Re: TAG election reform

Hi David,

On Apr 17, 2012, at 09:43 , L. David Baron wrote:
> On Monday 2012-04-16 17:15 +0200, Robin Berjon wrote:
>> The three suggested changes are:
>> 
>>    • Make the list of candidates public so that they can be publicly discussed (this not being the case prevents some communities from following their normal decision process, e.g. Mozilla). This should be trivial to implement (a simple W3C news items would be enough). I don't think that this requires a change to the Process, it's probably just a W3M decision to make.
> 
> To be clear, I don't think this is the only reason Mozilla would
> want to use a different process; I think it's probably unwise to
> discuss the pluses and minuses of people in public the same way we'd
> discuss charters.  That said, we could certainly discuss the TAG
> election more openly than we do now (e.g., by soliciting feedback).

Sorry if I wasn't clear; I didn't mean to imply that this was the only reason why Mozilla was not comfortable discussing its vote in public, nor that it's a problem only Mozilla has (others have reported it as well).

I actually took the feedback you'd given me then about discussing the plusses and minuses of people in public (which I agree is problematic) and it shaped the part of my proposal that establishes a location for candidates to convey their platform (as it were) and be interrogated by the community. It's not a perfect solution, but it makes it possible to move on from "I really don't like that Robin guy" to "I really sass that froody Robin guy, but he's clearly completely wrong on the privacy implications for the semantic range of polyglot XForms as representations of 409 errors — that just makes him wrong for TAG in my book".

>>    • Use preferential voting so as to produce a fairer, better-informed vote (the current system encourages voting for only one candidate if you strongly favour her). In order to avoid an endless debate as to which voting system is best, I suggest simply using the preferential voting system that WBS already supports. Implementation-wise, I believe that this is a simple matter of picking a different WBS question type. This probably requires a small change to Process since it describes the vote. That said, if there's a way to weasel it past without changing the Process I for one won't complain :)
> 
> There are many different ways one could process a list of ranked
> preferences.  Which are you proposing?
> 
> (I'm actually generally a fan of approval voting, which would mean
> keeping the current system, except eliminating the rule that one
> can't vote for more candidates than number of open positions.
> Though in this case I'd be less opposed to a system involving ranked
> preference than I normally would be since I suspect there's less
> likely to be well-informed strategic voting (e.g., based on polls).)

As indicated, I was hoping to skip over that debate by using "whatever WBS already does" :) I have no idea how it makes its decision, but it does support ordering a number of responses to a question. Maybe Yves can find out more? The point here being mostly that I'd rather not a) have a long debate on an unsolved problem and b) make more work for the systems team — unless enough people really care about it.

-- 
Robin Berjon - http://berjon.com/ - @robinberjon

Received on Tuesday, 17 April 2012 16:06:16 UTC