Show HN: Reveddit.com: Improving online discourse with transparent moderation

Yeah, I think that line of thinking really only works well for outright shadowbans, not singular shadow removal. Single removals might have been justified, and the time people spend on further comments might not actually be wasted (because they might not be removed).

That said-- getting rid of outright shadowbans would be a useful step forward and establish a norm against shadow moderation more generally.

Re: government-- I did say civil cause of action for a reason: The argument there would go "you intentionally deceived me by making me think my comments were visible, when they never would be-- and in doing so caused me to waste a lot of resources, I'm entitled to recover from my losses". No new law would be required. And the moderation being correct or not isn't the concern-- just the fundamental dishonesty of shadow banning. (though, ironically, outright spammers might have an easier time making the legal argument because they'd be more able to show monetary damages).

But as you note, the single comment removals are more common and more insidious, so perhaps its just a distraction.

I am tremendously grateful for reveddit, and I think it's made a contribution to better the world. But I'm also not sure how far we can get with awareness. Even after all this time the whole subject can't even come up without people constantly repeating the false claim that subreddit mods can't shadow ban users (or their comments)!

But don't see anything better to do right now-- so education it is!

I know the opacity is ultimately reddit's fault (and I've complained to them about it), but reddit these days seems almost completely unresponsive to small scale public feedback.

Thanks for the citations.

(from your comments)

> Since I regard secrecy from authors as the bigger problem,

I guess part of the problem with the admin removals is that while the removal itself may not be secret (though it's often hidden, e.g. requiring a direct link to a removed thread), the rational and basis for it always is. It increasingly appears to be automated based on user reports and if any human is reading the text at all in many cases they must not be native english speakers. (I've noticed that a lot of the more obviously bad admin removals seem to happen in deep in the night, pacific time).

To the extent that user-community moderation is often bad and admin moderation grows in frequency in response I think we can only expect admin moderation to become more incompetent and kafkaesque: As you note, it fundamentally can't scale but they can (and I think have) respond to the lack of scalability by just doing a worse job of it rather than not taking actions.

But regardless, I understand your focus better now. If you ever do come across a way to give posters access to their own admin removed comments without creating risk or distraction for your focus on shadow moderation, I know it would be appreciated by many.

Thank you for going through it! Yours is the first comment I've seen mentioning the talk. You've provided some great feedback here that I will incorporate into future versions of any related presentations.

> However, I am not sure what to do with this information except perhaps sharing it with some people I know own Reddit (as per your FAQ),

Right, I think we need to do two things. (1) build these systems for other platforms, and (2) raise awareness about what's going on by telling stories about it.

I've got some idea about how to do both of these, but I don't own the patent or anything. So to the extent that someone's motivated to collaborate, I'm willing. This is very grass roots. To get to the point where I could so openly collect my thoughts has taken some time, however now I've got a good idea about how to go about it. In fact I've always felt that way about this work! Every couple of weeks I'm like, oh, I should be doing that. Then I go do it and I reach a new milestone, but real public awareness remains elusive.

I've tried reaching out to hundreds of journalists and several organizations who support free speech and so far haven't had success, so if you have any connections and want to fire off an email or tweet about shadow moderation, Reveddit, or the talk, please do your worst.

> but then that makes me think "then what"? What happens then? Getting Reveddit exposure is good, but there does not seem to be a clear stated strategy of how any of that traction will be practically used.

Well, I can say with confidence that when the site is shared on Reddit, groups do change. Users create new groups, moderators become more open, or admins may even step in. So my expectation is that sharing it on other social media sites might lead to more discussion and possibly more openness from platforms, because again this is happening everywhere on every side of every issue in every geography.

For example, I have virtually no Twitter following and don't know how to do Twitter, especially since I'm usually awake while you all are sleeping.

Does that provide some ideas for direction? I know it's not all buttoned up like a campaign from FightForTheFuture. I'd love for it to be like that, but right now it's still just me, and maybe you all. I don't have good ideas for what to do with money so I am not asking for donations.

Another to-do is to start doing video shorts on stuff that gets removed, like Cheyenne did here [1] [2].

