I think there could be a balance here, allow that which is legal but distasteful but don't promote it or advertise beside it. You could probably repurpose some already pre-existing ML models searching for users to ban over takes that are too hot for twitter to mark as un-monatizable. Cleaning house is the way to make this vision of twitter a reality.

Bots I think is both the harder and existential problem.

I agree this is definitely possible, but if you do all that, is it still the public square open to debate, if you are selectably promoting / monetizing "tasteful" tweets?

I think distasteful nonsense gets deprioritized in all public squares as a natural consequence of our social mores.

Not really. Look at pretty much any reality TV show, or the magazines at any checkout line. Nonsense and drama is big business from an attention and advertising standpoint.

Plus, you can look at most biased news sources and see the drama that they selectively cover under the guise of news. "1 person did a bad thing" and "here's another reason why this entire group is conspiring against you" are not uncommon tropes.

My reference to distasteful was meant to reference a much more vile flavor of speech than gossip and celeb junk news.

Fair enough, but that’s also why it’s so hard to define your content moderation policy, because the definition of “distasteful”, etc. varies between people and over time.

> I think there could be a balance here, allow that which is legal but distasteful but don't promote it or advertise beside it

OK, but then it's more like blacklisting rather than banning. You're still free to show up and participate, you just can't get access to the biggest audiences.

So then the conversation turns from "why is Twitter blocking my free speech?" to "why is Twitter penalizing my free speech?".

Plus, what's legal speech is a real grey area. Free speech cases constantly go to the Supreme Court because we're talking about a near infinite number of ideas that can be expressed. And intent usually matters. So when Kanye writes his "death con 3" tweet, is it a direct threat to incite violence (not protected), an ambiguous threat that's not actionable (possibly protected), a joke (very likely protected), or as a brand new idea that hasn't been expressed yet (possibly protected)?

At the end of the day, a human has to make a decision. So, we're just swapping Jack's and Parag's vision of content moderation for Elon's, but it'll still be content moderation - perhaps better, perhaps not.

I think your example in Kanye might be a good one.

Maybe his tweet should stay up, ambiguous speech isn’t clearly illegal. Illegal speech is much easier to recognize. But Twitter probably wouldn’t want to put that tweet on the trending front page or advertise next to it but he has an audience and can continue to interact with and grow his audience.

Minimal regulation as required by law, no moderation, and maximal participation.