That's not what I've observed. The discussions actually turn out to be quite inflammatory when these matters are involved. Even folks that claim to have PhDs in

That's not what I've observed. The discussions actually turn out to be quite inflammatory when these matters are involved. Even folks that claim to have PhDs in language design consistently confuse these concepts. That's the only way I can explain all the condescending remarks when it comes to things like type systems, monads, curry-howard correspondence, etc. In fact the above article is a response to one such inflammatory and condescending blog post by a programmer who clearly values theoretical beauty and specific patterns that typeclasses enable above software engineering pragmatism. Although I can see how typeclasses help in cutting down boilerplate in this case his tone does not help make that point. I'm no stranger to abstract and theoretical beauty. My undergrad and grad training is in pure math but I also understand that theorems and proofs don't necessarily lead to good software engineering in the large. So valuing the abstractions is great and all but using it as the ultimate arbiter of good language design is not.