[dimva]

His argument makes no sense. If this is indeed why they are banning people, why keep the reasoning a secret? Honestly, every ban should come with a public explanation from the network, in order to deter similar behavior. The way things are right now, it's unclear if, when, and for what reason someone will be banned. People get banned all the time with little explanation or explanations that make no sense or are inconsistent. There is no guidance from Twitter on what behavior or content or whatever will get you banned. Why is some rando who never worked at Twitter explaining why Twitter bans users?

And how does Yishan know why Twitter bans people? And why should we trust that he knows? As far as I can tell, bans are almost completely random because they are enacted by random low-wage contract workers in a foreign country with a weak grasp of English and a poor understanding of Twitter's content policy (if there even is one).

Unlike what Yishan claims, it doesn't seem to me like Twitter cares at all about how pleasant an experience using Twitter is, only that its users remain addicted to outrage and calling-out others, which is why most Twitter power-users refer to it as a "hellsite".

[roblabla]

From my understanding, he's not claiming this is how twitter currently works. He's offering advice about how to solve content moderation on twitter.

[dontknowwhyihn]

He’s offering advice that differs from what Reddit does in practice. They absolutely ban content rather than behavior. Try questioning “the science” and it doesn’t matter how considerate you are, you will be banned.

[CountHackulus]

He covers that further down in the tweets, near the end of the thread. He doesn't necessarily agree with the Reddit way of doing things, but it has interesting compromises wrt privacy.

[pixl97]

Because no one has developed a moderation framework based on behavior. Content is (somewhat) easy, a simple regex can capture that. Behavior is far more complicated and even more subject to our biases.

[ascv]

Honestly it seems like you didn't read the thread. He's not talking about how Twitter itself works but about problems in moderation more generally based on his experience at Reddit. Also, he specifically advocates public disclosure on ban justifications (though acknowledges it is a lot of work).

[dang]

He also makes an important and little-understood point about asymmetry: the person who posts complaints about being treated unfairly can say whatever they want about how they feel they were treated, whereas the moderation side usually can't disclose everything that happened, even when it would disprove what that user is saying, because it's operating under different constraints (e.g. privacy concerns). Ironically, sometimes those constraints are there to protect the very person who is making false and dramatic claims. It sucks to be on that side of the equation but it's how the game is played and the only thing you can really do is learn how to take a punch.

[sangnoir]

> Honestly, every ban should come with a public explanation from the network, in order to deter similar behavior

This only works on non-adversarial systems. Anywhere else, it will be like handing over to bad actors (i.e. people whose interests will never align with operator's) a list of blindspots

[noasaservice]

"You have been found guilty of crimes in $State. Please submit yourself to $state_prison on the beginning of the next month. We're sorry, but we cannot tell you what you are guilty of."

[vkou]

"Look, I'd like you to stop being a guest in my house, you're being an asshole."

"PLEASE ENUMERATE WHICH EXACT RULES I HAVE BROKEN AND PROVIDE ME WITH AN IMPARTIAL AVENUE FOR APPEAL."

---

When you're on a platform, you are a guest. When you live in society, you don't have a choice about following the rules. That's why most legal systems provide you with clear avenues for redress and appeal in the latter, but most private property does not.

[10000truths]

Imagine for a moment what would happen if this rationale were extended to the criminal justice system. Due process is sanctified in law for a good reason. Incontestable assumptions of adversarial intent are the slow but sure path to the degradation of any community.

There will always be blind spots and malicious actors, no matter how you structure your policies on content moderation. Maintaining a thriving and productive community requires active, human effort. Automated systems can be used to counteract automated abuses, but at the end of the day, you need human discretion/judgement to fill those blind spots, adjust moderation policies, proactively identify troublemakers, and keep an eye on people toeing the line.

[sangnoir]

> Imagine for a moment what would happen if this rationale were extended to the criminal justice system.

It already is!

The criminal justice system is a perfect example of why total information transparency is a terrible idea: never talk to the cops even if they just want to "get one thing cleared up" - your intentions don't matter, you're being given more rope to hang yourself with.

It's an adversarial system where transparency gets you little, but gains your adversary a whole lot. You should not ever explain your every action and reasoning to the cops without your lawyer telling you when to STFU.

Due process is sanctified, but the criminal justice system is self-aware enough to recognize that self-incrimination is a hazard, and rightly does not place the burden on the investigated/accused, why should other adversarial system do less?

[Spivak]

Being cagey about the reasons for bans is

1. To keep people from cozying up to the electric fence. If you don't know where the fence is you'll probably not risk a shock trying to find it. There are other ways one can accomplish this like bringing the banhammer down on everyone near the fence every so often very publicly but point 2 kinda makes that suck.

2. To not make every single ban a dog and pony show when it's circulated around the blogspam sites.

I'm not gonna pass judgement as to whether it's a good thing or not but it's not at all surprising that companies plead the 5th in the court of public opinion.

[bink]

Sorta related to (1) but not really: there are also more "advanced" detection techniques that most sites use to identify things like ban evasion and harassment using multiple accounts. If they say "we identified that you are the same person using this other account and have reason to believe you've created this new account solely to evade that ban" then people will start to learn what techniques are being used to identify multiple accounts and get better at evading detection.

[bitxbitxbitcoin]

He’s specifically referring to Reddit’s content moderation which actually has two levels of bans. Bans by mods from a specific subreddit are done by mods from that specific subreddit and having an explanation isn’t required but is sometimes given - these bans apply to just the subreddit and are more akin to a block by the community. Bans by admins happen to people that have been breaking a site rule, not a subreddit rule.

Both types of bans have privacy issues that result in lack of transparency with bans.