Comment by armchairhacker

On one hand, I agree that internet comments tend to judge people unfairly, and “treating people as guilty first” probably leads to an unhealthy society (considered “unhealthy” by its own members).

On the other hand, GP is objectively right ("innocent until proven guilty doesn't extend to internet comments"). I also think that it’s better for random people to be able to post their terrible judgements than any feasible alternative, because such an alternative probably leads to good judgements also censored. We can mitigate (not eliminate) bad judgements, e.g. by educating people better and shaming those who shame others more; and we can minimize mob justice’s effect on critical government functions like welfare and prison sentencing, e.g. by running them on mostly objective procedures and with staff who aren’t influenced by mob opinion.

Targeted harassment and doxxing (and swatting, getting people fired/divorced/ruined when they don’t deserve to be, etc.) is different (and to be clear, IMO very bad). People posting opinions in a way that the target can block (which they can usually do with blocklists and word filters) is fine. The main point I’m trying to make is: if opinions in random internet comments lead to targeted harassment and real-world consequences even when the opinions are “bad” (e.g. bigoted, hypocritical), it's less effective to try and prevent the internet comments' existence, than to reduce the factors causing them to influence the real world and create factors preventing influence.