[anti-abuse-wg] Appointment & Removal of Working Group Chairs

On 9/29/2014 9:12 AM, Brian Nisbet wrote:
>>> This document aims to provide outline procedures to deal with the tenure
>>> of RIPE Working Group Chairs. It applies to all RIPE Working Groups.
>>
>>     This document outlines procedures dealing with the tenure...
> 
> I think I'm missing your point here?

Wordsmithing. This was merely an offer of more concise and direct
phrasing.

Self-assessment or self-prediction language in documents tends to add
only verbosity and distraction.  So for example, something like "aims to
provide" isn't needed.  Either the document does what it intends, or it
needs fixing.  At the least, I prefer documents that are self-confident
that they actually do what they intend, in terms of defining things.
Self-doubt about eventual efficacy is always a different matter...


>>> ** WG Chair Term:
>>>
>>> The normal term for a WG Chair is three years, after which the WG Chair
>>> must resign. A WG Chair may stand for re-selection after resignation. A
>>> WG Chair may resign voluntarily at any time.
>>
>> As phrased, this means that there is a gap between resignation and
>> filling the spot with the replacement chair.  And if the resignation is
>> 'forced' it isn't really a resignation; it's just the end of their term.
>>
>> Perhaps:
>>
>>     The term of a WG Chair is three years.  A current chair may stand for
>> re-selection at the end of their term.  A WG Chair may resign
>> voluntarily at any time.
> 
> The "gap" should be no longer than a RIPE meeting and isn't really a big
> thing. The wording change, yup, I've no problem with that.

My point is that I'm pretty sure there is no gap intended, or at least
there shouldn't be, for normally cycling of occupancy.  I'd expect the
group to want to have the chair seat always occupied.  (That is, for
cases other than mid-term resignation.)

The normal model for normal cycles is that a sitting chair holds the
position until the moment the next cycle of the chair puts the next
occupant into it.  (The fact that the current and next might be the same
person doesn't affect this bit of formal modeling.)


>>> ** Selection of a WG Chair:
>>>
>>> WG Chair vacancies, along with a call for candidates, should be
>>
>>     should -> must
>>
>> Seems essential to make the minimum announcement time mandatory.
> 
> Ah, yes, this isn't an RFC, so language thing, perfectly happy with must.

Hadn't meant to invoke RFC formalities as much as simple,
natural-language vocabulary semantics.  'Should' tends to be advisory
and the sentence here seems to call for an absolute requirement.


>>> be opened so long as the addition of a new Chair would not cause the
>>> number of Chairs of that WG to exceed the maximum number specified in
>>> this document.
>>
>> I didn't see where that maximum was specified.
> 
> There's a reason I left out this line:
> 
> Reference Documents: Working Group Chair Job Description and Procedures
> (RIPE-542)
> 
> I've no idea what that reason was, mind, but I'm sure it exists. Anyway,
> RIPE-542 specifies a maximum of three.

My main concern was that the text promised "in this document".  Perhaps
the next text is a revision to 542?  Sorry if I missed that bit of
context.  In any event, since the max is indeed specify somewhere,
rather than have text making the promise, perhaps just include a
citation to the text (in 'this' document or the external one)?


>> This obviously allows a form of ballot-box stuffing, by permitting votes
>> from people who have not been involved in the wg up to that point.  I
>> have no idea how to prevent this, however.  In addition, it can be
>> argued that anyone garnering enough ire to motivate the ballot-stuffing
>> effort probably is causing serious problems.  The counter to that is
>> that they might be standing up to efforts to coerce the wg...
> 
> It does, potentially, and this has been a bone of contention in other
> discussions. I have no idea how to prevent it either. Possibly more to
> the point I have no evidence to suggest it would or would not happen.
> Bussing in people at €350 a pop to remain co-chair of a RIPE WG would
> seem... frankly insane. That's not to say it'll never happen, of course,
> but wow.
> 
> There is, btw, the notion of popping in a consensus stage before the
> voting stage in a no-confidence motion. It doesn't remove the vote
> stuffing concern, of course, but it brings the whole thing more in line
> with preferred RIPE procedure.

For organizations trying to be dominated by overall community goodwill
rather that by artificial and strict procedural formalities -- and I
think ripe still falls into the former category -- I'm inclined to
suggest some sort of community 'smell' text.  That is, having a safety
mechanism which calls for the community at large to step back and decide
whether it thinks the overall tone of an activity feels acceptable.

However I've never seen this idea applied as a formal construct, and I
fear that the last time I saw it applied was as the ad hoc, spontaneous
effort after the IETF's Kobe blowup, 20+ years ago, that took authority
from the IAB and moved it to the IESG.  (The popular view is that this
was a successful exercise, and given the timespan from then to now, it
probably was...)

d/

-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net