Help me opine about a new election system

Question

I have been reflecting on some ideas around our current election system which were (originally) prompted by our pro tem appointment process. First, I think the Community Team needs to get out of the moderator-selection business. Currently, 45% of our moderators were appointed by someone from the Stack Exchange staff, and more than half (58%) our sites have yet to hold a proper election.

We need to give agency back to the communities regarding how you pick and maintain your elected leadership from the beginning and as the community continues to grow.

Is there an interest in broader election reform?

The purpose of this post is to gauge the general interest in taking a fresh look at our current electoral ideology and how you pick and maintain your moderation leadership. This isn't about working out specific implementation details or resolving every obstacle. I want to see if there is a general appetite for looking at broader reforms around how we elect and maintain moderators.

The first question to ask is: “Is this even a thing? Should I be looking at this?" Or am I stepping on customs so ingrained and sacrosanct, I should just leave it alone.

Why now?

Jon Ericson has been experimenting with pro tem elections to replace moderators who have moved on. As we were looking at improvements to that process, I kept asking if we should roll that [cool idea] back into the regular election cycle?

Starting with a clean slate, I started considering how broader electoral changes might improve some of the issues around our representative ecosystem. I think we might be edging into something more capable and adaptive — so I want to see if I should continue working on this.

Here are a few issues I’ll put on the table for consideration; I'm sure there are others. This is an open discussion and trial balloon to see whether there is any interest in looking at broader electoral changes. Please feel free to add your thoughts in the answers below.

• Appointing pro tems is icky
  There's no reason we (the community team) need to continue picking your moderators. We've already started experimenting with pro tem elections, so I'd like to improve on that process so ALL moderators are elected by the community — including pro tems — starting shortly after the private beta.
• Have one unified pro-tem/moderator process
  Jon's pro tem experiments show that pro tem elections can work, so I would like to see if we can merge both processes into one, unified election ecosystem. As the distinction between pro tem and "real moderator" starts to fade— read on.
• Moderator replacements take too long
  The CoGro Team (particularly Jnat and others) has been grinding away at a long-neglected backlog of sites needing moderator replacements (resignations, absentees, vetting candidates). Those delays will not likely end with the current backlog. The community team continues to be conspicuously short on resources to help communities with other issues. I believe this next bullet item can essentially free up those resources almost entirely.
• Simpler Elections, More Often
  Let's start with an assumption that I can simplify the election process to make it much less exhausting. The basic idea is to have much simpler, lightweight elections… more often… at regular intervals… maybe even yearly. The idea is to make elections much less epic and generational by creating more opportunity for the community to decide who they elect and retain. One way to help assure your best talent rises to the top is to provide more opportunity for avid users to apply.
• The community should run their own town hall
  This one seems like a no-brainer. Leveraging our current Q&A framework, we can model a town hall where communities are empowered to host their own event if they so choose, or skip it if they don't (with CM assistance where needed, of course).
• Should "Moderator For Life" always be automatic… and FOREVER?
  Our Moderator team is the best there is, bar none. Having senior moderators is important to retain valuable talent. But should communities have a say when that happens? And should they be empowered to decide if something has changed? We don't want to have a lot of needless rollover elections, so maybe we could come up with a system where long-term "senior moderators" are easily retained, while more established communities can avoid needless elections when it turns out: "Nah, we're good. We don't need another election."
• No change to the “single transferable vote”
  I love our STV voting system; I think the single transferable vote is inspired. I’m not sure if there’s room for improvement, so I mention it here but I would largely keep the fundamental mechanism intact.

Answer 1

The first Skeptics election had 64 voters in total.

That's a scary number to me, any lower and I'm not sure the election would mean anything.

The previous stance by SE was always that the community needs a certain size and activity to host elections, and I don't see any reason why this would have changed. Coupling elections to graduation doesn't make any sense anymore with the way they work now, and the endless beta many sites are in, but that is something SE already knows and has worked on. I don't see any reason to change much more than the decoupling of elections from the graduation state, as graduation is now coupled to criteria many sites won't hit soon, or maybe never.

I'm personally not a fan of the town hall questions for elections. They don't provide substantial insight about a candidate to me, compared to the rather large effort they are for the community and each candidate. The most reliable indicator for a mod for me is still previous activity on the site. So I would personally consider changes there potentially good, but it's also not something I would care a lot about.

Personally, I don't have any issue with moderators being appointed by SE. Of course I've been appointed twice by SE, and later elected in both cases. I don't see any alternative to appointments in early sites, there is simply no way actual elections would work at that time. SE isn't a democracy, and I think it's a bit misleading if we style SE sites as democratic. SE still holds all the power, and while they usually don't exercise it, the community only runs the site within the constraints set by SE.

Answer 2

Should "Moderator For Life" always be automatic… and FOREVER?

This would be a fundamental change in the relationship between the moderators (particularly the elected ones) and the company and, more importantly, the community.

