21

On some smaller sites...

Without pointing fingers, I've seen widespread diamond moderator closures, taken to the extreme where almost all closed questions are closed using a diamond hammer; not just for urgent closures, but virtually all closures.

I'm certainly not claiming the closures are incorrect [the questions are rightfully closed], but I'm bothered by how this behavior excludes the community from self-moderation. Virtually no question gets closed nor reopened without the few (one or two) highly active diamond moderators. Those few diamond moderators (not the community) are the gatekeepers of the site, and close votes function as flags.

The community seems to become dependent on the small number of diamond moderators. What happens when the site grows and it becomes overwhelming for these few diamond moderators? What happens if the diamond moderators were to unexpectedly leave?

Is this a good thing or a bad thing?

My instinct is that this is not how moderation is intended:

Moderators are human exception handlers, there to deal with those (hopefully rare) exceptional conditions that should not normally happen...

... Keep the site reasonably on topic by closing ... blatantly off-topic questions.

The ideal moderator does as little as possible. But those little actions may be powerful and highly concentrated. Judiciously limiting your use of moderator powers to selectively prune and guide the community — now that’s the true art of moderation.
Jeff Atwood, 2009

However:

When you see a post you think should be closed, close it. When you see a post you think should be re-opened, re-open it. If you're not sure, don't do either. We've entrusted you with the ability to perform both of those tasks instantly - so use them judiciously.
Shog9, 2012 (emphasis mine)

I guess when writing that post, Shog9 did not envisage a scenario in which virtually every closure is a diamond-hammer closure. For example:

If the community decides to allow it, they can vote to re-open, and if the discussion clearly indicates support for the topic, you can re-open.
Shog9, 2012 (cont.)

No they can't vote to re-open! The community does not have sufficient population nor experience in self-moderation (closing and reopening). Only those few diamond moderators can reopen; the community needs diamond moderator permission to reopen.

I'm torn here. On one hand, close-worthy questions are being closed. On the other hand, the community's ability to self-moderate is squashed.

So...

Should anything be done?


While I don't really want to point fingers, perhaps it'll be considered more seriously with recent numbers. Here's some examples on sites I'm not active on:

                 Action                  Moderators Community¹
---------------------------------------- ---------- ----------
Questions closed                                 20        241 [BiblicalHermeneutics.SE]
Questions closed                                 65        314 [Astronomy.SE]
Questions closed                                 65        318 [Space.SE]
Questions closed                                145        142 [Sports.SE]
Questions closed                                 55          2 [Poker.SE]
Questions closed                                200          3 [Parenting.SE]
Questions closed                                133          3 [Russian.SE]
Questions closed                              1,927         10 [Android.SE]
Questions closed                                 91          0 [Gardening.SE]

(Quoted from the relevant "2019: a year in moderation" meta posts.)

This is not an issue only relevant to a single site.

10
  • 5
    What does the site's meta regulars say (assuming there are some)? I.e. do people that use the site think it's a problem? Jan 1, 2020 at 11:26
  • 7
    How many users with close vote privilege are there on those sites and are they actively closing questions? It is worth considering how a mod "finds" that question. If they handle the close vote queue at least one community member flagged / close voted the question. I think this matters in judging how much of a problem this is. You could run stats with a SEDE query I wrote earlier: meta.stackexchange.com/a/323259/158100
    – rene
    Jan 1, 2020 at 11:37
  • 3
    I've seen this on more than one site. People on those sites don't raise it on meta, but at the same time, they don't raise much on meta at all. I don't think they're aware that this is extraordinary behavior. It's hard to tell how many users are actively closing questions because of diamond-hammering. (I feel it's also likely users are not closing questions anticipating the diamond hammer.) Jan 1, 2020 at 11:40
  • 2
    I have never been to Android.SE ... but if it is in any way "as worse" as the android tag on stack overflow, then I do see why you would need a gardener using an iron brush going through it all day long.
    – GhostCat
    Jan 1, 2020 at 14:24
  • While we are busy not pointing fingers or mentioning names let me say that at least one of the moderators on one of the sites mentioned above is an all around great caring person, I'm also not going to lay it on any thicker than that. --- It is only borderline stuff that should be unhammered, if it's within arm's reach the community should do their own work. --- If it's clear it should go it's great that the moderators keep a clean site. --- While this is a good question it suffers from numbers without pointing at the worst cases; places with lower hammers running closer to the line.
    – Rob
    Jan 1, 2020 at 14:38
  • 1
    How many of those questions had some close votes by the community even if the community didn’t cast the final vote that closed it? Moderators on smaller sites may need to be more active in the review queues, but that doesn’t necessarily mean they are acting completely unilaterally. Maybe those questions would have eventually been closed by the community.
    – ColleenV
    Jan 3, 2020 at 17:25
  • 3
    FWIW: only unilateral closures (no votes from non-moderators) are counted in the "moderators" column on these stats, @ColleenV. They may've been flagged for closure of course (just as moderator deletions are usually/often preceded by flags).
    – Shog9
    Jan 3, 2020 at 18:48
  • @Shog9 So by extension that would mean the community statistic would include all other closures, including the case where a moderator casts the final vote with prior close votes already on it? Jan 4, 2020 at 18:55
  • 2
    Right, @rubiks. Mods == one mod vote to close. Comm == 1 to n non-mod votes. A thorough analysis of how closing is done on a site must usually involve other data.
    – Shog9
    Jan 4, 2020 at 19:01
  • Here's a SEDE query to find (non-deleted) posts that were unilaterally closed. (I was surprised by the numbers on the site I mod too!)
    – Em C
    Jan 4, 2020 at 19:11

5 Answers 5

19

It really depends.

