Carl Mummert has suggested that we take some time to determine community standards for what we should require of a potential moderator. I don't know if it's too late to stop the train, but it's probably worth talking about seriously, if only for the next election.

I've previously seen comments that during the election may not be the best time for such discussions. I'm not sure I see the logic behind this; what changes between now and the end of the election?

What changes? we will learn more about the process by observing it. For example, we will have the complete list of nominees, the list of those who advanced to second round, and who won. Based on that, we can develop criteria that are more restrictive than what is in place now, yet do not inadvertently eliminate worthy candidates.

If the comment was this one, then I'd like to clarify: I don't mean we should not be discussing the issue, but rather we should not make final decisions during the campaign.

@Behaviour I seem to recall the opinion being expressed by multiple users, but don't have the links to back it up. Anyway, you make a sound argument.

I withdrew my nomination after reading some meta discussions on the... erm... bloated nature of the nomination page (I'd love to moderate here - but 3 months is clearly not enough to be what the community wants a new moderator to be). Do I have to make a concession speech or something?

Not at all.

There were quite a few users who nominated themselves who I thought clearly were not good moderator candidates; I do not recall you being among them. Nonetheless, if you do not think you would be a good moderator by current community standards, I appreciate you taking down your nomination.

@MartinSleziak I've had an aops account since 2011 and he was a mod then...

@AnalysisIncarnate I've posted a comment on his nomination about this:

math.stackexchange.com/election/5?#comment-2158458

Hm, it seems that I have to copy it manually.

In chat someone mentioned that you have been mod at AoPS. If this is indeed the case, it might be important enough to be mentioned in your nomination. In fact, there is also a related question in this list.

@GraceNote Hi :-) Just noticed the off topic comments vanished. Thank you.

No prob

@GraceNote Speaking of off-topic comments, there appear to be a number on Anastasiya-Romanova 秀's nomination. (I would call the first, 3rd-5th, and Sawarnik's, off-topic.)

Link?

Ah, I hadn't realized these were linkable. Here.

Done

Appreciate it.

Hi, @Jack.

Hi, @MikeMiller :)

Hi @JackD'Aurizio

Good luck for the moderator election :)

Thanks @Venus, I appreciate it.

@Venus Probably meant you're on your own. Good luck @JackD'Aurizio:D

Many thanks, @Integrator ;)

Uh hu. I wonder whether he really believes what he writes.

I've no doubt.

I'm glad to see @Arkamis nominate himself. I think he would make a good moderator.

@MikeMiller Good news.

@GraceNote: do you have a way to see what the largest election was? that is, the largest number of candidates.

@robjohn There's a cap of 30 nominees in the Nomination Phase

So, the largest was the times that SO hit that.

I didn't know there was a cap... a bit worrying

@GraceNote Ah, so we are over 2/3 of the way there... I thought it was getting kind of out of hand.

@MikeMiller What actually happens (I think) is that once there are 30 nominations, the next nominee must have more reputation that the one with the least reputation, who is then removed as a nominee.

ah, perhaps less of a big deal than I thought. We have five nominees with less than 1000 rep and eight with less than 2000.

I doubt I would vote for someone who doesn't have access to and use all the review queues, but especially not one that can't access the close/reopen queues.

(There are 11 below 3000, and then there's Kasper.)

@MikeMiller There are users I'd vote for without access to those queues on math. Most (if not all) of them are moderators on other SE sites. (And I doubt (m)any of them would actually nominate themselves here.)

Yes, based on that comment, I have a user in mind as a counterexample. But of the current nominees, at least, I'm sticking to my assertion.

Yeah, in the battle of theory v practice, practice tends to win in real-life situations.

How the capping on nominations works in case of withdrawals?

@JackD'Aurizio Withdrawn candidates are no longer candidates, so they're not part of that count. (Unless I misunderstood you.)

I was wondering about the following situation: the capping limit is reached, so no one is able to nominate himself anymore. If one of the 30 candidates withdraws, is the nomination phase re-opened?