I'm not sure I would have run for election - especially on Stack Overflow - if I was always having to think about re-election. I know in other walks of life elections aren't for life, but then we get the constant electioneering which distracts people from doing the job they were elected to do.

I know that Stack Exchange takes breaches of the rules around moderating very seriously and diamonds have been removed from several moderators for various reasons. Perhaps what we need is a simpler (but not necessarily quicker) process for this.

1. Inactive moderators - currently it seems to take too long for inactive moderators to be removed. I know that there are multitudinous reasons why a moderator might become inactive and not all of them require the immediate removal of the diamond, but should the system become more automated?

   The system already knows when you last visited the site and when you last performed a mod action, so do we want an automatic message/email if:

   1. You've not logged in at all in the last [N] months.
   2. You've logged in but not performed any (or very few) moderator actions.

   This would remove the need for the CMs to reach out (at least in the first instance) and this reminder might be enough to prompt the moderator to say "you know what, I don't have time for moderating right now and I should step down".

2. Unpopular moderators - I don't see this as being a reason to remove the diamond. We're not elected/appointed to be popular but to help run the sites. Sometimes that involves making decisions that might upset a number of people.

3. Harmful moderators - these are the cases where the moderator is taking actions that are actually detrimental to the site. These actions might be popular (at least with some sections of the community), but are against the overall ethos of the site or direction that the majority of the community wants to go. People are quick (certainly on Stack Overflow) to go to meta and raise issues they see with moderator actions, so I don't think there's a need to invent a new way to air grievances. If someone doesn't want to go public they can always raise a ticket with the community managers.

   Does there need to be something that collates the meta questions (it could use the score to see how valid the rest of the community thought the complaint was) and CM tickets to show if a moderator is heading in the wrong direction before something bad happens? I don't have an answer here though.

Answer 3

Could we get a yearly check in meta post by the Community user asking if the site would like to have a mod election?

On some sites, a mod election may be necessary while on others the community could be perfectly happy with the mod team. In either case I don't see a harm in asking and in the second it would help the moderators know that the community is happy with what they are doing.

The reason for having the Community user post it on a schedule is that otherwise the mods may feel like the asker of the meta post is irritated with them for some reason, because if they are happy then there is no reason for an election right? While if the Community user posts it then it's just a neutral question that gets asked every so often.

Answer 4

Alright, I can see where a lot of this comes from - elections are a lot of work for the CM team, as are the pro tem appointments. I get that with a reduced team and an ever growing network, you'd want to look at seeing what can be done to reduce your load and increase overall productivity as sites are happier when they have mods to steer the ship and have the number of mods they need to keep the ship afloat. That makes sense to me.

That said, a lot of this seems to devalue a lot of the work, time and energy that mods put into their sites, most of which their community never sees. There's a reason we joke about being janitors, after all. So when I read this and see "should it be forever", and the idea of having to re-run yearly just feels like a lot of mess and disruption. I get it, you don't want to risk the idea that the community feels like the mods are the kings and queens stepping all over the populace (cue that "help, help, I'm being repressed" skit here, if you like). At the same time, all of us who have been here for a while have put a lot of ourselves into making sure the sites are awesome, and this feels very "okay thanks for coming out, putting up with all the crap, dealing with all this stuff as we step away from being present, but here's the door".

Maybe that's not charitable of me, and perhaps I am supposed to hope that my community will recognize my awesome and keep me around. But here's the rub - like I mentioned earlier, a lot of what we do and what we handle, users don't see. We're supposed to be exception handlers and janitors and all that good stuff. A sign of a healthy site is that you shouldn't see the mods unilaterally dealing with and closing all the things. So...the community might not always know that I handled infinitely more flags than my fellow mods by a hundredfold, lets say - they might only see that they personally like ModX better because they're more fun in chat.

I am not sure how to balance all of this. I get that you guys desperately need to relinquish some control because this stuff isn't tenable as things get bigger and bigger. But I am not sure that this, as I am interpreting it, is the best way to go about it.

A further thought - if you are, effectively, forcing mods to step down if they're not constantly re-elected, this kinda flies right in the face of the fact that you have said many times to mods (of elected/non beta sites, who haven't like...blown up stuff or caused giant messes) that elect to stop moderation and step down due to burnout or other personal reasons that their diamond is, effectively, theirs if they decide they wish to step back into the moderation ring again. It feels like youre doing a bit of a 180, here, instead of supporting the people you have, you're just flipping them out in favour of others. This feels like instead of supporting mods who might feel burnout, that you are trying to preempt that by potentially setting us aside.

Answer 5

Consider allowing a long-serving moderator in good standing to have an "Emeritus" role instead of full retirement.

Currently, elected moderators who stop being moderators do so for one of three reasons:

• They drift away from the site, have less time for the job, etc. This is usually accompanied by a substantial decrease in activity. They're not around much any more. They are welcome to return later.