Ideally, the community should be closing question on their own, without needing a moderator to close everything. However... in the early days of a site, when there isn't quite as much of a userbase as you'd like, and there aren't very many users with the ability to cast close votes, it does fall on the moderators a bit to keep the site clean.

Remember, not everybody with the ability to cast a close vote bothers or cares enough to cast them. It's only a small percentage of the people with a certain privilege who wind up using them. On a small site, where the number of people with the ability is small enough in the first place, the portion of users who actually use them can be too small to actually be effective.

Personally, in the early days of Literature.SE, I would wind up closing... a large percentage of the posts that needed to be closed, either single-handedly or with a couple votes beforehand. This was because there weren't really that many people who could or were cast close votes for a while. (Remember, after the initial surge of activity when a site starts off, participation drops like a rock after private beta and it takes years for it climb up again.)

After a while, though, there were enough other people who could and were casting close votes that I often didn't need to hammer questions closed. I started leaving them and letting them sit a bit, in order to let the now-present community close it themselves - both to see if they would, indeed, actually close it, and to prevent a situation where they were simply depending on the moderators to close every question and not know or care how to close vote stuff themselves.

So... what do I make of that?

It really depends on how old and how large the site is, and how many users there are who can close vote, and how many closable questions are coming in a time... there isn't really a way, as far as I know, to determine the optimal moderators-closing vs community-closing ratio. You just have to let the site figure it out on its own.

As long as things are in fact getting closed, then there is no real problem. While it may be a good idea to teach the community to close and not to depend solely on the mods, it all depends on the age and size of the site in question.

1
  • 2
    This answer is spot-on. The one thing I might add is that, because it depends on a lot of variables, OP is probably going to have a much more productive discussion about any individual site's moderation on that site where people have the context to answer it fully. Jan 1, 2020 at 17:29
9

Should anything be done?

Nobody outside of that community should make that call easily.

Let's contemplate for a second on that most interesting of your examples, the one with 1,927 moderator and only 10 community closed questions.

Those two numbers tell us:

  • Overall, there is an extremely high number of questions that get closed
  • But there is an extremely low number of questions that the community "itself" closes

Now, there are only two possibilities about dealing with the first point:

  1. That is fine with the corresponding community.
  2. The corresponding community refuses the way some moderator(s) do business.

From there:

  1. That community is fine with what happens! And as long as the corresponding moderation team isn't in violation of "SE Inc." rules, what business do outsiders have to tell them to change course?!
  2. Then that community needs to come together in their meta, and discuss that problem, and find solutions. Only when that isn't possible, then it might be necessary to have SE Inc. look into it, or have MSE provide guidance.

Finally: when you have thousands of "close-worthy" questions in a community, but only a handful of users willing to do that moderation work, then well: be glad that there are moderators who do all that work. Chances are: without their efforts, trash would start piling up quickly.

Of course: that puts even more responsibility on the moderation team. But what is better in some remote wild west town? Having no sheriff, or having a sheriff that really takes care of everything?!

7

On a small site where I am active (Spanish.es) there has been some discussion about whether we should ask for the number of votes to be reduced from 5 to 3 precisely because (a) we only have one moderator (b) there are few of us who monitor the review queue. There are only five users with more reviews to close than the current moderator and of those one has left the site and some were previously moderators.

Here is the discussion on our meta Can we get the close and reopen voting thresholds lowered to 3? most of which is in English. Although there is disagreement most of those who contribute to moderating the site feel we should not leave it to our diamond moderator.

6

Irrespective of site size, I think the best time for anyone (including elected and appointed moderators) with the privilege to cast close and reopen votes to cast that vote is when they first see questions that warrant them.

This is because the longer a question that needs improvement remains open, the more likely it is to receive unfocused answers or answers focused on a question that was thought rather than actually asked. It is also because the longer a close question remains closed after it has been improved, the more likely it is to miss a quick answer that is focused on the question asked.

When non-moderators vote to close and re-open there is a lag awaiting 3-5 (depending on site) votes to be received, and in that lag, poor answers can arrive, which means an unfocused Q&A may result. That unfocused Q&A will need later clean up, or remain as a "broken window" on the site.

When moderators vote to close and re-open that lag is short-circuited, and more focused Q&As result. This may not always be or appear to be to the immediate benefit of the question asker but the more clear questions with clear answers that are on every site is to the great benefit of the many viewers looking for an instant answer to the same question later.

If anyone sees a question open or closed when they think it should be the opposite, and is unable to effect a change of its state via close/re-open voting, then the fallback is to ask a question on the Meta of that site so that they will either get it re-opened/closed or gain a better understanding of why its state is open/closed.

5

It can definitely be a problem.

I have been a member of one such Stack Exchange site and it has been a big problem. This is past tense because it did not get better and I left that specific SE site for that exact reasons. It may have improved, I don't know.

What happens is that since the moderators do everything the users should moderate, the users tools for moderation (downvotes as the prime example) are dormant and people don't see the actual need for them as tools of moderation and they degenerate into a tool for expressing Facebook-like approval or disapproval of good, but conflicting advice, instead of cleaning up the SE site. You can write the second-best, well-thought out and well-presented answer and be downvoted for it.

Regulars like myself have brought it to the site's meta(1), but nothing changed. With a reasoning of "we are right because we are the mods", and mods being on their job "for life" there is really little users can do other than wave that Stack Exchange site goodbye and find better uses for their time.

Now it does not have to be a problem. It might work. It works for some people that actually want that kind of moderation and that kind of forum. I don't, it's not what SE stands for, and it's why I'm on SE instead of a forum, but if enough people like it, I guess... good for them? We should just put some kind of warning sign, to not expect SE standards on all SE sites.


(1) Upon revisiting, I saw that moderators have changed. So don't take into account who has or hasn't a diamond now in that thread.

0

You must log in to answer this question.

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