@JackD'Aurizio It's always opened. Just that when the maximum number of candidates is reached there is a de facto increase in the min rep requirement. If there is a max number of candidates, and then someone withdraws (or is withdrawn), the requirements go back to the default.

Except now I'm certain this is the way it works.

@ArthurFischer: all right.

Hello.

@PedroTamaroff Oh, we have one of the future moderators here. Hi :)

Especially read Shog9's answer. Some changes have been made, but nothing that affects the current nominees.

This is an interesting choice, but a good one, I think.

Thanks for the changes, @Shog9.

I want to write a better query for "citizenship score", something to sort users by. In the spirit of the queries here but less rigid. Ideas?

I suggest using actual counts of flags, meta, reviews, rather than the badges, weighted appropriately.

(As a factor, that is.)

I don't think I SEDE has the number of helpful flags. I can use reviews.

The meta posts can be rants or useful posts. The count does not differentiate.

My current idea: minimal rep 1000. Get 1 point for every extra 1000 rep until 10000 (so, for the maximum of 9). Get 1 point for every "noteworthy" badge on the election stats page. Then somehow factor in the number of reviews.

But SEDE can check score, no?

There are a few of the "noteworty" badges that I don't think are particularly noteworthy. Other than that, that sounds like a nice scheme.

While SEDE has the score of meta posts, it's stored in a different database, the one for meta.math. The rest of the data is pulled from math. So this would be a cross-database query, which is a bit beyond my SEDE skill at present.

Outspoken is the obvious one; I do not think presence in chat really relates to moderation ability at all. I'd personally also scrap the voting things, since I don't think those relate either, but I understand quite a few people disagree.

That Pedro loses points for not having Electorate is a bit silly, seeing as he clearly votes on a question or two. But it's probably not possible to account for such edge cases in a rudimentary 'citizenship score'-type thing, and maybe it's not desirable, either.

Okay, I will replace Electorate with "600 votes on questions", i.e. drop the 25% requirement.

I never liked that requirement myself.

I'm not sure why it didn't occur to me to just remove that. I like that idea.

So, how to weigh in the reviews? I can get their total number from SEDE.

I'd suggest something similar to reputation (maybe a point per hundred, capped at 1000 - the gold badge? This might be too high).

I think that may work, because I'll take the total of reviews from all 6 queues. Gold badge is for 1000 in one queue.

Also, 1 point for Quorum badge (bronze version of Convention). Helps differentiate the folks with little, but nonzero meta activity, from those who never post.

I was going to suggest you add another badge if you keep Outspoken, because a 29 point maximum citizenship score is gross.

Wait, I lost count. Let's see: up to 10 for reviews, up to 9 for rep, one for each of Deputy, Civic Duty, Convention, Strunk & White, Electorate*, Sportsmanship, Pundit, Copy Editor, Quorum....

I suppose it was me that lost count; I thought you got one per thousand reputation.

So 2000 \to 1 pt, 3000 \to 2 pts, and so on?

Sure, that can work. So, 1-10 from rep.

Well, 0-10. Could be you have sub-1000 points.

Those I want to simply exclude. But it does not contradict giving 1 for 1000, etc.

Alright.

You seem to have removed Outspoken; now my desire for 30 points means I need to go find a badge to replace it...

I suppose one could throw in "Yearling".

Look, we have the edited-question-you-answered things.

Those are nice.

Which are not on the election stats page because they are so new.

But your suggestion is also very good.

Because 1 year on the site, while not required, says something. I'll go with Yearling.

You could also throw in Explainer and Refiner and drop the review- and reputation- count by 1 (2000 \mapsto 1 as before). Or just add them; I doubt the numerologists among us would object to 32.

I can't drop review count without disqualifying those with <100 reviews, which I don't want to do. Nothing wrong with 32. But I'd use Refiner and Illuminator, because Explainer is just 1 edit.

Can you run a query as to what proportion of users that have a positively scored answer, have the Explainer badge?

I understand using Illuminator instead, but I think the requirements are a bit high.

Maybe put in Explainer* with 10 edits instead of 1.