• Something bad happened and the diamond is removed.

• They get hired by SE and might get an employee-diamond to replace the mod-diamond. They still have privileges, but use them much less frequently because they have network-wide responsibilities now.

This proposal would create a fourth:

• They were actually doing just fine and are still active and interested, but you want them to step aside so the community can have another election.

The risk with removing a diamond from an active elected moderator is that it could cause some frustration and resentment, for both the moderator and the community. Here's somebody who's dedicated years to the site, who thought it was a lifetime position after election, and who's now being told "thanks but you're done". Even when people know intellectually that a change-up is good for a site, I anticipate that many will still struggle with it, because people are complicated multi-faceted beings, not logical machines. You'd be taking the people you most want to hold up as good examples and alienating them. That's no good.

Further, there are people whose wisdom and insight we want to retain access to. I could (but won't publicly) name a few individuals who I want to still be part of the mod team even if their activity greatly drops, because they bring history, particular problem-solving abilities, or special skills to the team. Currently they can remain on the team; your proposal would kick them out.

So let's create a way for them to stay without having sites go for years without elections because there's no need to increase the team. (Small sites are especially prone to this; on one of my sites the second election came four years after the first.)

For mods who are only stepping down (or being nudged out) to make room, and who have served the site for some threshold amount of time to be determined, let's create a "moderator emeritus" status.² Functionally, this is similar to the case of a moderator getting hired -- the person retains access to the mod tools and chat rooms and team,¹ is free to handle things when noticed or urgent, sometimes participates in mod chat (especially to answer questions or advise on tricky situations), but by and large does not participate in the active moderation of the site. A moderator emeritus wouldn't carry a normal share of the moderation load, wouldn't count toward a quorum when an official mod panel needs to be called, and would be separated out from the active mods on moderator lists.

Should this status be for life? I don't know. I don't think we are at risk of the emeritus mods overwhelming the currently-active ones, in the same way that the 15 or so employees with diamonds don't overwhelm the three moderators on a smaller site. So I think "as long as the person is active and wants to participate in this way" would be fine. I think it's also fine to go into this with that intention but knowing that we might need a time limit later -- anybody accepting the emeritus diamond knows things might have to change.

I got this idea from my congregation's honorary trustees. Past presidents (it's a big job and you had to work your way up through offices usually over ~ten years) become lifetime honorary trustees. They may attend and participate in board discussions, they have access to all the official doings that officers and trustees do, and they are allowed to vote for several years. After a set time they lose the vote but retain everything else. Most of them have drifted away by then, but it's a graceful transition, not a boot. And if somebody wants to return to "active duty", we'll elect them to the board again. (Under SE's current rules, a mod emeritus who wants to return as a full mod can just ask for it.)

---

Some commenters have objected, citing as precedent government elections (parliament, president, etc) which have term limits. This is a flawed analogy; those kinds of positions have real power and authority (and a treasury and an army) and make policy. Moderators are better compared to the public works department and first responders. Almost nobody elects these kinds of civil servants in the first place, because they're accountable to authorities who are themselves elected. While we do elect our civil servants (mods), they do not have the kind of power that elected governments do, and they are accountable to SE. And we don't fire civil servants every few years just to refresh the slate; we keep staff in the jobs they're doing well.

---

¹ Having your team account deleted because the community needs you to make room for another mod would be especially demoralizing, I would think -- you lose your account and body of work.

² It can be called something else; this is an idea, not a precise suggestion. "Ex-officio", "mod advisor", "venerated elder"... ok, probably not that last one, but this is the kind of function I'm describing.

Answer 6

I think it's worth doing a mix of what we do now on betas, and the election process.

Appointing pro tems is icky

It is. However on a new community, it's extremely useful to get things started on the right foot. The nice thing with the current appointment system is that you guys get to vet the candidates. On one hand, this hopefully means folks with the right mix of SE experience and subject knowledge expertise get picked. On the other hand, this, well, might not scale that well. Y'all can't know everyone.

On a smaller, more manageable site (the good ol' days?) it would probably be a great way to get to know the folks in the community (I'm not sure if betas get their own CM on overwatch any more) as well.

The pro-tem process also kind of got me used to being a moderator, so it's also a useful way to bootstrap members of a community who are not mods on a smaller site with different challenges. Not everyone makes it (that's true of some elected mods too), but it allows there to be a bit more diversity in the broader community of mods. Of the two "newbie" mods on Software Recs - one's a Stack Overflow mod now, and I'm a moderator on Super User, so the process might actually have helped in some way there.

Elections also seem expensive in terms of time and effort, but that's probably a CM might know better than us.

I suspect a mix - of an initial batch of hand picked moderators, with top-ups/replacements until graduation being elections - might be a good fit.

On beta sites, having elections for top-ups and replacements gives a chance for folks with the 'right' personalities to assert themselves before folks need to decide if they'd be a good candidate for moddyness.

