Right. It always kind of puzzles me when folks start nonchalantly talking about what kind of restrictions could be placed on speech in order to make it less to

Right. It always kind of puzzles me when folks start nonchalantly talking about what kind of restrictions could be placed on speech in order to make it less toxic. Talk about a slippery slope. The reality is that social media is exactly as toxic as the crowd that runs any one splinter of it, and it's not by any means a cohesive whole. I think that there is some irony to a group of computer scientists looking at a social problem from a context of "What filters can I apply to this bad data in order to have a good output?". I think that it's (mostly) irresponsible, and damaging as a trend. Basically : service providers are in a conundrum where the product they sell appears to be discourse and speech, but due to human nature the 'product quality' dwindles, so the service provider treats the 'product' like any other, and iterates over it until it sells better. Shaping discourse and speech in an effort to 'sell better' as a social media 'product' is going to silence a lot of people, and is in no way free discourse. 'How it sells' is not a valid metric for the judgement of discourse, unless you're profiting -- and then that position puts your motivation towards purely free discourse into question.