| > Like it or not, most problems you will want to solve are not "purely functional" but rely on sound management of shared mutable state. This is a claim which is certainly not true in all domains. For example, web applications and compilers, two of Haskell's "killer problem domains", do not heavily rely on shared mutable state. That said, Haskell's approach to shared mutable state is no less sensible than the rest of the language. > Just look at all of the theoretical mumbo-jumbo associated with something like Lenses...now look at the utterly trivial problem they solve. With all due respect if you believe this then I'm not sure you fully appreciate the problem that they solve. The terminology is indeed obtuse though. > Even for a FP tool, Haskell is riddled with cruft. Look at the set of compiler pragmas that are pretty much required to be in any useful Haskell. Including basic crap like overloading the String type. These are hacks, pure and simple. That's not fair. Haskell is a language that is (de facto) defined as a basis, the Report, plus a set of modernising features which have been implemented since. They're not hacks, they are deliberate optional language features. It's a different approach to most modern languages but no worse. A combined set of such features, called GHC2021, is the new "standard" and enables a huge swath of them by default. If anything I'd say there is very little hype around Haskell. People look at it and think it's cool, but that's not because of people hyping it up. In fact every post/article I've read about Haskell has been extremely measured and open about its shortcomings. Sorry if you just wanted to get something off your chest, but it makes me sad to see a perfectly fine language disparaged unfairly. |