Have one unified pro-tem/moderator process

Well, in theory, on mature sites, you're basically running elections as needed. It seems to work, and other than Stack Overflow, most sites are not going to need a massive, churning moderator team. In fact, if moderators are getting replaced often, it's more work, and maybe a sign that moderating the site is a bit too much of a chore.

Moderator replacements take too long

They might. On the other hand, it's a massively manual process. As mentioned earlier, elections are expensive, and personally I feel running them only as needed makes sense, even when it's cheaper.

The community team continues to be conspicuously short on resources to help communities with other issues

This is something that's been a problem for... quite a while. It seems to be getting a little better, but it does feel there's a point where I'd ask, "if it's short of resources, how far is streamlining going to help?".

At some point, there's a distinct lack of fat in the processes to trim.

Simpler Elections, More Often

Simpler elections sound like a good idea. On the other hand, it's worth considering if running them more often than "as needed" makes sense. On the short term, it might be worth using simpler, more streamlined elections to cut back on the backlog.

I like simpler. More often should need a clear reason why it's better.

The community should run their own town hall

Meta is kind of that. We also have a room on Super User for contacting mods, and it works reasonably well. There's a handful of users on many sites (who I shall not name since they would probably be embarrassed) who have an impact primarily through meta and such.

Having read through and realising I misunderstood this completely - They could - but other than initiating town hall question collection and picking the questions, the CMs act primarily to facilitate and organise. I'm not really sure that those processes mirror anything a regular mod comes across ... as a normal mod.

In a sense - the community does but I'm not very sure how devolving the process to the community would work better than it does now.

Should "Moderator For Life" always be automatic… and FOREVER?

The idea of getting mods to "re-up" and letting them take a break and come back has been mentioned before. As someone who's both a mod and an ex-mod - I actually feel the value of a experienced, engaged and accessible moderator kind of goes beyond the site they actually moderate on. So, I'd feel that a mod should be a mod as long as they're able to, and willing to put in the time.

That said, there's a website on which I've tried to resign being a mod on a few times... and I embarrassingly forgot my login, so if a mod wishes to quit, well...

I'd say, a mix of minimum activity and an option to step down at any time would be a great compromise.

I'm not sure of a "great" system of working out when a community needs more mods short of the mods asking. As long as the flags are handled (which is a "easy" metric in terms of statistics), they're able to keep ahead of meta and there's no major/widespread drama in the community over mod teams.

Even where some of our experienced mods are less active, they do form a really useful counterbalance on some decisions behind the scenes, so "activity" is not a great metric.

That's to say, we shouldn't have moderator turnover for the sake of having turnover on a mature site.

No change to the “single transferable vote”

No argument there.

Answer 7

The idea is to make elections much less epic and generational by creating more opportunity for the community to decide who they elect and retain.

This seems to be connected to your bullet point about moderators for life. Can you elaborate on that a bit? For example, if you have 3 moderators and the election comes, 3 new would be elected and if the old ones wanted to stay, they needed to nominate themselves again?

This sounds like a lot of extra work for then active moderators, who had to go through the election process while also volunteering to keep on moderating their sites.

Furthermore, as @fbueckert mentioned in a comment, moderators also have to make unpopular decisions. It shouldn't become too much of a popularity contest. I wouldn't want moderators to have to worry also about the popularity of their potential decision, because there's another election soon. If it's the best decision for the community should matter only.

Answer 8

I hope this is rare, but sometimes an election is the last thing a site needs right now -- there's some issue on the site, some factional dispute or the like, and an election will be contentious and perhaps produce results we'll all regret in a few months. I'm not sure how to address this, but we should somehow "take the pulse" of the community before scheduling an election, rather than just having a fixed schedule.

Of course, if there's a problem that significant, the community and the CM team should be working together to solve it. The "are we ready for an election?" prod would surface those issues if they haven't been visible to the CMs before.

Answer 9

Moderator appointment is important.

As I've always seen it, site elections are a sign of trust between the Community Team and the site itself; you effectively trust us to be able to select our own leadership without running the site and the network's good reputation into the ground in the process.

Early beta sites really haven't earned that trust. And there's also long-running beta sites which still haven't earned that trust. Appointing moderators instead of throwing it to the fickle winds of the voting public is still necessary to ensure that sites have proper leadership when they for whatever reason are unable to choose that leadership for themselves.

However, I do agree that there's no real reason for the Community Team to do this most of the time. Except for the very first moderator selection on a fledgling site, in most situations where appointing a pro-tem moderator is necessary you already have an existing moderator team, or at least part of one; these are the people that you have already trusted to know how to effectively lead the community, and whose judgement you're (presumably) already relying on when deciding who next to appoint. Why not just cut out the middle-man and let them deputize their own pro-tems?

