News for hackers. Stuff that matters.
Re:What I Wish I Knew When Learning Haskell 2.0 by AnimalMuppet on Thursday April 24, 2014 @01:10PM (#7641537)
There's some sound wisdom in this post, jerf. A couple of questions, though:

First, is it Haskell that drives you toward well-factored, reusable designs? Or is it functional programming? (I know, you can't do Haskell without functional programming. You can do functional programming without Haskell, though. Can you separate the two, and say which one is driving you here?)

Second, have you ever seen Haskell deployed in a large, long-lived production code base? How large? How long-lived? How did it work? (I ask because I remember reading an article on Lisp, which said that X in Lisp is just like Y in an object-oriented language, "with the proviso that in Lisp, all the types exist in the programmer's head". I'm pretty sure that won't work very well on a million-line code base that lives for 30 years. Haskell has a very solid type system, so it won't run into that particular problem but... does it have a track record for large-scale work out in the real world?)

Re:What I Wish I Knew When Learning Haskell 2.0 by jerf on Thursday April 24, 2014 @01:29PM (#7641649)
"First, is it Haskell that drives you toward well-factored, reusable designs? Or is it functional programming?"

In my opinion, it's the purity, and the general simplicity of the individual pieces which forces you in certain directions. Since this conversation is dead-ish now, it's probably not a big deal to link to my longer musing on this: http://www.jerf.org/iri/post/2908 Even functional languages that make impurity easy and available end up making it too easy to jump down the escape hatch.

"Second, have you ever seen Haskell deployed in a large, long-lived production code base?"

I personally have not. As I said, where I work I tend to tamp down on any efforts to use Haskell, and there's not many! But I trust the people who have and do, and who knows, perhaps may join them someday.