There's this constant tension between the principle of free speech, which allows people to freely express their opinions, and the individuals rights to not have to listen to speech they find unpleasant or otherwise annoying/distracting.

In everyday life we solve this by letting anyone say what they want where they want, and anyone who doesn't want to listen avoids going to those places where people are saying things they don't want to listen to. Newspapers and magazines will print whatever most appeals to their audience, and people only read the ones that interest them. Mostly it works out.

The same approach works pretty well in blogs. People can write whatever they want, but nobody has to read anything they don't like. Those blogs which are most unpleasant won't get many links and so will be difficult to find unless you're searching for them directly.

If you don't like what somebody's writing on a Whatsapp group, you can either ban them from the group, or leave the group yourself.

On twitter though this approach starts to break down. Somebody can reply to thousands of people, none of whom have expressed any interest in their views, and their responses might be seen by millions of people. Sure, you can block them one by one, but there's more people to block than you have seconds in your day.

The current approach taken by twitter in extreme cases is to ban the user from twitter. Whilst that's a reasonable approach in a Whatsapp group, where there's always another Whatsapp group to join, twitter has over 450 million active users - over 10% of the world population, and has an enormous impact on popular discourse. Stopping someone from being able to air their views there definitely significantly impinges on their ability to freely express their views - and the threat of being banned can have a similar impact.

A simple solution could be to, instead of banning someone from twitter, simply stop them from being able to reply to non-followers[1]. They can still express whatever opinions they want, and anyone who wants to could search for them, follow them, retweet them,  reply to them and link them. But they wont be able to use twitter to forcefully communicate with people who aren't interested. If they're popular enough (e.g. Donald Trump), they'll continue to get plenty of engagement on twitter, satisfying our desire for free speech. If nobody is interested in them they can continue tweeting into the void eternally.


 

  1. ^

    As far as retweeting is concerned, I would suggest that they would be able to retweet non-followers, but it wouldn't show up in the list of retweets.

New Comment
13 comments, sorted by Click to highlight new comments since:

I think the problem with this post is that it's not clear about what the actual issue happens to be. There are a few different stakeholders with different interests and this post tries to avoid thinking about the interests of the individual stakeholders but focus on abstract principles.

Twitters current approach of how to deal with people is a lot more complicated than either banning or not banning a person. 

I want LessWrong posts that touch on politically charged topics to focus on trying to deal with reality in the complexity that it has instead of dumbing down complexity. 

I want LessWrong posts that touch on politically charged topics to focus on trying to deal with reality in the complexity that it has instead of dumbing down complexity. 

Every post touching on political topics over the last week has some elements of 'dumbing down complexity' though?

This seems like an unrealistic expectation to impose on casual writers.

One definition I found for casual was "not serious or careful in attitude". I don't want to read political writing on LessWrong that's not serious and careful in attitude. 

LessWrong doesn't exist for casual political conversation. There are serious political topics where it makes sense to speak carefully about them, and I do think those have a place on LessWrong but LessWrong can go without casual political discussion. 

One definition I found for casual was "not serious or careful in attitude". I don't want to read political writing on LessWrong that's not serious and careful in attitude. 

The  majority of such writing on LW is not seriously or carefully prepared beforehand. And that's not limited to just political topics either. 

This solves only a 3rd of the problem - speech that listeners do not want to hear. There are two more

  • Speech with a negative externality. If somebody with a lot of followers starts advocating genocide, we might want to block it even if the followers want to hear it. But then the question becomes - how much negative externality, how much certainly in that externality existing, etc should speech have before we'd want to block it (this is the "COVID vaccine misinformation", "voting misinformation" and other similar categories where clear boundaries are hard to draw, and where any decisions - whether to block, or not to block - would always be controversial).

  • Posts next to ads. The "real" customers of Twitter are advertisers, not users, and they have strong preferences of what kinds of posts they want to be or not to be next to their ads.

Upstream from this: stop optimizing for user engagement.