[hairofadog]

One of the things that frustrates me about discussions of censorship here on HN is that there’s a lot of intense focus on censorship via deleting a tweet or Facebook post, but no focus given to the more insidious problem of censorship by algorithm.

I am wholeheartedly in favor of a free marketplace of ideas where (we would hope) good ideas win out over bad, but as it is, once you’re deemed by an algorithm to be susceptible to a certain category of extremist information, that’s all you’re ever going to see again; the competing ideas are never going to have a chance.

Algorithmic distribution of ideas is sorta like distributing ideas via gasoline-powered leaf blower directly to the face. I am free to speak my competing ideas, and so technically I haven’t been censored, but no audience is going to hear me over the leaf blower.

[visarga]

We need to get some level of control over the criteria for ranking and filtering. A third one is the UI - it is the place where all sort of dark patterns hide.

I'd like to see the browser put in a sandbox and its inputs/outputs sanitised and de-biased before being presented to the user. Could also protect privacy more. We need more browser innovation. A neural net should be in every browser ready to apply semantic rules.