Am 19.05.2012 12:00, schrieb wren ng thornton:
Exactly. That's what I was trying to get at re the problems of comparing Haskell to C++ (or indeed any pair of dissimilar languages). A legitimate comparison will involve far more than microbenchmarks, but then a legitimate comparison must always have a specific focus and context in order to be able to say anything interesting. The problems I see in language comparisons is that by and large they tend not to have any such focus[1], and consequently they shed little light on how the languages compare and shed a bit of darkness by serving only to confirm initial biases.
[1] A nice counter-example to this trend are papers like:
http://www.cse.unsw.edu.au/~chak/papers/modules-classes.pdf
There was another one comparing the expressivity of Java-style polymorphism vs Haskell-style polymorphism, based on an analysis of in-the-wild code; but I can't seem to pull it up at the moment.
Possibly "Software extension and integration with type classes" by Lämmel and Ostermann? http://doi.acm.org/10.1145/1173706.1173732 Best, Janis. -- Jun.-Prof. Dr. Janis Voigtländer http://www.iai.uni-bonn.de/~jv/