96

It's New Year's Day in Stack Exchange land...

A distinguishing characteristic of these sites is how they are moderated:

We designed the Stack Exchange network engine to be mostly self-regulating, in that we amortize the overall moderation cost of the system across thousands of teeny-tiny slices of effort contributed by regular, everyday users.
-- A Theory of Moderation

While there certainly are Moderators here, a significant amount of the moderation is done by ordinary people, using the privileges they've earned by virtue of their contributions to the site. Each of you contributes a little bit of time and effort, and together you accomplish much.

As we enter a new year, let's pause and reflect, taking a moment to appreciate the work that we do here together. To that end, here is how the moderation done here on Stack Overflow breaks down by activity over the past 12 months:

                 Action                  Moderators Community¹
---------------------------------------- ---------- ----------
Users suspended²                              1,631      1,269
Users destroyed                               4,750          0
Users deleted                                 5,484          0
Users contacted                               5,689          0
User suspensions lifted early                    55          0
User review-bans lifted early                    68          0
User banned from review                         439      4,519
Tasks reviewed³: Triage queue                    13  1,329,907
Tasks reviewed³: Suggested Edit queue         1,229  1,177,058
Tasks reviewed³: Reopen Vote queue               20    200,214
Tasks reviewed³: Low Quality Posts queue         71    631,186
Tasks reviewed³: Late Answer queue               11    312,478
Tasks reviewed³: Helper queue                     2     40,210
Tasks reviewed³: First Post queue                27    835,061
Tasks reviewed³: Close Votes queue              611    440,336
Tags merged                                     127          0
Tag synonyms proposed                           351        176
Tag synonyms created                            413         64
Tag highlight language set                        7          0
Revisions redacted                              517          0
Questions unprotected                            11         67
Questions reopened                            1,068      8,860
Questions protected                             184      6,212
Questions migrated                              459        914
Questions merged                                 52          0
Questions flagged⁴                            1,453    592,454
Questions closed                             22,180    331,769
Question flags handled⁴                      38,092    555,815
Posts unlocked                                  143        676
Posts undeleted                               2,382     79,917
Posts locked                                    381      8,886
Posts deleted⁵                               92,577  1,370,172
Posts bumped                                      0     34,820
Escalations to the CM team                    1,578          0
Comments undeleted                            1,275          0
Comments flagged                              1,634    345,828
Comments deleted⁶                           303,700  1,006,020
Comment flags handled                       211,510    135,952
Bounties canceled                               224          0
Answers flagged                               3,504    462,254
Answer flags handled                        219,741    246,017
All comments on a post moved to chat          1,569          0

Footnotes

¹ "Community" here refers both to the membership of Stack Overflow without diamonds next to their names, and to the automated systems otherwise known as user #-1.

² The system will suspend users under three circumstances: when a user is recreated after being previously suspended, when a user is recreated after being destroyed for spam or abuse, and when a network-wide suspension is in effect on an account.

³ This counts every review that was submitted (not skipped) - so the 2 suggested edits reviews needed to approve an edit would count as 2, the goal being to indicate the frequency of moderation actions. This also applies to flags, etc.

⁴ Includes close flags (but not close or reopen votes).

⁵ This ignores numerous deletions that happen automatically in response to some other action.

⁶ This includes comments deleted by their own authors (which also account for some number of handled comment flags).

Wishing you all a happy new year...

21
  • 10
    Happy New Year Shoggles. You happy to feature this?
    – user3956566
    Commented Jan 1, 2019 at 3:16
  • 27
    "Users destroyed" Ah ha ha, you've made my year ;) Commented Jan 1, 2019 at 3:17
  • 9
    @LightnessRacesinOrbit ikr. There's an option to delete or destroy them. Mods tools are pretty dangerous ;)
    – user3956566
    Commented Jan 1, 2019 at 3:20
  • 4
    I'd like to see how many users were rate-limited as well as post-banned at any point or length within the year Commented Jan 1, 2019 at 3:23
  • @usr2564301 there's been a few inactive at times. We get break downs for some time periods of how many flags per mods were handled. The issue with that is some flags are quicker to handle than others.
    – user3956566
    Commented Jan 1, 2019 at 3:23
  • 3
    You can feature it if you want, @yvette; donno how many folks are interested, but it makes a nice reference.
    – Shog9
    Commented Jan 1, 2019 at 4:05
  • 19
    Note: "user deleted" means the name/id are removed; "user destroyed" means everything is removed (posts, votes, etc.)
    – Cœur
    Commented Jan 1, 2019 at 4:15
  • 2
    Does "users deleted" include users deleting their own accounts, or only moderator interventions? Commented Jan 1, 2019 at 7:45
  • 5
    Could we get the Community column divided into two columns: non-diamond users and the -1 user? Commented Jan 1, 2019 at 8:18
  • @SamuelLiew questions rate limit and post bans are not really related to moderation, so don't belong here in my opinion. Commented Jan 1, 2019 at 8:31
  • Would love to see the top users who're leading in their respective activities (deletion, review, closure, etc).
    – cs95
    Commented Jan 1, 2019 at 12:28
  • 8
    Better than YouTube rewind 2018
    – Spikatrix
    Commented Jan 1, 2019 at 14:09
  • Would love to some summary added for "aged away" flags. Commented Jan 1, 2019 at 14:57
  • 1
    @ThomasG delete/destroy both remove a user account and anonymise their contributions. The main difference is that "delete" preserves posts based on certain logic (did they have something with positive scores that could be worth keeping kind of thing). "destroy" however, removes everything that account did positively scored or not. Commented Jan 1, 2019 at 17:20
  • 2
    You usually wouldn't, @ThomasG (although there have been exceptions, mostly involving successful trolls / rings of spammers). The important feature of Destroy isn't so much that it removes positive posts as it is that it removes all posts - and comments. Consider a prolific spammer who manages to drop a half-dozen posts on the site, or a troll who earns 50 rep through careful edits and then starts blasting out vile comments on dozens of unrelated posts: it is imperative that these be removed promptly, and Destroy ensures this is done.
    – Shog9
    Commented Jan 2, 2019 at 16:25

