Conceptually, Stack Overflow did something amazing for its time. I agree however that after all these years there are now visible shortcomings in the approach,

Conceptually, Stack Overflow did something amazing for its time. I agree however that after all these years there are now visible shortcomings in the approach, and simply starting the same thing from scratch wouldn't solve the problems. The primary format of SO is Q&A. Some Q&A is timeless, as in the answer is still relevant today as it was 10 years ago. Some Q&A invalidates quickly and would probably need to be touched up every year or so. How to account for both? Another issue SO strongly struggles with is the same entry-level questions being asked again and again, in different wordings. I wouldn't really put the onus on the newcomers here, as they're often new to programming to begin with and thus couldn't know what's a problem other people might have already had, how to search for things etc. But the tooling to help them avoid those situations is crucial. Currently, possibly related questions are shown when writing a new question, but the problem still persists. How to better avoid this problem? Is this simply a matter of nailing down on the related questions algorithm or is there a better approach? And lastly, the company. There are many very real cases where the change in management has hurt the SO both as a site and a community. I think the answer lies somewhere between SO, MDN and Wikipedia - with all the ups and downs it offers. Clearly, making a global knowledge base and then tacking on investor interest later doesn't work out all that well. To date, I don't think this problem is really solved, as Wikipedia also has a number of issues it struggles with. Without addressing this core issue, the rest are more or less irrelevant with time.