I used to hang out at the C2.com wiki, the first wiki ever, actually. It was devoted to software engineering. Unlike Reddit, the discussion topics remained "alive" for several years so that we could revisit ideas.
Discussions sometimes turned contentious, but it raised good questions deserving of more formal research, especially in terms of balancing tradeoffs (is getting less of X worth getting more Y?). I had what I call the "practioner's" viewpoint, which conflicted with what I'll call the "academic" viewpoint. We practioners viewed it more from a business expense standpoint while the academics wanted to build what seemed like an overly idealistic model of the domain and/or world. I didn't find such realistic without producing a million clones of Sheldon Cooper, but removing his arrogance so that he could work in teams.
However Ward Cunningham eventually pulled the plug on the wiki (it's read-only now), partly to experiment with a "federated" version of it (which is a failure in my opinion), and because he got tired of cleaning up after a nasty vandal. I think the vandal issue was solvable without shutting the whole thing down, but that's another story.