Reading posts like this make me so grateful for good moderation, and for not being responsible for nurturing and protecting an online community, because it soun

Reading posts like this make me so grateful for good moderation, and for not being responsible for nurturing and protecting an online community, because it sounds far more difficult than it should be. As a technologist it seems like we want a high-tech solution (and I'm surprised that AI-driven moderation isn't mentioned more in the discussion; it seems to me you could accomplish a lot with sentiment analysis plus other heuristics like account age and activity). But another thought is to treat trolling itself as a disease to be modeled as a diffusion process, infecting previously good actors and tempting them into trolling behavior themselves. Its remarkable how fragile communities are in the face of even one or two determined malefactors! The OP's sensible approach feels like a kind of vaccination itself, in that it should slow the rate of reinfection such that while it won't eliminate trolling entirely, it will keep the inevitable outbreaks small and localized and not self-sustaining. But, sheesh, what a PITA. Is anyone working on automating the distracting/annoying/frustrating task of internet forum moderation?