Re:Haskell Problems for a New Decade
by mumblemumble on Friday January 24, 2020 @09:29AM
(#22138546)
This has been a growing concern of mine about the whole functional-as-defined-by-John-Backus (viz, not including lisp) lineage of programming languages. On my grumpy days, I'm inclined to say that the current culture can be best described as, "Got excited and ran off to do some coding before making it all the way to the end of The Evolution of a Haskell Programmer." (https://www.willamette.edu/~fruehr/haskell/evolution.html)
That doesn't particularly bother me in Haskell, which was born as, and continues to primarily serve as, a platform for language research. (Secondary purpose, at least for me: A platform for daydreaming about a non-post-C++ existence. I suspect that professional use remains a small minority.) It gets more worrisome to me in an ostensibly industry-oriented language like Scala. There I start to fear that my colleagues choose ways of solving problems for about the same reasons that I prefer mechanical watches over quartz ones: Not because they're more practical or more economical - no in both cases - but because they're cooler.
I think it'll probably get better naturally, though, the same way that lisp programming culture eventually stopped being quite so enamored of macros, and Java programming culture eventually stopped being quite so enamored of elaborate XML configuration files.
That doesn't particularly bother me in Haskell, which was born as, and continues to primarily serve as, a platform for language research. (Secondary purpose, at least for me: A platform for daydreaming about a non-post-C++ existence. I suspect that professional use remains a small minority.) It gets more worrisome to me in an ostensibly industry-oriented language like Scala. There I start to fear that my colleagues choose ways of solving problems for about the same reasons that I prefer mechanical watches over quartz ones: Not because they're more practical or more economical - no in both cases - but because they're cooler.
I think it'll probably get better naturally, though, the same way that lisp programming culture eventually stopped being quite so enamored of macros, and Java programming culture eventually stopped being quite so enamored of elaborate XML configuration files.