Some Senator who's name I can't recall was arguing otherwise. He said something to the affect of, if social media companies, or websites in general want protection under the law from content generated by their users, then they should not be able to dictate what their users can and cannot post barring illegal content.
It's an interesting argument regardless of what you believe, but certainly your first amendment rights don't currently extend to Twitter.
He wasn't arguing the current interpretation of the law as much as proposing an amendment.
I don't think that HN, nor the community at large does as much as say twitter or reddit to stifle opposing points of view. Relegating extremist to their own echo-chambers is a good way of proliferating their beliefs. Were their ideas exposed to the world where they would have to defend them, they may be dissuaded from holding such extreme beliefs, or a passerby may see the flaws in their argument.
> He wasn't arguing the current interpretation of the law as much as proposing an amendment
I wanted to point out it would require radical changes to our current laws and maybe the Constitution. I wasn't sure if he was aware of that. (Stronger versions of this argument, including ones deployed by members of Congress, have said it described current law)
> Were their ideas exposed to the world where they would have to defend them, they may be dissuaded from holding such extreme beliefs, or a passerby may see the flaws in their argument.
One interesting study mentioned in that summation found that temporary internet outages in Germany reduced hate crimes against refugees.
> This paper investigates the link between social media and hate crime. We show that anti-refugee sentiment on Facebook predicts crimes against refugees in otherwise similar municipalities with higher social media usage. To establish causality, we exploit exogenous variation in major Facebook and internet outages, which fully undo the correlation between social media and hate crime. We further find that the effect decreases with distracting news events; increases with user network interactions; and does not hold for posts unrelated to refugees. Our results suggest that social media can act as a propagation mechanism between online messages and violent crime.
With regards to the paper you posted, I don't think it does enough to escape the fallacy of correlation/causation. It stands to reason that a news event implicating refugees for a crime will attract anti-refugee sentiment, likewise why would such sentiment spread to unrelated news posts? Facebook has already been caught prioritizing controversial content in user news feeds to increase interaction. Facebook will quickly learn a user that displays such views interacts more with said content, and display more.
It's an interesting argument regardless of what you believe, but certainly your first amendment rights don't currently extend to Twitter.