On Thu, Jul 17, 2008 at 10:09 AM, Neal McBurnett <[EMAIL PROTECTED]> wrote:
> On Thu, Jul 17, 2008 at 08:20:36AM -0300, Cody A.W. Somerville wrote: > > * Normal elections will start at an agreed date relative to a > development > > milestone and polls will remain open until a second agreed date that is > also > > relative to a development milestone. Once polls close, results will > become > > available and a short period to allow for grievances and/or disputes to > occur > > takes place. Finally, a third agreed date, which will also be relative to > a > > development milestone, will mark the normal conclusion with the MOTU > council > > officially announcing results and updating team memberships. If a > grievance of > > dispute arises, the MOTU council will resolve the issue in 15 days of the > third > > date or escalate the issue to the technical board. > > > > Although a voting system has not been agreed upon yet, two systems which > have > > received a lot of discussion include Single Transferable Vote (used by > some > > governments) and the Schulze method (used by Debian and Wikipedia among > > others). It is generally agreed that a preferential voting system, where > voters > > would rank their preference of the candidates instead of voting for or > against, > > is best. > > I've been active in voting method discussions for a few decades now. > If we go with elections, my advice is to go with either Approval > Voting or Range Voting. One benefit of them is the focus on > "supporting" folks (rather than a competition among folks), which can > contribute to a feeling of consensus. > > http://en.wikipedia.org/wiki/Approval_voting > http://bcn.boulder.co.us/government/approvalvote/center.html > http://en.wikipedia.org/wiki/Range_voting > > I don't know if I agree with that. The reason why I feel a preferential system, in particular STV, is a good fit for us is because you never vote against someone like you would in approval voting (by not voting) - you simply express your preference of one candidate over another. I feel this is mandatory because it should be a requirement to becoming a MOTU that you *could* be able to fill these roles. As for range voting, that is a plausible idea but both range voting and approval voting are classed as single seat (ie. used to elect single individual) system whereas STV is a multi-member system. > I'm not sure why STV is on the list since it is for party voting - do > we have parties in MOTU now? I mean besides the great festive > gatherings that Daniel organizes? ;-) Single Transferable Vote is *not* for party voting. Infact, it *explicitly* ensures votes are for candidates and not parties. STV is on the list because is a preferential voting system designed to minimize wasted votes and provide proportional representation. > > > The ranked methods like IRV (and STV) and Schulze suffer from > increased complexity and confusion, and the risk of counter-intuitive > results. E.g. with IRV it is possible that raising the rank of a > winning candidate on some ballots, which originally had ranked that > candidate last, could counter-intuitively result in the winning > candidate becoming a loser. See > > http://en.wikipedia.org/wiki/Instant-runoff_voting > http://en.wikipedia.org/wiki/Schulze_method > http://en.wikipedia.org/wiki/Single_transferable_vote Yes, this is called the monotonicity criterion. IRV and STV both fail this criterion but Schulze method does not. However, I don't think we need to get caught up in weird corner cases due to the scale and nature of the elections we'll be holding. > > > > An alternative proposal by Emmet Hickory would have members be attached > to a > > "release" and favors replacing team members through a process more > > closely related to apprenticeship than any sort of election. The team > would > > define goals for a release, handle freeze exceptions for the release, and > then > > follow the release as the SRU team until the release is no longer > supported > > (all together, roughly terms of 2 years). His full e-mail can be found > here: > > https://lists.ubuntu.com/archives/ubuntu-motu/2008-July/004169. > > Make that > https://lists.ubuntu.com/archives/ubuntu-motu/2008-July/004169.html > > Neal McBurnett http://mcburnett.org/neal/ > > -- > Ubuntu-motu mailing list > [email protected] > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu > -- Cody A.W. Somerville Software Engineer Red Cow Marketing & Technologies, Inc. Office: 506-458-1290 Toll Free: 1-877-733-2699 Fax: 506-453-9112 Cell: 506-449-5899 Email: [EMAIL PROTECTED] http://www.redcow.ca
-- Ubuntu-motu mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu
