully. while it's not as clever
    as i like, i hope you will find it convincing...    
    
    many thanks,
    
    /mtr
    
				  #######
    
the current system
    
    the existing system isn't perfect, but i claim that it more or less
    works just fine. further, i'll claim that many "improvements"
    suggested thus far either fix non-problems or cause worse problems
    than the ones they solve.
    
    the thing to understand about the existing system is that there are
    lots of checks-and-balances to promote things like continuity,
    evolution, stability, innovation, candor, and privacy. although in
    conflict, all of these serve the good of the community.
    
    in terms of continuity v. evolution: only half of the positions are
    up for consideration each year. hence, presumptive arguments for
    incumbants are, on their face, irrelevant. the existing system gives
    the nomcom considerable flexibility to put someone new in, simply
    because there's still someone experienced in the role.
    
    in other words, the downside of change is pretty much minimized.
    more important though, it gives the community the ability to grow
    mid-level people into senior people. in an earlier note, ted
    argued against term limits because people like fred and harold are
    far and few between. what was left unsaid is that good people don't
    just appear at random, instead they have to be "groomed" by being
    given the opportunity to do things. the existing system does just
    that. by bringing in new people, you pair them with a more senior
    person, and this gives them the opportunity to grow professionally.
    
    in terms of stablity v. innovation: the nomcom doesn't operate in a
    vacuum, there are four ex-officio participants who can provide
    considerable guidance. so, even if every voting member of the nomcom
    is absolutely clueless, the ex-officio's should be able to moderate
    their decision processes. further, if any of the ex-officio's
    generate too much noise, then the others should be able to deal with
    that.
    
    all systems, including the current one, are going to get complaints
    about being gamed, spun, or perverted. it's all part of the mix of
    having competing interests. what the current systems does is provide
    numerous feedback loops to minimize this behavior. besides the four
    ex-officio's, you also have the confirming body to provide push
    back. can the voting members ignore all this? yes. will they? it
    depends on how strongly they feel about the choices that are
    available to them. the point here is that there are multiple stages
    of review and comment along the way.

    some have argued that when the confirming body rejects a candidate,
    then this should be treated as the final say on the matter. this
    perspective isn't consistent with the philosophy of checks and
    balances. if the confirming body rejects someone, the nomcom should
    be free to re-think their choice, and, if they feel strongly about
    it, then to submit the name again. if this leads to a deadlock, then
    the position doesn't get filled -- and frankly, i think that's just
    fine. if both sides feel so strongly on the issue that they can't
    agree, then the position should go unfilled until they can
    agree. and if that means the position goes unfilled for a year, i
    applaud them both for acting on the strength of their
    convictions. (of course, we all know that it's unlikely to go on for
    a year. maybe a month or two. and if it is an entire year, that's
    okay, the iesg and iab have operated at less than full strength
    before.)
    
    i claim that to give the confirming body an absolute no, effectively
    allows the confirming body to ignore the nomcom entirely. the
    confirming body has veto power right now, it does have, nor should
    it have droit de seigneur.
    
    in terms of candor v. privacy: i'll be honest, the original model i
    had for nomcom confidentiality was that an outsider shouldn't even
    be able to tell if an existing officer put their name back in the
    hat for re-consideration. given the widespread reports of rumors,
    counter-rumors, misinformation, disinformation, etc., from the
    current cycle, i was hopelessly naive and am now deeply
    saddened. perhaps i expect too much from people.
    
    however, i think that candor is important because if someone is
    doing a lousy job, i want someone to be able to tell this to the
    nomcom, without worrying that this will get out'd and, a week later,
    they might end up with that very special someone as their new
    boss. further, because you have the four ex-officio's there, any
    comment that is seriously twisted will get debunked fairly
    easily. again, more feedback loops to stop nonsense.
    
    to sum up: we have managed to survive 9 generations of nomcoms
    without having to hire ted olsen or david boies. this is probably a
    good thing.
    
    
a proposal
    
    one of the thing that surprises me about the way the nomcom works 
    is how little turnover we get on the iesg. i'm not saying this as a
    criticism of the incumbancy, but rather as a concern that if the
    iesg gets run over a by a bus, we're in a lot of trouble. the whole
    point of putting half the positions up for consideration each year
    is to be able to rotate mid-level folks in to give them an
    opportunity to become senior folks.
    
    of course, there are a lot of reasons why it may be necessary to
    select the incumbant, fine. however, i suspect that a lack of
    history between each generation of the nomcom is causing this
    behavior. (of course, replacing the nomcom each year is a good
    thing; it's just another one of those pesky trade-offs...)
    
    so, here's my proposal: 
    
    1. a member of the iesg or iab must not serve for more than six
       years without taking a break; and,
    
    2. a break must last at least two years before that person can be
       considered for the same body.
    
    i call this the "fred baker precedent", because fred served for six
    years as the chair of the ietf and then went to the iab. (maybe he
    took a break in between, i'm not sure.)
    
    so, what this means is that once you get put on the iesg, you can be
    there for at most six years. after that, you can be selected to go
    onto the iab, but you can't come back to the iesg for at least two
    years.
    
    so, if someone is a superstar and their six years is up, and if the
    nomcom feels that person should still be providing service to the
    community, they can select them for the other body.
    
    i suggest that the at-most-6/at-least-2 rule counteracts the lack of
    history between generations of the nomcom, whilst not penalizing
    people who do good things.
    
    your comments are solicited.
    
/mtr

				  #######