On 2009-12-22 10:43:17 +0100, Walter Bright <[email protected]> said:
One of the big problems with the presentations of the functional qsort is that it is presented as a shining success of FP, when it is anything but.
There are better examples of 'shining succes' in FP. For example, consider the very clear and terse examples of balancing binary search trees or heaps in the Okasaki book. DSLs are also another good example where FP works very well. However, these examples are probably too long to catch attention from a casual reader browsing the haskell.org, or as an introductary example of functional languages.
Furthermore, an introductory FP programming text should be about how to write useful programs in FP, not about how sorting algorithms work.
'ML for the working programmer' is certainly one of the better FP programming texts I have seen, and it discusses various sorting algorithms and how they work (with the head as a pivot in quicksorting :p). But how is that problematic? Some of good programming books will end up being used in CS curricula, and then it's useful to learn something along the way, rather than have a dry treaty about yet another language.
-- Daniel