I would still want the Community Team to sign off before giving any deputized pro-tem access to PII and such, but the majority of flags and work that would necessitate having a pro-tem in the first place doesn't involve more than just having elevated mod-level privileges (e.g. one-vote close/delete/undelete powers). The community leaders should be able to decide if and when there's need for a few extra hands while they're waiting for the next election; there's no reason to backlog the Community Team with this. If a new "real" moderator is needed urgently, we'll let you know.

Answer 10

We should do a better job of giving "drive-by" voters relevant information about the candidates.

Right now the election screen contains very little information about how a particular candidate interacts with a community – instead, it provides "objective" information like reputation, time on the site, candidate score, and badges, in addition to the candidate's blurb.

But besides what the candidate wishes to reveal in her bio, there's no information about the candidate's views on the site, her approach to moderation, her approach to conflict, and so forth.

This puts us in the situation where high-rep users, after a long hiatus or period of conflict with the community, can attract votes on the basis of name recognition and "good stats" from drive-by voters who got their minimum voting reputation some time ago and haven't been engaged recently. They get a notification that there's an election, and they vote, and that's that.

One way to help address this is by developing better "objective" stats that incorporate things like recent review queue activity (as opposed to years ago), recent flag count and acceptance rate (not just total count), recent meta posts count and score, and so forth, and displaying them prominently, rather than only showing metrics based primarily on rep and long-term badge achievements.

I believe a better informed electorate would help make elections less stressful for communities. I'd also speculate that this would reduce the need for CM involvement following controversial elections, though this is just a deduction with no data or even anecdote to back it up.

Answer 11

The recent pro tem election on the Vi site has left me with mixed feelings.

There were four candidates, but:

1. Candidate 1 is an active member of the community.
2. Candidate 2 is a sporadic contributor (mostly from when the site was new). I would still consider him to be a good mod, but that's mostly based on contributions I've seen on other sites (he's a great mod on another site).
3. Candidate 3 is a prolific contributor on other sites, but not on ours. I think few people knew who they are and thus if they're a good mod. Besides, every site is (subtly) different and what works well on one site doesn't necessarily work well on another other. You need to be in touch (and agree with) the site culture to some degree.
4. Candidate 4 has very few/sporadic contributions on any site spread out over many years.

There wasn't really a choice: only one candidate was clearly qualified. I think no one was surprised by the outcome: candidate 1 got 72% of the vote.

It's not that I'm unhappy with the outcome; they'll make a great mod. But it seems to me that the entire election process was just ... unnecessary? I'm not active on the parenting site, but looking at their election it seems there was a little bit more choice, but one candidate getting 60% of the votes seems it wasn't a difficult decision there, either.

Honestly I'm not sure if the election really improved anything over just appointing the mod. On the other hand, I don't think it did any harm, either. At least, this time it didn't; but we'll have to see how future elections turn out (what if there are only four candidates who aren't really qualified?)

---

Small update: one of the nominees self-deleted their account a day after the election results were announced. While we can't be sure of the reason, I think it's not unreasonable to assume that that disappointment with the election results is a factor. No content is lost and the user's contributions were sporadic, so not a huge deal; but it does highlight that elections can be a risk and have negative effects.

Answer 12

I like the overall idea, but I'm nervous about a few things with replacing pro-tem mods, in particular right out of private beta.

• Lack of solid knowledge. The community doesn't know their candidates that well, they probably haven't seen them run into many tough situations yet.
• Lack of voters. There won't be that many users yet, and if all goes well, the site will quickly grow to where most users didn't vote in that first election.
• Lack of risk mitigation. If we get bitten by the previous two points, we don't have a good safety net. If a mod elected early on does turn out to be a bad choice, it's difficult to avoid a lot of drama, which is always harmful, but perhaps even more so for a site early in public beta.

All of that said I absolutely think that we should have elections eventually, and pro-tem mods should be a temporary solution, and making elections simpler and thus easier to have (e.g. for replacements) sounds great.

Answer 13

Have we considered the effect on private information?

I don't know if this is neutral or negative, but more people cycling through the moderator positions means more people who can see all the mod-private information, like suspension histories and PII. Everybody is bound by the moderation agreement (including after they step down) so I'm not concerned that things are more likely to leak out, but over time more people will see private communication between mods and users. This could become uncomfortable for the users involved; that "be nice" suspension of a prominent user with all the juicy details in the mod message is visible to an ever-increasing number of site users. What will the effect be? This concern already exists whenever we add a moderator, but we're opening the door wider now.

Do we care? I don't know. Just from an information-security perspective, we should ask ourselves the question.

Answer 14

I think the mod for life position works quite well, and I would not change it to have mods reelected annually, or some such.
There are plenty of cases where mods step down, or get removed and new users step up and are elected to replace them. Or in another case, a site grows and it gets more mods. That all works (could use some tweaking on the inactive mods steeping down part) but the mod for life bit is central to how our sites are run.

