-60

A moderator has various bots that capture a bunch of comments, and these comments land up in various rooms that are in plain view of the community. I think this tactic is creepy surveillance and violates the CoC; it's not nice to keep everyone's comments on record, and then let other people (especially non-moderators) in chatrooms bring up the comments again and again for discussion, causing more and more unnecessary and unwanted drama. This is particularly important when some users may post comments that they subsequently wish to delete shortly after. Some things in the comments are said in the moment, some things are said but meant to be transient and not meant to be logged permanently in plain view of the community.

If moderators with diamonds and the higher ups, e.g. Stack Exchange staff members already have access to pretty much every piece of information that is around the network, the least they could do is to keep deleted comments away from public view -- they are deleted by users for probably good reasons.

Do you think these bots are problematic?

Can we put an end to that practice or at least fix the way in which this creepy surveillance is being implemented?

  • 10
    The Code of Conduct does not contradict the Moderation. The bot is for catching up too chatty, irrelevant and useless comments for the site. If one posts them, they need to deleted. – Nog Shine Aug 9 '18 at 3:15
  • @NogShine seem my edit and my emphasis on enabling other people, non-moderators, to stir things up, when deleted comments should really be away from public view. – user393596 Aug 9 '18 at 3:20
  • 41
    "This is particularly important when some users may post comments that they subsequently wish to delete shortly after.", why do they even want to do that? If you want to post a comment which you also want to immediately delete, then please don't post it in the first place. – Bhargav Rao Aug 9 '18 at 3:24
  • 1
    If you believe there is a violation, you could ... you know, raise your concerns with SE using the "Contact Us" link given on the CoC page. meta.stackexchange.com/conduct – Masked Man Aug 9 '18 at 7:07
  • 3
    FYI, most spam and R/A posts deleted from SE since a few years ago are publicly visible to anyone. This includes random drugs, fake tech supports, garbages, trolls and a lot more. Does that violate the CoC? I don't think so. – ɪBᴜɢ Aug 9 '18 at 9:27
41

There's a long and very very honourable tradition of folks automating slightly more annoying moderation tasks, and to hasten handling of potential issues. By its nature, IPS has a tendency to attract chatty comments, and tools like this make life easier for moderators. In a sense, these are improved tooling for mods and engaged users, built by them, to help them clean up and maintain sites in good order, made by the people who use the site. Sometimes, it's a nice way to get a site-specific tool, or even something that can be developed quicker outside SE by the community (like SmokeDetector). They're useful tools.

It's worth remembering that in the context of IPS - Mithrandir isn't a mod. But that's fine. They didn't write the bots. (Which is also fine.) They don't actually own the bots (though apparently they can give them a kick if needed), and they help the community clean up comments as needed. So erm... ya, most of that bit of your question seems to be incorrect.

These comments that I see posted by the bot are a subset of all comments, anyway.

It's worth remembering...

All of these comments are public. This is roughly about as creepy as sitting at a café commenting on the cars passing by.

Once the comments are handled, they're cleaned up from the room when a moderator or RO has the time, inclination, or boredom, so there's less worry of folks stirring things up.

Folks who want to stir things up will probably find some way to do so anyway.

27

They do not violate the code of conduct

Seriously. Read through it. There's no part of it saying they aren't allowed. There isn't in the ToS either.

If you say keeping comments around is not nice, then by definition, the site isn't nice. Deleted posts, comments, and chat messages are still around. Only 10k users and mods can see deleted posts, only mods can see deleted comments, and ROs and mods can see deleted chat messages. They're still technically open, as anyone who can see it could theoretically share it with someone else who can't.

Also, the comments are public until deleted. Anyone could see them before that happens. As far as I know, moderators can see deleted comments.

A moderator has various bots that capture a bunch of comments, and these comments land up in various rooms that are in plain view of the community. I think this tactic is creepy surveillance and violates the CoC; it's not nice to keep everyone's comments on record, and then let other people (especially non-moderators) in chatrooms bring up the comments again and again for discussion, causing more and more unnecessary and unwanted drama.

There are tons of moderation bots. There are fewer for comments, and some delete the comments, others don't. It's not surveillance though. If you post something on the site, it doesn't take long before web crawlers find them. Some even go into web archives.

As for bots, I personally don't see a problem with them. They aid in moderation to keep unfriendly/chatty/otherwise unwanted comments off the site. For an instance, Queen (SOBotics bot) does this. Interpersonal.SE has a bot too, which also is designed to help with moderation. And it's also worth remembering that everything you post on SE (questions, answers, comments, chat messages), are licensed under CC by-sa 3, which means anyone can do anything with it, as long as there's attribution.

