I sincerely doubt that the discussions which began on the leading edge would return anywhere near the same amount of progress as those which start in the scholarly middle. After all, those problems are on the edge because they're difficult to solve given the intellectual tools we have today. Though Less Wrong is often insightful, I suspect it's the result not of discovering genuinely new tools, but of applying known tools in ways most readers haven't seen them used. For Less Wrong to make progress with a problem that a lot of smart people have been thinking about in detail for a long time either requires that the entire field is so confused that no one has been able to think as clearly about it as we can (probably hubristic), or that we have developed genuinely new intellectual techniques that no one has tried yet.