I enjoyed this article enough to wander about James' site for a bit. But I didn't appreciate his core point until I read several other articles: judge a language by the problems it solves. Or even more harshly: the problems it has solved. He hammers it home in this gem, which may or may not have appeared already:
...And I reluctantly agree with him. I love functional programming. The code is art.
But I cannot justify building a business around it: he hits many good points, but to me the killer is library availability (it's the natural extension of judging based on achievement). When I start a new project, I want to write as little code as possible. I can't do that if slightly-more-complicated-than-basic structures and patterns aren't available to me.
Sadly, this isn't even a criticism of the languages I love. It's circumstantial. But it's still killer.
http://prog21.dadgum.com/13.html (http://news.ycombinator.com/item?id=516718)
...And I reluctantly agree with him. I love functional programming. The code is art.
But I cannot justify building a business around it: he hits many good points, but to me the killer is library availability (it's the natural extension of judging based on achievement). When I start a new project, I want to write as little code as possible. I can't do that if slightly-more-complicated-than-basic structures and patterns aren't available to me.
Sadly, this isn't even a criticism of the languages I love. It's circumstantial. But it's still killer.