As others have stated changing the mod position to focus on reelections is dangerous (just look at politics).
If instead of curating the site we help run, which includes some actions that are unpopular, I was just filling my term looking to the next election, I'd always be thinking about how this would effect my votes (again look to politicians).
Take an example from our site, blender has an all new still experiential version 2.8, we are getting questions about that new version. There is an ongoing debate about whether or not 2.8 questions should be on topic yet.
If I was thinking about getting re-elected, I'd say sure ask away (that is the popular thing), however if I'm thinking about the longevity of the site, and the usefulness of the questions/answers I say no. All that is still changing, next week that button might not be there.

---

Ok, back to the problem of elected forever. Do we realistically expect a mod to volunteer for decades? (Some people have life to do, and simply can't.) No, nobody does.
We need to make it very clear this is an opened agreement.

Moderators are elected to help run the site for as long as they want to.

When a mod gets busy (new job, school, baby, family emergency, moves, etc) they should simply be able to leave.
No pressure, no expectations from the users or SE. They could very well be here for decades, or just a few months. Whatever it is, mods should feel free to come and go as they want.

Which brings me to my next point.

Once a moderator of SE, always a moderator of SE.

After a moderator is elected and signs that Moderator Agreement, they should be able to return as a mod any time they want (say they finished school, or that baby grew up) even if their position has been filled. Granted they left in good standing, and have stayed even marginally engaged with the site (this does not go for the few that get asked to leave, or are kicked out).
You can't have too many mods. So if a mod who helped start the site years ago now has time again and comes back, it is a win win. There is less workload on all the other mods; there is less to learn then a brand new mod; and the site keeps its energy with "fresh blood."

Lets say there is a mod "Frodo Baggins." Frodo is more busy then he use to be, he's still on the site nearly every day handling flags and all, but if he knew he could step down, let one of the super active power users take over, and yet still be able to come back as a mod, he probably would step down.
See at that point there is no reason not to. The site benefits because the more active a mod the better, and Frodo has lost nothing by leaving.

In our current system, there is a huge disincentive to leave as a mod. If you leave, you leave all that power (even if you don't use it that much any more). Our sites suffer when semi-active mods do not want to give up their seats, because there is no guarantee that they will ever be able to come back.

---

So instead of forcing the mods to keep getting reelected (which has a whole host of its own problems), just remove the reasons the mods do not want to leave.

---

TLDR
Mods keeping a re-election mindset is dangerous. Sites' quality will suffer.
Enable mods to freely step down and rejoin when they want. Solves most of the inactive mod problem, keeps sites fresh and active, and the special role mods play safe.

_{Any references to "The Fellowship of the Ring" or "The Lion, the Witch and the Wardrobe" may be incidental.}

Answer 15

If the downsides of more frequent elections could be avoided…

There's one notable advantage of more users cycling through moderator teams: increased perspective among non-mods.

There are many good points here so far about concerns with having more elections. I agree with almost all of them.

If the problems with more frequent elections could be figured out, I see one big advantage: the more users who have experienced having to moderate a site, the more “forged in the fires of moderation” users we will have. I think this would be good for the sites.

Becoming a moderator has an interesting side effect: the rules look very different when you're the one trying to judiciously apply them. This change in perspective has been remarked on by and is very valuable for myself and my fellow local mods. If we could get some of that outside the set of mods, I think it would be a great benefit to a site.

After all, it's one thing for the people who control the levers judging and enforcing the rules to have a big-picture view of tending to site health and operation. It's even more valuable to have that perspective among the general membership. It makes for a lot more cooperation and a lot less friction. But, it's less likely to spread from “The Man” of lifelong mods though, and more likely to spread among non-mods by seeing other non-mods model it.

As much as the idea of standing for election every year (or whatever) gives me a headache thinking of how it would politicise every mod action, if we could somehow get more regular users to have moderator experience under the belts, I think that one thing would be of great benefit to site health.

In short, moderator experience makes for more conscientious users, and we can always use more of those. If we can figure out how to get more users real moderation experience without breaking anything or anyone else, that would be good.

Answer 16

Okay, I am throwing a suggestion out there. It’s very rough and mostly food for thought. Also, it pertains to “running” sites with an established moderator team and probably does not translate well to sites on the scale of SO or the community of SciFi. I am aware that this goes against the premise of sticking to STV, but while STV is great, it is simply not made for what I am suggesting.

Do not hold moderator elections anymore, but have suitable users approved or disapproved by the community, when there is a need. In a bit more detail:

• Introduce a score that quantifies how much a new moderator is needed. This can be calculated from flag-handling times, some special button each active moderator can push, or recent retirees.

• Introduce a score that quantifies how suitable somebody is for moderatorship. This could cover some basic community-moderation achievements (like the moderator score in current elections) but most importantly should be based on anonymous votes by experienced users. The latter could roughly work like this: Once a user reached a certain reputation, they can nominate up to e.g. three other users for moderatorship (and change that list at any time). The value of these nomination votes could somehow mildly scale with reputation or be higher for existing moderators.

