21
$\begingroup$

Thanks for all your feedback on this format and design document. The 2010 Mathematics Community Moderator Election is now underway.

We're preparing the election page now. Here is how we think it will work, based on our existing moderator elections to date.

Our general criteria for moderators:

  • patient and fair
  • leads by example
  • shows respect for their fellow community members in their actions and words
  • open to some light but firm moderation to keep the community on track and resolve (hopefully) uncommon disputes and exceptions

All of the below will happen on the same "page" --

Phase I: Nominations

  • Any user in good standing with a minimum of 300 reputation may nominate themselves as a community moderator.
  • All nominations are by definition self-nominations. Nominating others is absolutely not supported.
  • Entering a nomination requires a brief description of why you wish to be a moderator. This can be edited as a wiki over time, but only by the nominee.
  • Nominees will appear in most recently nominated first order.
  • Nominees are allowed to withdraw, if they so choose, during this phase.
  • Commenting on the nominations is allowed and shown in this phase.
  • There is no voting in this phase, only commenting.
  • This phase will last ~7 days.

Result: Nominations are gathered and polished. Through commenting, nominees get a sense of what kind of community support they may (or may not) have in the primary phase, and what they want their description to say. Nominees may elect to voluntarily withdraw at any time, otherwise, they proceed on to the Primary phase.

Phase II: Primary

  • Any user in good standing with 150 reputation may vote in the primary.
  • Candidates will appear in random order.
  • One vote (either up or down) is allowed per voter per candidate.
  • Vote up/down counts and total vote scores are public.
  • Candidates are allowed to withdraw, if they so choose, during this phase.
  • Candidate descriptions may be edited.
  • No commenting is shown or allowed on the candidates in this phase.
  • This phase will last ~4 days.

Result: Initial up and down voting should provide a sense of which candidates are strongest and which may be controversial. Candidates may elect to voluntarily withdraw at any time, otherwise, the top 10 candidates based on primary vote score will proceed onward to the Election phase.

Phase III: Election

  • Any user in good standing with 150 reputation may vote in the election.
  • Candidates will appear in random order.
  • Three candidate votes, in order of preference, may be cast per voter.
  • Voting details are fully private until the election is over.
  • Candidate descriptions are locked in and cannot be edited.
  • Candidates may not withdraw.
  • This phase will last ~4 days.

Result: After the election is complete, the vote tallies will be available on the page for public download and analysis. We will use OpenSTV and Meek STV to calculate the 3 election winners.

Site-wide banners will be in place for 2 days for each phase change. For the final election vote, per-user messaging will be used to publicize the election and ensure that users -- particularly those users who have the requisite rep to vote -- are aware of them.

Some of this we are still hashing out, and it's all subject to change. Feedback on the live election is still welcome.

The first 2010 Mathematics Community Moderator Election is now underway.

$\endgroup$
5
  • 9
    $\begingroup$ "All nominations are by definition self-nominations. Nominating others is absolutely not supported." - why not? $\endgroup$ Nov 25, 2010 at 13:36
  • $\begingroup$ Are the values of n in all three phases supposed to be the same? $\endgroup$ Nov 25, 2010 at 13:37
  • 2
    $\begingroup$ If a nominee withdraws, can they reenter? $\endgroup$ Nov 25, 2010 at 16:04
  • $\begingroup$ @J.M. Somehow, I doubt it ;) $\endgroup$
    – badp
    Nov 25, 2010 at 16:32
  • 9
    $\begingroup$ @J.M. Probably so you don't nominate someone who doesn't want the job, without their knowlege. $\endgroup$
    – tzenes
    Nov 25, 2010 at 18:02

7 Answers 7

12
$\begingroup$

All done!

Here is the output of OpenSTV for the elections (thanks to badp at the m.SE chatroom).

Congrats to Robin, Qiaochu, and Willie!

"394 voters were eligible and 174 voted"

Wow, that's only 44.2%... :( Is this typical for SE elections?

$\endgroup$
1
6
$\begingroup$

How many moderators will be elected?

(This answer is community wiki; so someone can edit and add the answer in.).

$\endgroup$
2
  • 2
    $\begingroup$ How many moderators will be elected? — The latest news (2 weeks ago) I've heard is "3 to 5". $\endgroup$
    – kennytm
    Nov 25, 2010 at 18:12
  • 2
    $\begingroup$ On math.stackexchange.com/election the panel on the right says 3 moderator positions available $\endgroup$
    – WWright
    Dec 2, 2010 at 4:30
6
$\begingroup$

The election looks well designed, especially the nomination process.

One point not documented so far: does a submitted vote immediately become a cast vote (frozen and irrevocable)? If not, can users undo or edit their votes without limit until the end of the election? Once a vote has been frozen, can it be changed if the slate of candidates changes, e.g. due to withdrawals?

Also, can the votes on different candidates be made at different times?

