True but nobody is refuting the logic of how it works. They're refuting the reasoning behind why it works that way. As is, you get more points and labeled as an expert of a language if you choose to answer the easy, low hanging fruit questions. On the other hand, if you actually are an expert of a language and tackle those questions that only experts can answer, you get 10x-100x less points.
But aren't the problems which only experts can answer by their very nature both problems that most people won't encounter and problems that most passers-by are not likely to feel confident passing judgement on possible solutions for? There is the "bounty" system which I think tries to account for that a bit but I don't see it often used, other than that the only way around the expert requiring problem problem is some form of manual curation by domain experts, which would potentially introduce a small shipment of other worm cans.
