[x38iq84n]

Level 3 and beyond are nonsense. Just give users the tools to moderate what they see. Don't want to see slurs? Here's a toggle for that. No need to censor those who say them. Memes fall under Fair Use, nothing needs to be taken down. Government of Malaysia can start their own Twitter. $TWTR needs to follow US laws, not laws of Malaysia.

Here is a radical content moderation policy: Block/take down things that are illegal by law and leave the rest.

Why is it so hard?

[jwarden]

No open online community with such a radical content moderation policy can survive. It is too easy for a small number of people to flood a community with spam and abuse. It has happened countless times. And there is a lot of good writing on the phenomenon:

Clay Shirkey, A Group is its Own Worst Enemy: https://www.gwern.net/docs/technology/2005-shirky-agroupisit...

Lesswrong, Well-Kept Gardens Die by Pacifism: https://www.lesswrong.com/posts/tscc3e5eujrsEeFN4/well-kept-...

[latexr]

> Here is a radical content moderation policy: Block/take down things that are illegal by law and leave the rest.

That’s essentially what every social network catering to the far right claims as their shtick. They soon realise it doesn’t work: https://www.techdirt.com/2021/07/12/it-appears-that-jason-mi...

[pjkundert]

You know what doesn’t work?

Attempting to moderate all human communication.

Impossible, by definition - it is produced at a rate as high as the humans involved, so short of a Stasi-level police state, it can’t work.

So instead of trotting out the “far-right” boogie man every time someone doesn’t give you the impossible, let’s figure out how to give the local police the tools to do their jobs?

[themoonisachees]

1) you'd be surprised at how many things are illegal to say somewhere. Sure, malaysia can mame its own twitter, and so can india, and the EU, and russia, and china, and japan, and soon you only have a costumer base in the united states.

2) okay. Do that. Make a platform that stops at 3). Get sued despite fair use (you have a case, but do you have sony money to defend it in court? If you fail, you are now responsible for the legal precedent that memes aren't fair use). Allow people to say the n-word and see how quickly you'll get dropped by advertisers who think having an ad next to a tweet asking for racial genocide is maybe not exactly good brand image.

3) spam is a considerable part of user experience. If every second tweet on your timeline is spam, users aren't going to block and move on, they're going to stop using twitter. Your antispam is not flawless.

It's hard because "block things that are illegal by law and leave the rest" is unfathomably hard to follow for a human, let alone an automated system.

[x38iq84n]

1/ Twitter is a US company. It needs to follow US laws and it can ignore laws of all countries that are not the USA. Why should a US company dance to the tune of foreign governments anyway? Not to mention that those requirements may be mutually exclusive or contradictory.

2/ Twitter, not being a publisher, is not responsible for user-generated content. Just like Verizon is not responsible for what people say on the phone to each other. Advertisers should have some controls over what content their ads display next to. Giving controls to users and advertisers is the key, not heavy moderation.

3/ once again, controls. Twitter already has an option to show (in feed) only tweets from the people you follow and their connections, that's a good start and I see no spam at all.

"Block things that are illegal by law and leave the rest" is hilariously easy but must be accompanied by tools that allow users and advertisers to tailor their own experience.