> Also, on slide #60 of your presentation you have a slide title of "There is no evidence that censorship doesn't change minds." I don't know if I am just having a slow brain day, but I was not sure how to interpret this in the context of what you were saying and the bullet-points on the slide itself. Is this just meant to be a more non-committal version of "Censorship changes minds"?

Yes, that's a much better way of putting it, or maybe "Myth: Censorship might change minds" ? I think I was thinking of this conversation [3] where I said "countering negative ideas with suppression does not work", and someone else said they don't believe it. I've seen the same argument made elsewhere too.

Since this talk, I have developed a stronger opinion on that. I now believe censorship creates the toxic environment that mods seek to eliminate [4].

[1] https://www.tiktok.com/@cheyenne.l.hunt/video/71115053899502...

[2] https://www.tiktok.com/@cheyenne.l.hunt/video/71118581026385...

[3] https://news.ycombinator.com/item?id=33055132

[4] https://old.reddit.com/r/TheoryOfReddit/comments/ymeqaz/hate...

Then let's make it big! Can do #ShadowModeration on Twitter or ... ?? Once you figure out how to share this issue, it's an easy victory. Nobody wants their content to be shadow removed, so there's no justifiable reason for the practice. Platforms will say it helps with bot spam, but that's bologna because coders who make bots can code to detect shadow bans and removals. So it's clearly designed to manipulate real people, including moderators who use it and people who work at these social media companies. They think they're creating environments free from toxicity, but what they're really doing is creating toxicity by making it more and more difficult for people to reach across ideological spectrums.

> I guess part of the problem with the admin removals is that while the removal itself may not be secret (though it's often hidden, e.g. requiring a direct link to a removed thread), the rational and basis for it always is.

I completely agree. It's just something that's out of focus for me right now, and as I wrote, I don't want to compromise what I have so far. Undoubtedly I lose some traffic over it, and I just accept that.

> It increasingly appears to be automated based on user reports and if any human is reading the text at all in many cases they must not be native english speakers. (I've noticed that a lot of the more obviously bad admin removals seem to happen in deep in the night, pacific time).

That's interesting. I'm pretty sure the Philippines is a pretty big hub for moderation because they are low cost and know the language quite well. Plus that was reported in 2014 [1].

> To the extent that user-community moderation is often bad and admin moderation grows in frequency in response I think we can only expect admin moderation to become more incompetent and kafkaesque: As you note, it fundamentally can't scale but they can (and I think have) respond to the lack of scalability by just doing a worse job of it rather than not taking actions.

Yeah, that's effectively what is going on. They are building tools that remove mass amounts of comments in different ways [2] and now allow users to cut each other out of conversations too with the new user block described in the same link. Just today I came across a post from some 3rd party building a tool to identify hateful groups of users to be shared with subreddits [3]. And the website is all done-up too as if that's an idea that would be popular. It's crazy. I replied on that thread and the author of the tool just completely sidesteps the fact that any actions taken against users on his list will be occurring in secret. Like, that is not his concern because "that tool is just made for spam" which is just such a weak argument.

There is absolutely no concern for how much gets removed as long as users don't know about it. The good thing is the problem then becomes easier to point out. It's a race to do that before more harm is done.

> But regardless, I understand your focus better now. If you ever do come across a way to give posters access to their own admin removed comments without creating risk or distraction for your focus on shadow moderation, I know it would be appreciated by many.

Noted.

[1] https://www.wired.com/2014/10/content-moderation/

[2] https://www.reddit.com/r/reveddit/comments/sxpk15/fyi_my_tho...

[3] https://old.reddit.com/r/TheoryOfReddit/comments/ymeqaz/hate...

That's deep. I hadn't thought about what happens to the ego of users whose content is approved. One lie in the chain really does screw up everything.

> b) the 25%, seeing the vigorous debate about the cost of the King's new crown (and not the debate about the King or the crown themselves), don't understand why having the King as a mod could have a chilling effect on speech.

You're right, it's like a newly manufactured Overton window that can be adjusted on a per-group basis. A simpler term is brainwashing but I think you do need to spell it out as you do here. This is good stuff, I can use these principles to tell stories using real examples and make what's happening more relatable. Or you or anyone else could too.