Camestros Felapton

FELAPTONL

Today’s Infographic: moderating comments

Cat Rambo has an unsurprisingly insightful essay on the Baen’s Bar mess:

“I am so tired of this argument, which so often gets used by people who have, indeed, fought the good fight but somewhere along the line also acquired the idea that only people who’ve gotten punched in the face for speaking get to talk. That’s what underlies someone talking about “swooning” or “pearl-clutching” and don’t even get me started on some of the gender stuff that gets draped onto that rhetoric like a six year old putting tinsel on a Christmas tree.”

http://www.kittywumpus.net/blog/2021/02/18/opinion-on-baen-books-moderating-discussion-boards-political-expression/

Well worth a read but it also made me think of showing pictorially what was wrong with some of what Eric Flint said. I keep circling back to the kinds of stuff actual white supremacist murderers have posted on line, like the Christchurch murderer making jokes about the Spyro the Dragon video game in a manifesto calling for a race war. Flint is literally deluded about there being a common sense way of spotting real threats from jokes.

The delusion is that there’s a nice clear distinction between the two (or at worst a small overlap) so that there’s a nice clear point or zone where a moderator can step in and outlaw the truly bad stuff from the innocent jokes or just-arguing-hypotheticals etc. It is an utterly wrong headed perspective that is literally helping violent people organise online.

I thought this illustration might help. The ‘moderate here’ line is intended to be the point where you can be pretty confident that all the bad stuff gets moderated. I call the bottom one ‘reality’ but even that is an idealised situation and doesn’t reflect how broad genuine threats can be.

camestrosfelapton

39 responses to “Today’s Infographic: moderating comments”

  1. It’s like the plot to kidnap the governor of Michigan. When I looked at some of their early conversations, I would have assumed it was just some right wingers blowing smoke and making plans they really had no intention of carrying out. However, it was obviously advanced enough that the FBI got involved. On the other side of the spectrum, comedian John Mulaney says he was investigated by the Secret Service for a joke he made about Julius Caesar and the POTUS in his monologue on Saturday Night Live.

    We live in interesting times. Just because you think you know that the people talking are just joking or talking out of their hats doesn’t mean that everyone will take it that way and doesn’t mean that they are.

    Liked by 4 people

  2. I thought of something similar to this when Weisskopf, in her statement on the bar, says
    “We do not endorse the publication of unlawful speech.
    […]
    But we will not commit censorship of lawful speech.”

    The implicit message is that she wants the moderators to thread the exact same line as courts. And that is certainly going to be challenging – in particular when they have already attracted a clientele that clearly wants to toe the line.

    Liked by 1 person

  3. Ask Kathy Griffin about jokes and their consequences.
    At least when the joke is about a powerful old rich white man who happens to have the Secret Service watching his back.
    Strange how situations like hers don’t happen to non-powerful, non-old, non-rich, non-white, non-men, isn’t it?

    Liked by 6 people

  4. The way I’ve usually seen this sort of thing done successfully is pretty much:

    – Have a clear comments policy, but don’t be afraid to treat it as guidelines with some wiggle room, because certain people will attempt to see how far they can go over the line before they get reined in. Be sure to be up front that some things just won’t be tolerated.
    – Everybody’s first comment gets automatically placed into a special queue that can only be released actively by a moderator, and no other comments will be posted until after that user’s first comment is released. This filters out not only most spam but also the people who can’t even pretend to be civil for long enough to post a single comment.
    – Certain things (URLs, language) can also result in auto-moderation.
    – Actively moderate, and be clear about what you are kicking off.
    – Encourage an active community and make sure they know where to send notifications so that if somebody who got past all the above is still being a horrible person it can be acted upon.
    – Allow a certain amount of leeway even towards the ‘horrible person’ side for multiple reasons: sometimes they may actually have points that should be dealt with; being too strict or arbitrary may make people you want feel less welcome; and sometimes the community likes having ‘chew toys’.

    Of course, none of this works if your moderators are actively horrible people already, which is documented in the case of Baen’s Bar. It also doesn’t work if you explicitly make ‘free speech’ part of your moderation policy, because at that point the people who yell the loudest will just shout everybody else down, and there’s no way to moderate without looking arbitrary. (See also: Gab, Parler.)

    There are issues on how well this scales up, of course, but the idea is that each ‘room’ should have at least one moderator that sets the tone.

    (PZ Myers at Pharyngula was mentioned in one of the Debarkle Vox Day thread, and he’s a decent example of doing this reasonably well despite controversial subjects. Many of which shouldn’t be controversial in any sane society.)