• When the ratio of the two above scores exceeds a certain threshold (i.e., there is sufficiently much need given the qualities of a candidate), the respective candidate is:

  • asked whether they actually want to be a moderator
  • vetted by Stack Exchange
  • …

• If all of the above is positive, the candidate writes a brief text to introduce themselves, is subject to questions by the community, and most importantly subject to a vote. In the latter, voters are only asked to express their approval of this particular candidate. If the result of this is positive, the candidate becomes a moderator.

This could address the following issues:

• “Moderator replacements take too long” (see question)
• “Simpler Elections, More Often” (see question)
• No moderator elections if there are no suitable or willing candidates.
• Excellent candidates can become moderators when they are ripe.

Answer 17

Should "Moderator For Life" always be automatic… and FOREVER?

I'm incredibly leery of this aspect. It completely changes the paradigm of moderators, and not in a good way.

I get that SE is trying to make elections take less from their staff to enact. But...that's kinda the cost of doing business in this system. If you want final oversight and control of the site, you need to keep your hands in it. Less involvement implicitly means less control. See chat and the various debacles that have happened there.

If you want to take a step back and automate the process more, while requiring moderators to stand for re-election, you're telling the community that their input and decisions matter more than the overall health of the site does. You're going to face massive backlash when you see something isn't working right, and stick your oar back in.

If you want the oversight, you need to put in the work.

Answer 18

Should "Moderator For Life" always be automatic… and FOREVER?

There are, and should always be, procedures by which Stack Exchange or the community can remove a diamond moderator appointed "for life". (Existing procedures are: removal by a two-thirds vote of the rest of the site’s moderators,¹ and removal by Stack Exchange, Inc. for violating the Moderator Agreement.²)

You seem to be proposing a procedure by which the whole community can remove a diamond moderator by majority vote, when there are more candidates than there are seats. A better alternative to consider is removal by a two-thirds vote of the whole community.

I suggest that if you do establish a yearly election cycle, it is a great idea to list existing diamond moderators on the ballot, and give them an opportunity to campaign, but they should be removed if they receive twice as many downvotes as upvotes.

Why? Because moderator elections are not democratic. I am not criticizing this. It is just important to keep in mind. A community cannot set the rules of its own elections. It cannot set terms, term limits, voting procedures, it cannot determine the powers of its elected officials. Robert’s Rules do not apply. The closest the community comes to establishing bylaws is voting on meta posts which propose community policy, and even then community policy cannot contravene Stack Exchange policy. The moderator agreement is with Stack Exchange, not with the community. In short, elected moderators are community representatives, sort of. But it is closer to the truth to say they are appointed volunteers, and that Stack Exchange uses elections as a way to crowd-source and automate the decision of who is safest to appoint.

So the question becomes, how can Stack Exchange crowdsource the question of which diamond moderators they should remove because they are not an asset to Stack Exchange? For that there should be a community consensus, that is, the two-thirds majority which is conventionally the minimum for removal from any office.

Notes

1. This procedure is detailed at “Handling Calls to Remove a Moderator”. A two-thirds vote is fairly consistent with what Robert’s Rules calls for. Robert’s Rules begin with confidential investigation by a committee that prefers charges, followed by a trial of the facts by the assembly (censure a.k.a. impeachment), and, if censure passes by a majority, a second two-thirds vote on removal then takes place.

2. “Stack Exchange Inc. reserves the right to terminate my privileges as a moderator at any time without warning.” This is a clause in the Moderator Agreement. Termination could presumably take place whenever a diamond moderator violates the agreement. I do not know if this procedure has documented steps that provide for a fair hearing or due process.

Answer 19

Is there an interest in broader election reform?

From the number of posts, it seems obvious there is interest. Not a lot of consistency in what direction it should take but there is interest.

A lot of the posts opposed to recurring elections are about the drama (and pain) of the current process. The suggestion is for a simpler less painful process.

Long before I was active on SE, I had a mop at a couple of the en.wikipedia.org sites.

The process used at https://en.wikisource.org/wiki/Wikisource:Administrators was not overly painful, quickly removed the inactive, and provided a route to return. It is one of the most graceful and pleasant processes I have seen.

Much like here, everything there is CC, so it can be leveraged on easily.

Assuming the process is not painful, consider the statement below about inactive (source link)

It is a common occurrence on wikis to create an account, be active for a certain period of time, then become inactive. While this is to be expected due to the nature of this site, accounts with restricted access should not be left unattended for a long period of time. Users who return after an extended break may no longer be in touch with community practices or policies, which is not desirable in an administrator. Inactive accounts are also more open to account hijacking, and this may be difficult to detect as compromised if the original user is not there to notice edits they didn't make. Furthermore, the list of users with restricted access should only contain those capable of responding to queries within a reasonable period of time.