By the way: since both @ArthurFischer and @MichaelGreinecker commented on my post with those queries, perhaps they could add their input to the above?

Hm: my profile says I have ~580 reviews, while the close vote queue says I've done ~560 of those, the low-quality post queue says I have ~100 of those, etc. Any idea why the discrepancy shows up?

@Behaviour I like the queries, but one should have maybe several options side by side. Otherwise users may become vary that the many parameters have been fine-tuned to support specific candidates.

@Behaviour @MikeMiller Before I comment on (or really look at) these queries, perhaps your conversation can be moved to a different room. Nominee requirements aren't going to change for this election (save the implicit increase in min rep due to being at the max number of nominees).

@ArthurFischer The discussion is meant to provide a way to easily rank the users (both nominees and potential nominees) by their "citizenship". It is intended for the current election.

Nonetheless, it might be fair to move this conversation to a new room and come back with the resuls.

For example, the early stages of "informal nomination" thread consisted of users looking at reputation list and posting everyone in the first two rows. After I posted the first SEDE query, someone suggested hardmath and Jyrki Lahtonen and Jonas Meyer... and user7530 even began thinking of moderator role for themselves, specifically because of being on the list.

@Behaviour First a preamble. Today I went back and looked at my own nomination from 2013. My reaction was that now I wouldn't vote for that milquetoast loser. I don't really have of a problem that I would be disqualified (or penalised) based on not having the Deputy badge, or other factors. I didn't have the goods at that point.

@Behaviour Rep can be easy or hard to obtain, and I'm not keen on putting much stock into its meaning. Happy that the bonus is capped, but uncertain where it should end.

I can make the cap a parameter in the query. The main use case is to identify and highlight potential nominees, those whose work toward improving the site does not make headlines.

Once nominations are made, people should use their brains rather than a query.

@Behaviour Badges that I would put meaning into on math: Investor, Scholar, Enthusiast, Quorum, Deputy, Critic, Custodian, Strunk&White, Supporter. The answer badges I generally dislike for math, on other sites they might be fine. Except Explainer. And bonus for the Reinfer. Missed those.

@Behaviour I kind of wish people used their brains before nominations. <shrug>

@ArthurFischer Don't undersell yourself. From the nominations alone, my three votes would have gone to Mssrs. Fischer, Greinecker and Gruber [listed in alphabetical order], had I joined two months earlier. Experience has shown that these votes would not have been wasted.

@DanielFischer I'm pretty self-critical: it was a terrible nomination. The questionnaire response may have been worse, saved only by my use of links. I've been told that's a good thing. Possible even ones this like. But I was among the rep leaders. A fairly common refrain in the Teachers' Lounge: if it wasn't for rep, I wouldn't be here.

@ArthurFischer Good thing that nobody reads those, then.

@MikeMiller Mods are supposed to moderate chat, too, so I can see some worth in this badge. But I don't see many chat flags (and hardly ever act on them since they invariable come from somewhere else).

@ArthurFischer Not so terrible. It showed that you had some idea about what moderation might be about. Of course now you know much better, but that's natural.

@Behaviour Don't. Or, if you can, only care about the reviews where "edit" is chosen. The rest are probably garbage. I suspect there's a lot of "reviewers" who just check to ensure that it's not an audit and then click their pre-determined favourite.

We also seem to have some users who stalk the queues.

Good point. I'm skeptical of reviewers who blast through the First Posts queue leaving "hi, welcome to MSE :)" and not editing.

Then I can simply count edits, from reviews or not. 1 point for each 100 edits, up to the max of 10 points.

Or simply min(edits/100, 10), if we are not afraid of real numbers.

I'm terrified of them.

@DanielFischer That, and I see how little most users actually do in terms of user moderation. Why edit a single solitary expletive out of a post when you can flag it for someone else to do it? Why spend time looking through the tag synonym suggestions?

@ArthurFischer What a question. If you flag it, that increases your helpful flag count, one closer to a badge.

@DanielFischer Only if someone marks them as helpful. Making more work for me is not helpful.

But how many flags does it take until people learn that?