    Liked by 2 people

    • In the old days, the commentariat at Pharyngula was “fierce”, with PZ’s approval. On the flip side, they didn’t stalk people around the web, unlike some other groups. (And they did call out PZ when he mistook satire for reality, and posted it as an example of “Abrahamists behaving badly”.)

      Like

      • That’s part of why I said ‘decent’ example rather than ‘good’ example.

        But yes, the commentariat is a little more sedate than they used to be as a whole, and a couple things that used to be fairly regular got called out and are now off the list of acceptable things to say.

        Which is another requirement for this sort of thing: the ability to change stances as the surrounding context changes. Accepting that you might be wrong is a harder lesson for some folks than others.

        Liked by 1 person

    • Also, if you resort to algorithms to help ease the abuse queue’s load, for Bog’s sake make them robust enough to distinguish bad faith mass attacks.
      @Jack, I’m looking at you…

      Liked by 1 person

    • The approach Jenora describes is almost exactly how my last moderation gig operated. They even had an antispam bot which graded messages according to a certain algorithm and referred any that crossed the line to the moderation queue for human inspection before the message would become visible. It worked pretty well, although it was a bit frustrating when the profanity filter picked up perfectly innocent words that just happened to contain a forbidden sequence of letters.

      One tool we had which she does not mention was a moderators’ forum that consisted of two major discussion areas. One was for general discussion regarding moderation – for instance, if one of us wanted to suggest a new policy or just say that we would be offline for a while and needed the rest of the crew to keep an eye on “their” forum(s), that’s where it would happen.

      The other, as I recall (it’s been a few years), consisted of two threads. One was a record of banned users, including descriptions of the circumstances that caused their bans. (Naturally, this did not extend to spambots.) The other tracked “problem children,” those who made a habit – whether accidentally or deliberately – of skating close to that fuzzy line between “edgy” and “actionable” discourse. Each such user had a post – sometimes even two – dedicated to their infractions, and mods were encouraged to review a user’s record there (if any) when assigning punishment for a new infraction, as well as create a new message if needed.

      Towards the end of my tenure, they formalized and published a point-based set of guidelines, involving both temporary and permanent points, in the interests of letting people know what to expect of they crossed the line.

      This worked pretty well, but even then, I could see that certain people got preferential status. This usually meant “people that frequently freelance for or are employed by us,” and it was never really formalized… but it was still there. “Oh, Bill writes two books a year for us. Yeah, he’s an ass, but go easy.”

      And even with all those measures in place, they STILL ended up removing the moderation-light “off-topic” zone before my departure. There was a big discussion about that, weighing the headaches it caused against the sometimes interesting discussions it generated, but ultimately we had to admit that most of the moderation took place there, even given its “wild west” nature. It was just so much easier to kill it, endure a period of outrage, and let the place settle into a new normal.

      Given that even Baen’s defenders talk about the Politics area being a place best avoided, perhaps they should consider a similar solution. Amputate the finger to save the body. People will complain, but that goes away after a while.

      Liked by 1 person

      • Just to make it clear, this is me. I signed in with one account and then got switched to the other when I decided to Like a comment. No deception is/was intended.

        Liked by 1 person

      • They can’t just kill of the Politics section, because if they do that, they will have to kill off Tank Marmot’s forum too; you can’t say “the Politics section is far too much work to moderate due to the amount of line-crossing going on” while at the same time providing a safe space in the forum of the worst inciter on the boards.

        Liked by 1 person

      • The other, as I recall (it’s been a few years), consisted of two threads. One was a record of banned users, including descriptions of the circumstances that caused their bans.

        Pharyngula used to have one of those. PZ Myers eventually dropped it because, during the heights of ‘Elevatorgate’ and ‘The Slymepit’, there were a large number of people who would go in, troll the board, get banned, and then go off and point to their public position on the banned list as a badge of pride and proof of ‘owning the libs’.

        Which, of course, takes us back to being willing to change when the current setup is showing flaws, something you also touched on. There are, sadly, always going to be people who abuse any apparent ‘weakness’ in the system to get what they want. There are also going to be people who declare victory no matter what actually happened

        Liked by 2 people

    • I’m one of the chew toy people – I get banned from left and right blogs (I’m banned on scifiwright and pharyngula) because I like to argue. I’m a very strong believer in free speech, but that’s for individuals. A blog or a forum is ultimately the work of whoever is running it, and if it’s to stay coherent and on topic, it has to be carefully moderated. The Volokh Conspiracy hosted at Reason allows all comments, and as a result the comments are mostly the same half-dozen people shouting at each other, including someone making outright Nazi anti-Semitic posts.

