3

Quora has bots that do a lot of moderation, such as: auto-question-tagging, marking questions as "needs improvement," auto-question-merging, auto-answer-collapsing, and even auto-question-editing (like grammar/spelling correction). If bots like these were implemented on Stack Exchange, would they make communities better, or are they unnecessary?

Improve this question
6
  • 2
    Required: xkcd.com/810
    – Catija
    Commented Aug 7, 2017 at 5:05
  • 1
    Can you give some examples? We have automatic low quality flags, for example - they just go into a queue for real people to agree with or dismiss. Question merging is extremely rare here and easy to mess up (and can't be fixed if it is). I'm pretty sure that we don't need a bot for that, unless you mean flagging duplicates... we don't merge questions, usually... we mark them as duplicates. We don't have a way to collapse answers, so I'm not sure what a bot would do in that case.
    – Catija
    Commented Aug 7, 2017 at 5:07
  • 1
    What answers other than "if you create one which is actually useful, the answer is yes" can you expect? I've been meaning to try out one which automatically checks shell scripts in questions against shellcheck.net and posts a pointer there if it finds any errors. Hard to tell whether it will be more helpful than annoying on balance but I'm sure it could be tweaked with thresholds etc to the point where the balance is definitely positive. Who will do that, though? Maybe me, one day ...
    – tripleee
    Commented Aug 7, 2017 at 5:54
  • Anyway, it's already happening. chat.stackoverflow.com/rooms/111347/sobotics
    – tripleee
    Commented Aug 7, 2017 at 6:00
  • 2
    See Can a machine be taught to flag spam automatically?.
    – Glorfindel Mod
    Commented Aug 7, 2017 at 6:11
  • 2
    Another example is Wikipedia with its dozens of bots, which I believe the answer to this question is: yes, it could be beneficial. However, I see different policies/formats that it might be hard to be applied on here, e.g. privilege based on reputation (only high rep can do more moderation tools), and editing bumps. I believe there are more reasons, but those examples are just the surface.
    – Meta Andrew T.
    Commented Aug 7, 2017 at 6:25

1 Answer 1

Reset to default
6

A lot of moderation is already done automatically; the fact that there aren't specific user-visible "bots" that do the automated moderation is besides the point...

There are already auto-flags raised for content quality, there are auto-flags raised for things like excessive numbers of comments, user activity (e.g. for mass deleting content) and probably a bunch of others I can't remember or don't know about. Questions are automatically protected if they are the target of low-quality answers, posts are automatically bumped if they meet certain conditions, tags are automatically removed if they are unused... and probably a lot more, those are just the first things that come to mind.

Automated processes that need to be associated with a user are attributed to the Community user, which essentially is a bot that does a lot of moderation (or at least the public facing attribution for the automated moderation that happens; I'm not sure about the internal mechanics).

There are also a lot of community created bots that help with moderation. See for example, SmokeDetector (which helps detect spam posts), Charcoal (which runs SmokeDetector) and SOBotics (A SO chat room for testing moderation bots).

So yes, bots can be useful and beneficial... and a lot already exist.

Some related reading:

Improve this answer

You must log in to answer this question.

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