$\endgroup$
4
  • 2
    $\begingroup$ votes can be undone in the standard 5 minute window after casting, only. After that they are locked in forever. $\endgroup$ Dec 2, 2010 at 6:24
  • 3
    $\begingroup$ @Jeff: thanks. Can a user vote at different times on different candidates (e.g., on different days) or does voting mean a single opportunity to submit a set of votes, one per candidate, as in a voting booth or an online checklist? $\endgroup$
    – T..
    Dec 2, 2010 at 6:29
  • $\begingroup$ . the primary voting works pretty much exactly like voting on answers in this question does. The election voting is, cast 1st/2nd/3rd choice with a 5 minute undo window once clicked. If you don't click you can wait forever to cast each vote. $\endgroup$ Dec 3, 2010 at 1:20
  • $\begingroup$ @Jeff: to disambiguate: do you mean that the primary voting on N candidates is N independent clicks, performable at different times? The final election has to be a one-click deal because it's a ranking. For the primary, the question is relevant mainly if the running vote totals are updated and visible throughout the process. If only the final totals are disclosed then voters will tend to wait until the end to decide, but with "live" disclosure and independent voting on different candidates, they will vote on some candidates earlier as a signal to other users. $\endgroup$
    – T..
    Dec 3, 2010 at 1:40
6
$\begingroup$

I believe elections shouldn't be "gameable" in that your behavior or knowledge of how the election works shouldn't make your "vote" count more. But I'm worried that if you allow people to both vote UP for someone and vote DOWN for other people, that the most logical thing would be to vote UP the people you like and DOWN all the other people... if you don't do that, you would essentially be casting a less strong UP vote. I'm not sure all people voting would know to do that and so I wonder if allowing DOWN votes is desirable.

Good thing this is on the math.se so people can figure out if I'm mistaken...

UPDATE: We will only allow downvotes during the Primary phase, not the Election phase, so my concern is essentially dealt with.

$\endgroup$
5
  • $\begingroup$ This is a good point. Allowing both up and down votes is equivalent to having people vote on a {0, 1, 2} scale, in a way that might not be obvious to a casual user (I probably wouldn't have thought of it if you hadn't pointed it out). Perhaps just allowing up votes would be better; that would make it a variant of approval voting. $\endgroup$
    – user856
    Dec 1, 2010 at 16:22
  • $\begingroup$ The UPvote/novote/DOWNvote is equivalent to an ordinary vote system with half-votes, where each voter gives 0, 1, or 0.5 votes to each candidate. With some fractional votes now allowed, it would be more informative (and may be more intuitive) to allow all, and have a continuous voting scale from 0 to 100 percent support, or -100 to +100. "Monetary" voting, where each voter has a budget of one vote-unit to divide among the candidates is also intuitive and expressive. Generally, the more information (in the sense of "information theory") allowed per user the more revealing the election. $\endgroup$
    – T..
    Dec 1, 2010 at 16:32
  • 6
    $\begingroup$ When we conducted moderator elections on MathOverflow almost nobody used downvotes to vote for all of the other people they didn't like. My recollection is that they were limited to candidates that users felt very strongly negatively about. $\endgroup$ Dec 1, 2010 at 16:55
  • 1
    $\begingroup$ This is only relevant in the primary phase, where we are trying to establish, from a field of potentially dozens, who is actually electable and should continue onward to the election. Not relevant to the actual election phase. $\endgroup$ Dec 3, 2010 at 1:27
  • $\begingroup$ It is exactly the electability determinations that would benefit from finer-grained voting in the primaries. The voter and candidate populations are too small now for any of this to be worth changing for one election, but in the future it could make a positive difference. $\endgroup$
    – T..
    Dec 3, 2010 at 1:48
5
$\begingroup$

Question: Will the elected moderators replace the current (pro tem) moderators, or serve in addition to them?

Answer: Elected moderators will replace the current (pro tem) moderators, however current moderators are welcome to run in the elections.

$\endgroup$
2
  • 3
    $\begingroup$ The former. [ ](http://.) $\endgroup$
    – kennytm
    Nov 26, 2010 at 20:15
  • 8
    $\begingroup$ whoever wants to run as a moderator, including pro-tem moderators, should nominate themselves $\endgroup$ Nov 26, 2010 at 20:46
5
$\begingroup$

It would help both the users and the moderators-to-be if StackExchange management would specify a calendar date by which the next election will be held.

This would give users the confidence of having moderator replacement in finite time, and moderators would get an upper limit on the duration of the commitment they are taking on (I guess they can resign at any point if it's too much, but it is better to know in advance). It is also ordinarily a part of any recurring election to have a known date or deadline for the following election.

This is purely a request, not a criticism. The handling of this election has been exemplary --if anything, it seems unusually careful and well-considered given the fallout from the pro tempore pseudo-elections during the beta -- and a first-of-its-kind election on a growing and evolving site is not expected to have all details solved in advance.

$\endgroup$
3
$\begingroup$

Question: According to the faq, anyone with over 10,000 rep has access to moderation tools. What, if any, is the difference between being a moderator and the privileges that come with having that much rep?

Answer, from the link provided by Jeff Atwood (thanks!):

"Moderators do have some special abilities; these abilities are necessary to handle those rare exceptional conditions that can’t effectively be dealt with by the community.

  1. Moderator votes are binding. Any place we have voting (close, open, delete, undelete, offensive, etc), that vote will reach the threshold and take effect immediately if a single moderator casts a vote.
  2. Moderators can lock posts. Locked posts cannot be voted on or changed in any way.
  3. Moderators can see all data in the system, including votes and user profile information.
  4. Moderators can place users in timed suspension, and delete users if necessary.
  5. Moderators can perform large-scale maintenance actions like merge questions and batch retag."
$\endgroup$
3
  • 3
    $\begingroup$ see blog.stackoverflow.com/2009/05/a-theory-of-moderation specifcally the section titled "moderators do have some special abilities.." $\endgroup$ Dec 2, 2010 at 5:06
  • 7
    $\begingroup$ @Jeff Atwood: On said blog you wrote: Moderators can see all data in the system, including votes and user profile information. Does this mean that mods have access to tools that reveal the author of every vote? $\endgroup$ Dec 2, 2010 at 5:44
  • 5
    $\begingroup$ @bill no, they only have access to vote statistics, eg, who has voted for whom in aggregate, largely to detect voting anomalies $\endgroup$ Dec 2, 2010 at 8:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .