54

Last spring, I ran an election on Parenting for a new moderator pro tempore. It went pretty well, so this fall we are going to extend the program to sites that have been in "beta" for years. Here's how it works:

  1. We're reaching out to moderator teams to see if they would be interested in participating. Priority will go to sites with the longest history or that we've identified as having an urgent need. The ultimate goal is to have elections for all sites.

  2. Once I've determined how many moderators the election should add, I'll schedule the election and post an announcement on the per-site meta.

  3. For these elections, we won't be creating a candidate questionnaire. Instead, we encourage voters to ask their questions on the site's meta use the and tags. Moderator candidates should monitor meta for these questions.

  4. Current moderators will not be required to nominate themselves—they will continue as moderators unless they ask to step down. (NB: Existing moderators currently get an email saying they must nominate themselves, but this is because the automated system does not yet know about pro tempore elections. Please ignore the instruction to nominate yourself if you get this email.)

  5. Unlike most elections, we won't require pro tempore elections to be competitive. If the number of nominees is less than or equal to the number of open slots, we'll end the election before the voting stage. (It should be very rare, but the community team retains the right to remove candidates at any time. In particular, we disallow nominations from people who are recently suspended.)

  6. When the election ends, I'll appoint the new moderator(s) as pro tempore moderators. (The distinction matters mostly for the Constable and Sheriff badges.)

The long-term goal here is to increase democratic participation in all of our sites, break down the barriers between beta and graduated sites and eliminate one community manager task. Over the years, we've appointed many moderators and while we seem to have a good track record, that probably has more to do with the population of users interested in the position more than our skill. Therefore, there's no reason for us to maintain control over who becomes moderator on beta sites.


After several successful pro tempore (and one that's not yet succeeded) elections, I'm extending the program to Tezos, a newly launched public beta site. We'll see if this is a bridge too far.

  • 6
    Just to be clear here, if the site does graduate, moderators elected as part of a pro-tem election will have to be re-elected to retain their moderator privileges, right? – Sonic the Inclusive Hedgehog Aug 21 '18 at 17:39
  • 4
    @SonictheInclusiveHedgehog: As things stand, yes. It's possible we'll change the way graduation works in the future. But in any case, elected pro tems will be identical to appointed pro tems in this respect. – Jon Ericson Aug 21 '18 at 17:48
  • 3
    I heard that current mods on beta sites are also considered "pro-tempore" before the site graduates. Does this new way considers those mods as the "current mods", and thus won't require them to be re-elected and so? – DarkCygnus Aug 21 '18 at 18:01
  • 4
    @DarkCygnus: Yes. On graduation, all moderators must stand for re-election if they wish to continue in the role. There will be no difference between moderators who ran for an election during beta and moderators we appoint. – Jon Ericson Aug 21 '18 at 18:04
  • So the long term goal, once this has trickled to all sites, is that a new site will have an election pretty soon after emerging from private beta? – Chenmunka Aug 22 '18 at 8:18
  • How is the number of new moderators calculated? Is it based on current moderator load? current moderator activity? site size? black magic? Mostly just curious to see what the trigger is. – hazzey Aug 22 '18 at 14:41
  • 1
    @Chenmunka: That's not determined yet. The main argument against is that starting a new site sometimes requires a different skillset than moderating an established site. It's also useful to give people time to get to know each other before running an election. Still, the original vision was we would have elections ~90 after the start of public beta and I think this is a step toward making that sensible. – Jon Ericson Aug 22 '18 at 16:24
  • 3
    @hazzey: I interpret the pattern of the coffee grounds in my mug and resent the implication it's black magic. ;-) We like to have at least three mods and add more when the existing team feels they need another hand. I look at flag handling statistics as a proxy for moderator workload, but that can be no better than the coffee grounds method. Some sites have more difficult flags than others and quite a bit of the work can be on meta rather than in the flag queue. Being able to run elections (and not worrying about them being competitive) should make it easier to add mods when the team needs it. – Jon Ericson Aug 22 '18 at 16:30
  • @JonEricson: That 'main argument' was exactly my concern. 90-odd days is a sensible settling period. – Chenmunka Aug 22 '18 at 17:22
  • Hi Jon, re this meta answer on BH.SE, is this experiment still ongoing and are you able to say/guess if there are likely to be elections on BH.SE before graduation? – Jack Douglas Jan 17 at 20:56
  • @JackDouglas: Yes we are running this experiment and it's entirely possible BH will get an election if y'all need a new moderator. We're talking about changing graduation (yet again) so that sites get various customizations as needed. So it may very well be that long-running beta sites will get a custom color scheme and lose the "beta" label without the other aspects of graduation. See also: meta.stackexchange.com/a/307988/1438 – Jon Ericson Jan 18 at 21:33
14

Elections as a catalyst for discussion

Thank you! I've been waiting for an announcement like this for a while, and the big thing I'm really excited about is the potential to kindle interest in community moderation as a whole amongst the userbase. Elections have a tendency to bring the community together and focus on common goals. There's collective self-reflection about some of the big questions: What are the biggest problems on our site? Are the policies we've implemented successful? Where do we go from here?

As a long-term beta moderator - does three and a half years count? - these are questions that I haven't seen people talking about with gusto for some time, on some of my sites. Now, part of that is that we haven't had many problems. The road's been a quiet one, but smooth enough. I'd like to see some more proactivity - talking about even minor issues that we can address to polish the sites a little more. Elections often bring old and new faces alike out of the woodwork, which could mean a more diverse set of opinions.

Keeping this in mind, in lieu of a questionnaire, I'd be interested in maybe having candidates just write up a paragraph or two addressing the top two or three problems they want to address, just to imbue the whole thing with a sense of purpose. Essentially, they'd have to answer the question "What are [sitename]'s biggest problems?" briefly. No need to dig into a candidate's ideas on every issue; there just might not be enough substance there. I also doubt we could drum up support for a whole ten-question questionnaire, even if we wanted to. But this is smaller, and it can probably be implemented casually, on a per-site basis; I do intend to try it out and see what happens, if HSM or Mythology wind up having elections (we'll see).

So . . . in a sense, I look forward to the elections not just as a way to get new people involved, but also drum up new ideas and maybe kindle some discussions. This could flop, of course, but I'm optimistic. We'll see what happens.

What to look for in a candidate

I don't think I plan to do this myself, but I'm actually kinda interested to see what would happen if an existing mod decided to nominate themselves - to continue on if elected, and to step down if not. Part of my feels like it would either go really well or really poorly - really well because folks are (hopefully) happy with the status quo, or really poorly because normal things we look at during elections - meta posts, review queue stats, etc. - are going to be virtually nil. We get meta posts on HSM less often than a local monthly book club sends out a newsletter.

In a sense, this exposes something I'm worried about: candidates in general will have little to point at on some sites. When community moderation tasks are needed, they get done, but there aren't many to go around. I'm curious to see what characteristics will swing voters which way. The possible downside? Rep could swing voters' minds even more than it normally does. And . . . I worry about that. Not that my old betas have any high-rep users who aren't good people, but rep isn't really correlated with moderator ability, unless it's indicative of an old hand with plenty of experience with the scope and community moderation tools.

Again, this is something I guess we'll get a better grasp on going forward, and I'm not inclined to worry yet. I'll be interested to see if we can get some data or anecdotal evidence on this when the first wave of elections starts, but it'll be in the back of my mind.

7

I think that's a good idea. There's no reason why moderator should be selected differently based on the number of questions per day.

Given that the initial moderators will still be nominated, when would the first election takes place? If I understand correctly, either when a nominated moderator steps down, or when the site needs more moderators?

Current moderators will not be required to nominate themselves—they will continue as moderators unless they ask to step down.

Why? If the site is old enough to sustain elections (meaning that candidates have had a few months to get used to the site and to demonstrate their behavior), is it still useful to retain the initial nominated moderators?

For these elections, we won't be creating a candidate questionnaire. Instead, we encourage voters to ask their questions on the site's meta use the discussion and election tags. Moderator candidates should monitor meta for these questions.

Why? I don't see any reason to treat these elections differently.

I do have reservations about the current format of election questions. The selection is determined by voting, which is largely determined by how soon the question was posted to the question collection thread. But if anything this is less of a problem on younger sites, where there would rarely be 10 questions anyway.

  • 11
    The main reason for not including the questionnaire steps is that it adds manual work and we need to either automate or eliminate work if we are to have more elections. The secondary reason is I'm not sure how much value the questions add to the process. Part of the reason I structured the experiment this way is to see what sort of election activity we see on meta. We're still hashing out our plans for regular elections and fixing our long-running-beta oddity. It may be that moderators will stand for reelection every so often. For the moment, this plan causes the fewest changes on sites. – Jon Ericson Aug 22 '18 at 22:31
  • 2
    Forcing the currently appointed pro tempore mods to stand for election amounts to a recall election in spirit. No other mods need to stand for elections once they are seated. Not having to be concerned with "popular" opinion in the future allows the mods to concentrate their energies on their work, and to make the occasional unpopular, but necessary decisions. Removing that sanity net from pro tempore mod would be less than fair to both the mods and the site entrusted to their care. – Gypsy Spellweaver Oct 9 '18 at 1:48

You must log in to answer this question.

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