As a practical matter, we cannot ignore the effect of technology on speech. To wit, what I see going on are the production of more and more powerful self-sustaining nuclear reactions of ignorance. The nuclear fission analogy is apt, because it requires a certain level of "critical mass" for any idea to catch on. Technology accelerates greatly the rate at which like-minded people can accumulate in the same, dense space; the neutrons are likes and comments; the explosion is rage and violence.<p>Technology has had this problem for a long time - the yellow newspapers of the 18th century arguably started it; the phenomena of toxic radio personalities began immediately after radio was invented. Goebbel's movie propaganda was a critical part of Hitler's rise to power.<p>The internet, and platforms like Twitter, Reddit, Gab, 4chan, take this to a whole new level. They are the ultra-centerfuges of online discourse, distilling and concentrating the extremes of human discourse into a dense, self-sustaining mass of negativity.<p>And yet we generally <i>don't</i> get an equal and opposite mass of positivity. I believe the central reason is that humans are super biased to address human negativity, rightly judging it to be the greatest threat at small scale, but wrongly judging it at scale. For example, the 9/11 attacks killed ~2,000 people. Car accidents in the US in 2001 killed ~30,000 people. But one event was rooted in negativity - hatred, rage - and the other was not. So one got trillions applied to it's solution, and the other got nothing.<p>In a worldwide search of 7B people, I believe you can find a self-sustaining mass of hatred on every side of every topic.<p>We may need a modification to the free speech law in the presence of a sea-change in speech technology. Perhaps "slow speech" (offline, or online and decentralized) should remain perfectly free, but "fast speech" (online and centralized) should be censored - to defuse any given critical mass that threatens to explode into hatred and rage.<p>I think it would also be useful to enforce a convention that differentiates between speech and paid-for speech. In this way we can more accurately identify who is speaking as a form of expression, and who is speaking as a means to an economic end. The former should be protected from censorship, and the latter should not.