On Wikisource, an inactive sysop will generally have his or her rights removed. An "inactive administrator" is one who has not edited during the past six months and has not made more than 50 edits during the last year. Inactive users automatically lose their restricted access in their next scheduled confirmation of the voting unless the community supports continued access. Any user who has lost access due to inactivity may reapply through the regular processes.

Answer 20

There are too many words, comments, answers to read them all. So, I'm going to focus on two points. If this has already been said by someone else, apologies.

But should communities have a say when that happens?

This section is unclear. Are you saying that users should be able to voice dissent? That users should have the possibility/right to remove elected mods (after a reasonable amount of time)? If I've interpreted this correctly then…

How long a time?

May I suggest two years? If someone is a good fair-minded mod whose presence is unintrusive until needed or an unreliable biased person, I believe two years is sufficient time to either confirm or remove their title. If their status is confirmed, then it's guaranteed for life 5-7 years. During which time, a mod can freely choose to resign whenever RL intervenes and new elections should be held, if and when the community feels that necessity. Turnovers are necessary for the well-being of any public institution, private company and workforce, the same principle applies to websites.

Who votes?

What about the users who have the right to vote?

Should a user who has only posted twice in five years have the same say as someone who has regularly contributed to the site? Should a user who has never posted on meta have the same privilege as someone who has participated, not every week, not every month, but at least (hypothesis) four times on meta in any given year EDIT: this includes posting comments and voting. But for candidates, I want to vote for someone who sticks out their neck every once in a while, and I expect them to have been an active member also on meta, not only on the main site.

It's a fact, but what I'm about to say will not be liked, not all users are created equal. Some have earned more of a voice than others.

Answer 21

I think that regular elections are simply a horrible idea. Having to go through the circus of elections on a regular basis would be a very strong incentive to step down. I am currently a mod on 3 sites (elected on two, appointed on one) so I would have to go through three elections every year if we made this a yearly thing. That would be very tiring.

I also think it is very counterproductive to make mods a more political position than they need to be. Moderators need to feel they can do their jobs, which sometimes involves making users angry at us, without fear of this leading to them losing the next election. I really don't want to have a situation where moderators will consider their actions through the distorting lens of how they would affect their reelection bid. I can't see that having a beneficial effect on the quality of moderators' work.

However, I do think it could be useful to give the community a simpler way of expressing its feelings about current moderators. To that end, I propose we have a running mod score. Each mod's profile could have two special voting buttons and each user would have the right to up- or downvote the mod (users should be able to change their vote at will). If a moderator's score falls below a certain threshold, and if the total number of votes cast represents a significant proportion of the community, that could trigger some sort of action. Perhaps a mini-election (like a vote of confidence) where users would be asked to vote on the mod retaining their position. Of course, the score should be completely ignored until enough users have voted one way or another that the score can be taken to be representative of what a large part of the community feels. The score also doesn't need to be visible to anyone but SE employees.

I think it is important that the communities feel their mods (at least their elected mods) always represent them and there should be a better way for the community to remove a mod when they feel the mod's no longer right for the job. As it stands, the only way a community can de-mod someone is by contacting SE or raising a huge stink on meta. This also means that a few vocal users who dislike a mod can have a disproportionate impact and cause all sorts of drama by complaining on meta. If we had a moderator score as I describe, that could give a much better, more objective and more representative measure of how the community really feels about its moderators.

Answer 22

I think these are good ideas for the smaller exchanges. Moderation on low volume sites can be difficult for only a few users to shoulder, and the community is often hard pressed to meet the numbers themselves.

Moderation of the smaller sites has been problematic in the past, so taking a slightly different angle could help keep the sites clean while also increasing interest in participation and contribution.

That said... I don't think this makes sense for the larger sites. The moderation at Stack Overflow, by far the largest site of the exchange and the lifeline of the company, is in a pretty coalesced state. Its overall process, lineage of term, and selection is being handled very well. I haven't seen really any complaints about it there, and I am an avid follower of their meta site, a frequent contributor, and have ran in a few elections. They were fair, the elected moderators have done well, there was a lot of community participation across the board, and the frequency seems balanced.

tl;dr; Good for small, avoid for large. Please do not do this for Stack Overflow.

Answer 23

Suggestion: stop using down votes for mod election votes (primaries¹).

Does it really matter who you are against? Who you are for is what matters.

The use of negative voting in SE primaries, the down vote button alongside the up vote button, remains a sign of the flaws in SE community's attitudes towards people. We have done a thing lately in trying to improve that. Carry that momentum forward.

Down votes during mod primary phases in elections: not necessary.

Add up the "who I am for" votes (the up votes); we each get three in a given election. The winners will still win.

The down vote tool for questions and answers retain their utility for the things -- questions and answers -- that is this site's attraction.

People aren't things, so please stop treating them that way (in the mechanical sense).

¹ primaries point edited in thanks to feed back in comments from @MonicaCellio.