To be honest, Journeyman Geek sums it up pretty good in their answer:

All of these comments are public. This is roughly about as creepy as sitting at a café commenting on the cars passing by.

Also, it's not just mods creating bots. There are regular users who create them as well. I'm guessing SE has some private bots as well.


This is particularly important when some users may post comments that they subsequently wish to delete shortly after. Some things in the comments are said in the moment, some things are said but meant to be transient and not meant to be logged permanently in plain view of the community.

If a user wants to keep something private, they shouldn't post it in the first place. Posting something, and deleting it, to keep something hidden is usually not allowed. The exception is for accidental posting PII. But if someone posts a rude comment, and then deletes it to not get caught, that's technically abuse of the system. The same applies in chat. But it doesn't necessarily have to be rude, it could be something they want to keep private for other reasons. In which case, it's still not a good idea posting it. The API catches it up, moderators can still see it, and users can read it before it's deleted.

I do agree, there are things you might wanna remove (i.e. accidentally posted PII. Happens sometimes, when i.e. an asker forgets to remove an API key or hide a username and password, or IP, or something else), but most bots don't pick those up.

If moderators with diamonds and the higher ups, e.g. Stack Exchange staff members already have access to pretty much every piece of information that is around the network, the least they could do is to keep deleted comments away from public view -- they are deleted by users for probably good reasons.

They're still accessible through the API before deletion. Even if there is a good reason for deletion, why should it be hidden? Anyone could technically keep a full store of all the comments on SE, and post them. It wouldn't be a problem as long as it's attributed (and doesn't otherwise violate the license).

Do you think these bots are problematic?

No! In fact, they're the opposite. Bot assisted moderation has helped a ton, especially with comment moderation. There aren't a lot of good tools for finding them automatically (like search). Charcoal has a tool that creates a real-time feed of comments. SOBotics has a bot too. Some sites, such as Interpersonal.SE, have their own bots. They're all created for use on different sites, but they still help with comment moderation.

Can we put an end to that practice or at least fix the way in which this creepy surveillance is being implemented?

This isn't surveillance. It simply creates a copy of a comment somewhere else. If you worry about comments sticking around somewhere, you shouldn't post it in the first place.

Also, it's highly likely your comment(s) have been picked up by search engine caches already. These are also available to everyone.

22

Based on your description, you're combining two concerns that aren't really dependent on each other.

  1. The bots posting comments in chat.
  2. Users bringing up comments repeatedly to cause unwanted drama.

The first is positive. The second is not - but the second doesn't rely on the first. Users can post comments in chat manually and then continue to talk about them long after they've been removed.


There is absolutely a benefit to this bot - it helps remove unwanted content from the site more quickly than would otherwise be possible. We're already struggling with vast amounts of comment content that doesn't need to exist on our sites, either because it's chatty, attempts to answer questions in comments, or is unfriendly or abusive.

Because of how comments function, they gain high visibility to later visitors to the site but do not get much attention for oversight. There have been many requests over the years to find ways to lock comments on a post or require local reputation for commenting, particularly in the case of protected questions. I've even gone in search of ways to be notified of new comments on inactive posts.

Particularly in relation to the site I think you're talking about, Interpersonal Skills, this is a huge issue and one that would be much more difficult to address without having a bot like this - or having the comment system overhauled in a huge way. The latter would be amazing but, for the time being, having users willing to put in the time and effort to track down and flag these comments is a boon and something we should appreciate.


If the latter is happening, though, that's not what we want but - as I mentioned at the top - it's tangentially related to the bot. That behavior is what needs to be addressed, when it exists. If you see it, flag it. Use custom flags to explain why you're flagging the chat messages and be clear and explain it fully so that whoever responds to the flags can know to look into the problem more closely.

The solution may be removing the chat messages, asking the discussion topic to change and encouraging users to avoid such discussions in the future, or - in the case of extreme situations where behavior is very unacceptable - issuing chat suspensions.


That said, while we assume good intentions and allow users broad latitude to be forgiven and continue using the network after making errors we also expect users to learn from those errors and work to avoid repeating them. So, when someone is posting comments that they know shouldn't be posted - and then deletes them - if they do it regularly enough to come to the attention of moderators, they can expect to be warned or suspended for that behavior and, if it continues, they can expect those penalties to increase over time.

Even short-term comments reach their targets in many cases and do harm. "In the moment" doesn't make saying them acceptable. While it's appreciated when a user recognizes that and self corrects, we'd prefer that they take time and think before posting such content and if a warning or suspension is necessary to help them slow down, then that's the step we recommend.

Poor behavior on one side doesn't excuse poor behavior on the other by any means but the solution here isn't to ban the bot, it's to address the behaviors on both sides.

You must log in to answer this question.