2 Answers 2

36

I had mentioned something when these were posted:

I'm looking at the Answer flags handled 273,000 154,891 values. It looks like the moderators handle more of these than the community. Can we do something to get the balance correct there? (Increasing the number of reviews a user can do or decreasing the number of reviews needed to do away with the post).

And 2 years later,

Answer flags handled                        219,741    246,017

Which clearly shows that the 4 reviews needed to delete a post needed certainly worked. That said, one other number is worrying me, the number of reviews has decreased. The triage queue, for example, had 2,212,710 reviews, as opposed to 1,329,907 this year. Ideally they should all have increased with the number of posts. (or perhaps we had really good content this year, and thus less posts entering the review?). Given that the review link was changed to an icon from a text link sometime in 2017, 2018 was the first "full" year for it. Can we blame the new icon for the decrease in the number of reviews, or is there some other reason? Some statistics about how many posts entered the review queue would also be helpful

7
  • 8
    Also, in case anyone is interested, I just tabulated the 2016 and 2018 numbers side-by-side here gist.github.com/Bhargav-Rao/3eae32a19ce4185c4ef5f1bbddf30127. Commented Jan 1, 2019 at 10:57
  • 3
    All I concluded was that moderators are generally slacking in review tasks ... ;) .... don't you love statistics ... :D
    – rene
    Commented Jan 1, 2019 at 11:19
  • 5
    In all seriousness: I'm a bit disappointed that the awesome push you and other mods gave in getting the backlog of burninate requests down isn't reflected in any of these figures. You should get credits for that, in writing.
    – rene
    Commented Jan 1, 2019 at 11:21
  • 10
    Hey hey, moderators in total reviewed just around 100 less in triage as compared to 2016, whereas y'all reviewed a million less. ... and yes, we love numbers. ;p Commented Jan 1, 2019 at 11:23
  • 14
    Triage reviews down? I don't bother doing triage reviews because it's broken, with a whiff of it being cynically broken by design.
    – Raedwald
    Commented Jan 1, 2019 at 17:50
  • 5
    Questions entering Triage are down 40% from 2016 - that's gonna account for pretty much all of the difference here. Now... Part of that drop is simply that I tightened the restrictions on what can enter to compensate for the less visible top bar. The other part is simply that the heuristics we use are now very old and badly need to be updated - essentially we trained a model that no longer fits well.
    – Shog9
    Commented Jan 1, 2019 at 18:05
  • 55
    "or perhaps we had really good content this year", haha, ahahaha, HAHAHAHA! Commented Jan 1, 2019 at 20:42
2

Thanks for the data!

I noticed that it's almost impossible to compare between Stacks because it's absolute numbers. Could the number of active users/questions/answer/comments of 2018 somehow be figured in?

Is that a SEDE query that you could share so we could do that ourselves?

I think it would be useful to compare sites.

3
  • 1
    SEDE doesn't have deleted comments, nor deleted users, but postswithdeleted has all records you need to do questions and answer counts over 2018. What does "active users" mean?
    – rene
    Commented Jan 1, 2019 at 11:16
  • I guess it should be a user that acted that year. For example 10 users suspended might mean different things for sites with 11 users than for sites with 100.000 users.
    – nvoigt
    Commented Jan 1, 2019 at 11:18
  • 2
    something like this then: data.stackexchange.com/stackoverflow/query/957215 but SEDE was last updated last sunday, so we miss a good day.
    – rene
    Commented Jan 1, 2019 at 11:37

You must log in to answer this question.

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