      On the other hand, unless you want the forum to be a bunch of like-minded people congratulating each other on their perspicacity (and there are plenty who do), moderation has to allow strong disagreement, assuming the topic isn’t something purely factual. Scalzi’s Whatever blog tends to overdo the moderation, at least on me. I feel like he’s trying to use aversion therapy to train me to say only things acceptable to him by allowing through some of my posts and not others. (It hasn’t worked so far.)

      Like

      • No offence intended Hyman but you are a bit trollish. I don’t mind a degree of contrariness as it keeps people chatting but you seem to spend a lot of time focused on what will annoy poeple .

        Liked by 4 people

      • Dude, you’re a self-admitted troll. That is why you’re “banned from left and right blogs”.

        You like to argue, but the vast majority of the time, you do it dishonestly. There’s a difference between honest disagreement and dishonest argument, and either you don’t understand that difference, or you enjoy pretending that you don’t understand it.

        I’m sure that I’m far from the only person who has noticed that the e-mail notifications for your comments on File 770 are always timestamped many minutes, or even hours, after the actual timestamps on your comments… because they’ve been sitting in moderation.

        Mike is an unbelievably generous moderator. When you’re on permanent moderation there, it’s for a very good reason — and it’s not because of “a difference of opinion”.

        Liked by 4 people

      • I get banned from left and right blogs
        {snippage}
        I feel like he’s trying to use aversion therapy to train me to say only things acceptable to him by allowing through some of my posts and not others. (It hasn’t worked so far.)

        Oh my. If only there was some kind of process of receiving feedback and then allowing that feedback to influence future behaviour.
        Perhaps we’d call it ‘learning’?

        Liked by 1 person

      • Being curious, I checked Pharyngula’s archives and found the thread that got you bounced. Yeah, I can see that you making repeated normative claims about the lesser value of the lives of people with Down’s Syndrome would push PZ past the point where he wanted anything to do with you. Oh, and it looks like you later pushed him even further by sock-puppeting

        There were a few bannings that I thought were mistakes by PZ, but not this one.

        Scalzi’s Whatever blog tends to overdo the moderation, at least on me.

        Or maybe he’s being overly lenient in letting you comment there at all.

        I feel like he’s trying to use aversion therapy to train me to say only things acceptable to him by allowing through some of my posts and not others. (It hasn’t worked so far.)

        So at the cost of a little effort on his part, he makes sure that all visible comments by you are acceptable to him. Sounds like a win-win! At least until you piss him off.

        Like

  5. I started wondering what Timothy McVeigh’s blog comments would’ve looked like if he’d been extremely online in 1994. But then I realized that of course there was no such person as Timothy McVeigh, since, as Weisskopf says, military veterans are too wise about violence to commit any violent aggression.

    Liked by 3 people

  7. Grr. This always makes me angry: Weisskopf said, ““We do not endorse the publication of unlawful speech.
    […]
    But we will not commit censorship of lawful speech.”

    She implies in her wording (but doesn’t come out and state) that any effort to moderate speech in her forums would somehow not be lawful. But that’s not true at all. And this conflation is incredibly harmful today when so many extremists utilize the ‘Second Amendment solutions to First Amendment problems’ tactic to threaten and supress perfectly legal reactions to violent language.

    Liked by 1 person

    • Not quite. She’s saying that the law requires some content to be prohibited (which is true) but that she will not go further than that (which is false; the Bar already has banned discussion topics which would not violate any law). She just wants to have it both ways: a statement that makes her look like a Free Speech Warrior in public, plus the freedom to ban whatever she wants to in private.

      Liked by 2 people

      • I get what she’s saying. I was trying to indicate that I thought she was saying it in a way that can be taken advantage of by people who want to conflate a private organization’s censorship of speech with a violation of the First Amendment. I personally think she worded it that way on purpose as a signal or a nod to those people (some of whom seem to be active in the Politics thread), but I’m open to the idea that she didn’t intend that conflation.

        Like

  9. […] It cannot be mentioned often enough that the events of January 6, the ones Republicans and other conservatives are working so hard to downplay and erase, was not a case of a rowdy bunch busting up a Starbucks. It was an organized effort that destroyed and stole government property, in which people died and it would not have taken much more for a pre-planned section of that mob to use the chaos in order to kidnap and kill government officials. Camestros Felapton illustrates this in an infographic here. […]

    Like

  10. […] So what does matter? Rules need to be clear and enforceable. They need to admit some degree of nuance and context but they simply can’t be as flexible as the kind of more dynamic moral judgements we make. They also need to fail-safe and err on the side of caution. It is an issue that I discussed earlier in the year in this post: https://camestrosfelapton.wordpress.com/2021/02/19/todays-infographic-moderating-comments/ […]

    Like

Blog at WordPress.com.