[vuanotino]

Muting words is dumb. It's not really complicated to write n1gger instead of nigger.

Suspending accounts is the dumbest, though—the account of an abuser is likely to be in the block/mute list of hundreds of people. By suspending his account, the user will just create another and continue annoying the very same people that had blocked him already.

I wonder who in Twitter, Inc. uses Twitter at all.

[ohyoutravel]

A crowdsourced "blocked list" similar to ublock origin for twitter?

[TorKlingberg]

This already exists. Of course, you get plenty of disagreement about who should be on the list.

[huac]

these exist, but because 'annoying account' is a subjective decision, while 'advertisement url' is an objective decision, it is difficult to create a list with a clear brightline between harassment and disagreement

[nailer]

There is at least one hard-left feminist who ended up on GGAutoblock because Randi Harper just disliked her.

[jansenv]

This sounds like the kind of thing that will get abused if widely adopted.

[digler999]

then you could harass people by falsely getting them on a block list.

[strictnein]

I'm already on some block lists. I'll run into some journalists who have me blocked. Never interacted with them. I hardly interact with anyone, but some block lists just take the entire list of followers from people and add them.

[joezydeco]

By suspending his account, the user will just create another and continue annoying the very same people that had blocked him already.

Why can't Twitter do what Metafilter does? $5 to sign up and a 24 hour delay on any posting.

[ionforce]

I like the idea of that kind of approach (maybe not for Twitter), but I don't have any experience with it.

How is it working out for Metafilter?

[joezydeco]

I think it works pretty well. Granted, Metafilter is also heavily moderated to prevent conversation derails and general abuse, but that's also what the money pays for (not like money is a problem for TWTR, or is it?)

I don't believe the payment is used towards enforcing identity (MeFi values privacy and anonymity when needed) and locating abusers, but if things got really bad I don't see why that would be a huge problem.

[matt4077]

Doesn't twitter require confirmation via sms now to sign up? It's quite a hassle to find a different phone.

[snerbles]

For a determined user, rolling burner/VOIP phone numbers is not much of a hassle.

[anigbrowl]

An obnoxious user can create another account but it takes work and the new account won't have the same reach as the old one. Of course there are strategies to circumvent this, but the economic costs for the troll steadily increase. Once the costs exceed the propaganda benefit, it drains resources from the propaganda effort.

[marxidad]

But they are also harder to find.

[nailer]

Sure, but there are libraries to identify visual similar words. And yes, you would fight them with being even more obscure, but at some point they'd almost become unreadable.

How about a bayesian approach, like email?

Contains masked content +3

New account +4

No profile pic +5

Follows accounts with low reputation +4

Greater than out imaginary threshold of 10 so maybe hide the mention from whoever's timeline?

[leereeves]

That's not a Bayesian approach.

[nailer]

There are degrees of belief attached to each aspect and the proposal works exactly like existing acknowledged-as-bayesian approaches eg spamassassin does. So how isn't it?

[leereeves]

A Bayesian approach would learn the degrees of belief empirically from examples of spam and "ham" messages.

There are other machine learning approaches that would also learn from data - specifically, a Bayesian approach would use Bayes theorem.

[nailer]

Well yes, you'd base those weights on actual, human confirmed account bans.

Nothing about my post ever suggested the value was arbitrary. You've just assumed that.

[leereeves]

The inclusion of arbitrary values suggested you would actually use those values.