The problem of online communities is not solved. We have a lot of data points of failure but I am not sure we have a good overview of why everything failed. Th

The problem of online communities is not solved. We have a lot of data points of failure but I am not sure we have a good overview of why everything failed. There should be like a wikipedia table like this one [1] but where explanatory factors (funding model, corporate structure, moderation tools, algorithmic timelines, discovery modes, anonymity, upvoting/downvoting mechanisms etc etc) would be tabulated against failure modes (toxicity, inanity, obsolescence etc etc) You then run an ML model and find the pattern. The problem is that our sample is totally imbalanced. We don't know what good looks like. The solution might be to take life expectancy before enshitification into account. [1] https://en.wikipedia.org/wiki/List_of_social_platforms_with_at_least_100_million_active_users https://en.wikipedia.org/wiki/List_of_social_platforms_with_...