On 12/15/2012 03:39 PM, Greg Freemyer wrote:

Michael Catanzaro wrote:

...

On Sat, 2012-12-15 at 02:36 +0100, Pascal Bleser wrote:

...

Could you explain what you mean with a "runoff" ?

cheers

In an election to fill one position, when no candidate gets 50% of the votes, a runoff election may be held with just the top two finishers. E.g. say 40% of the people like Candidate A, 32% like Candidate B, and 28% like Candidate C. But B and C are really similar and B/C supporters don't like A. If there's no runoff, then A gets 40% of the vote and wins, even though 60% of the electorate would prefer B or C. (That's what happens in almost every election in the USA.)

Every election maybe, but not every race. So where I live if we have 15 races to vote on, typically one or more of the races won't get 50% and there is a run-off election a few weeks later.

For openSUSE we can make up any rules we want.

eg. A winner must have 40% of voters minimum in the first vote to be declared a winner.

If only one candidate achieves 40%, then the candidates with the 2nd and 3rd most votes will be in a runoff.

If no candidate achieves 40%, then the top 3 candidates go to a runoff.

fyi: i have no idea what good cutoff would be. Where I live it is 50% for political elections, but it can be whatever works best for openSUSE.

Hmm, sounds very complicated. The openSUSE rules are pretty straight forward, https://en.opensuse.org/openSUSE:Board_election_rules#Voting: . Those with the most votes (by count) win, just a straight tally of the numbers, no calculation of percentages, run off votes, multiple rounds etc. I think this has worked pretty well for us so far. I am not certain why we would need to expand the procedure. AFAIK, to this point we have not had a situation where 2 or more candidates had the same vote count. I suppose that scenario is possible, and then there could be a second round of voting between tied candidates for 1 seat. The second round could then also just follow the simple vote count rule we already have. Anyway, this is rather theoretical and we probably do not have to worry about it until 2 or more candidates actually do have the exact same vote count. My $0.02, Robert -- Robert Schweikert MAY THE SOURCE BE WITH YOU SUSE-IBM Software Integration Center LINUX Tech Lead rjschwei@suse.com rschweik@ca.ibm.com 781-464-8147 -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On 2012-12-16 08:48:35 (-0500), Robert Schweikert wrote:

On 12/15/2012 03:39 PM, Greg Freemyer wrote: [...] Hmm, sounds very complicated. The openSUSE rules are pretty straight forward, https://en.opensuse.org/openSUSE:Board_election_rules#Voting: . Those with the most votes (by count) win, just a straight tally of the numbers, no calculation of percentages, run off votes, multiple rounds etc. I think this has worked pretty well for us so far.

Agreed. We do have a lot of candidates this time though, more than for past elections, afaicr.

I am not certain why we would need to expand the procedure. AFAIK, to this point we have not had a situation where 2 or more candidates had the same vote count. I suppose that scenario is possible, and then there could be a second round of voting between tied candidates for 1 seat. The second round could then also just follow the simple vote count rule we already have. Anyway, this is rather theoretical and we probably do not have to worry about it until 2 or more candidates actually do have the exact same vote count.

I think so too. And anyway, changing the rules of the board election is something 1) we shouldn't do while the election is on-going ;) 2) the board should decide before the next elections (possibly with a vote by members to ack/nack, although that's implicit through the board election process, I'd prefer to not add more process to it ...) If such a scenario arises, the election committee should discuss with the current board to see what to do. A "cutoff" (or "second round" as it's known e.g. in France) is an option indeed, if we find a simple and straightforward variant :) But unless that happens, and it's really unlikely to happen, one can win with 0.001% of votes, and IMHO that's still simple enough, and works. cheers -- -o) Pascal Bleser /\\ http://opensuse.org -- we haz green _\_v http://fosdem.org -- we haz conf

Le lundi 17 décembre 2012, à 11:38 +0530, Sankar P a écrit :

IIRC the GNOME election process also uses a calculation like this. May be someone like Vincent Untz may get more details (I am not following all the mails in the thread so sorry if I have missed an earlier mail on the same). I also believe that this is a much better electoral system.

The GNOME Foundation now uses one of the STV variants (Scottish STV, according to the result page). See https://en.wikipedia.org/wiki/Single_transferable_vote for discussion on STV. There are pros and cons, obviously. Cheers, Vincent -- Les gens heureux ne sont pas pressés. -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org