Even with infinity resources available for manual human moderation, you eventually hit a wall where different sub-communities will simply have different standar

Even with infinity resources available for manual human moderation, you eventually hit a wall where different sub-communities will simply have different standards for what is acceptable to them— what might be reasonable debate in one circle is gaslighting and triggering to people elsewhere. It's not really up to the platform to impose a global code of conduct, and attempting to do so (outside of banning the most obvious of bad behaviours or things that are actually illegal) never seems to go well for platforms. So yeah, I agree with TFA in the sense that these are problems to be solved largely at the system level. For example, compared to Twitter (where anyone can reply-to, quote, RT, and @user anyone), Twitch and Tiktok seem to do well at permitting individual creators to have their own space with their own exclusive authority over what is and isn't okay in the space. And they have (or at least enable to exist) lots of tools for exerting that authority— witness things like "bye trolls" scripts on Twitch that do have to be set up in advance, but then can be used at the drop of a hat in response to brigading to immediately close the stream to new followers, and disallow posts from non-followers, plus delete chat posts from anyone who joined the stream in the last X minutes.