The average user that does lots of these things is probably not good at them; I don't think that's a good reason to leave them out of the count. To tell apart the good from the bad we can use our eyes.

Almost OT but the synonyms page is really a pain.

@DanielFischer Learn?

@quid Here, yes. I almost feel like going through and cancelling everything suggested pre-2014.

Sometimes, the optimist in me surfaces.

... and then you look at the site again.

:-)

@quid I'm glad to see you active on MSE and its meta lately.

Thanks. I also enjoy being around.

So, @quid: When's your nomination going up?

Then after you're elected you can handle the flag on meta your last comment just brought up. ;-)

My last comment?

> "...there can be difference of opinion what is useful."

(More than 20 comments in the last three days.)

I see.

Such a flag.

But qute in full "we seem now back to square zero, perhaps time to stop. "

Kind of anticlimactic. I went to the profile, saw this and thought WTF?

@quid They're a pain, but I sort of like them. Shunting stuff off to chat isn't the worst thing in the world.

@quid Just intended for you to know which comment I was talking about. Sorry.

I apologise, @Arthur, for causing those a couple of times in the last weeks.

Yes they are not bad. Also one gets alerted to where there might be some hidden issue.

No worries. @ArthurFischer

@DanielFischer I honestly don't recall any from you, but I don't recall every flag I handle.

@quid Do you get a lot of arguments on MathEd?

@ArthurFischer A couple of tutoring sessions you moved to chat.

Two, iirc.

I am a passive reader there and the site seems fairly calm. There are no great debates like the one or two we have here.

@ArthurFischer sometimes yes. But not that many.

I suppose most (meta) conflict here is really stemming from one debate.

@DanielFischer And you didn't raise a meta thread about mod abuse? You might be suitable for a position that's opening up around here.

The meta there is super idle. If anything people discuss on main; on the subject not meta issue that is.

@ArthurFischer I should have. Would have brought me closer to "convention" ;)

I think Tag Editor is also something to consider. My current list, taking some (not all) of Arthur's suggestions: Investor, Enthusiast, Quorum, Convention, Deputy, Strunk&White, Copy Editor, Explainer, Refiner, Illuminator, Yearling, Civic Duty, Sportsmanship, Tag Editor, Reviewer* (250 reviews in total), Electorate* (600 votes on questions)

The asterisks mean "redefined" badges, where I'm going to tweak the definition to better suit the purpose.

Sorry, all. But I've gotta run. Ciao.

See ya.

@Behaviour I still suggest changing Explainer to something with a slightly higher bar to entry: maybe 5 or 10 edits.

Only 1.9k users have Explainer, vs 19.9k with Teacher. And, say, 3.1k with Enthusiast.

Only 1.9k have Explainer? Shocking. Alright.

I think @Behavior needs to run in this election, if only to give him something else to occupy his time.

If you want to occupy his time, you could suggest he gets started on the SO review queue...

@Shog9 Many had that idea, but he declined.

I think he's just being coy

@Behaviour Here's one passion of yours that would benefit from moderator tools: crushing Thierno M. Sow posts.

@Shog9 Maybe if you pay him ;)

The comment was one of the most hilarious aspects of the election process sofar.

@Shog9 Atwood proposed "citizenship score" way back when. I want more visibility for users who give a шит about the site (quoting Arthur; rare use case for Cyrillic). Not just during the election.

@Behaviour eh. I'll think about it in the unfortunate case that there still isn't a good selection of more qualified nominees in a few days.

@Behaviour well, it's an interesting idea - though I'm not too hopeful as to the utility of an opaque number. Badges are actually a pretty solid way to see who has been well-rounded.

@DanielFischer if only I needed to pay folks to set that gal off

@Shog9 Blackmail, "run for mod or you'll be network wide suspended for a year"?

Though, if there is the slightest risk of it not working, that's not a good idea.

"run for a mod or I'll publish your helpfulness score"

That'll do it.

Ah, the PH index. My activity on Meta.SE dropped recently mostly because of that thing. Don't get me started again. There are opaque numbers, and then there are stupid, misleading, opaque numbers.