I think the last paragraph of the "What Popper actually Said" section is just as important to note:
> It makes sense, doesn't it? Free speech is all fine and dandy, but let's stretch that to the limit. A and B are promoting their ideologies. A-ism is based on reasoned arguments — they may not yield correct conclusions, or they may, but A is speaking in good faith. B-ism is based on calls to violence and insurrection. If both are afforded the right to speak freely, modelling things out, B will necessarily inflict violence, or threats of such, on A — but violence and violent threats have the effect of silencing others, which indirectly impedes their right to speak freely — you are not 'free' to speak if someone will hurt you for doing so! Thus, free will is replaced with coercion, and society suffers as a result.
A line in speech has to be drawn somewhere. There's a difference among "I hate $RACE", "I hate $RACIAL_SLUR", "Someone needs to deal with the $RACIAL_SLUR problem", and "I want to kill $RACIAL_SLUR".
The first two on their own are not violent, and they're not written to stoke violence, but a person that sees as a lot of that content who agrees with it may eventually decide to escalate to violence.
The latter two are easily considered calls to violence, which even most free speech advocates might not agree with. The most aggravating thing about the people that post that kind of content is the blindingly obvious bad faith claims that "Oh, I don't mean we should solve the problem by KILLING them!" or "Well, I'd never ACTUALLY kill someone".
It's extremely easy to say "Well until the murder happens, no crime has been committed" when you're not the target.
Namely (for this case); true threats, fighting words, soliciting crime and/or imminent lawless action. The last two are in clear violation of this principle (3. soliciting crime 4. imminent lawless action), and whether or not speech is in violation of that can be determined in court. When people advocate "free speech" they're really advocating freer speech, as circumscribed by exceptions to the first amendment.
For the first two, the "[not written to cause violence but may cause it by misinterpretation]" argument is a problem to me. Intent matters, the writer is not responsible for nut jobs. Calls to violence & coercion are already impermissible by the law. I absolutely support these rules in real life.
It gets difficult when you consider that anonymity online means people will behave worse, since they have no reputation to protect or real-life consequences (just look at 4chan). Social media is not a great platform for Popper's 'reasoned debate', yet >48% of Americans receive news from it. Persistent, non-anonymous comments raise other problems, since it's a new phenomenon that everyone can essentially know every view you've ever expressed and attempt to get you punished for it. It's tricky, and I see the censors' point somewhat. I'd err on the side of free speech with a mechanism like anonymous accounts can't interact with verified, real humans. This should reduce bigotry & hate to the level experienced in real life (or perhaps lower), which is protected speech, but not what a reputable person would ever say.
Social media to me, really has become a public square - where politicians campaign, news orgs post articles, people & activists have discussions. 500 million tweets are sent per day. Once most discourse moves to these platforms, it should be necessary to allow controversial views, even if it is a private company. As such, censorship of "possibly false" information shouldn't be permitted, especially when this has been at the request of government officials, and some of it has later been confirmed true.
