relatively recent discussions

http://www.reddit.com/r/ocaml/comments/fmpl8/i_have_yet_to_l...

http://www.reddit.com/r/programming/comments/gd4xq/james_gos...

http://flyingfrogblog.blogspot.com/2011/03/pros-and-cons-of-...

http://groups.google.com/group/sayeret-lambda/browse_frm/thr...

Looking at the shootout, it would seem that the Java 7 stuff is rocking the house. Hrmm, now when Java gets real non-blocking IO, this will be very interesting.

Also, this article is from 2002.

OCaml losing its luster to me as I age. I just wrote down counter-points to this article: http://news.ycombinator.com/item?id=2983248

If F# worked well on Mono I would be using F# for a lot of my work, and F# isn't a million miles away from being OCAML.

What doesn't work?

As far as I know that combination is working fine and even supported/embraced by the teams involved?

I've been meaning to try this. In meantime: Only 1 critical open ticket:

https://bugzilla.novell.com/buglist.cgi?query_format=specifi...

Not a lot volume on google group and stackoverflow

http://groups.google.com/group/fsharp-opensource/topics

http://stackoverflow.com/questions/tagged/mono+f%23

I haven't tried with the latest version of Mono but, with the previous version, some useful and common libraries (e.g., FParsec) weren't fully supported. I've heard that F# support has improved in the latest version, but that it's still not complete.

I haven't tested it myself because I got tired of working around Mono issues, and just wrote a little script that sends .fsx scripts to my windows machine, runs them there, and returns the output on my linux machine. I can see why you wouldn't want to do that in a production environment, though.

The article is from 2002. I think now Haskel might be the best alternative. In 2002 though OCaml tools were probably better and more mature than Haskel ones.

I find that the Haskell language is too vast to be directly useful to me. It appeals to my programming language fetish, which is a curse when I want to get something done. I rapidly end up devising between different ways of laying my types down, which is brain-gasmic, instead of how to get a result. And in the end I find out that it's not efficient enough and that I have to re-write the program with less "pureness" to get some juice out of it.

In contrast, OCaml's syntax and model is quite simple. The IO is not pure but at least I don't have to choose between 20 different ways to do it. There is not as much activity in the community but at least you don't get distracted by the latest monad library. For somebody who just want to crunch numbers, I think it is a good choice.

> ML is a strong, statically-typed version of Lisp.

Meh? How so?

The author explained why.

http://en.wikipedia.org/wiki/Type_inference

Not to put words in other people's mouth but I believe he is more concerned with 'version of Lisp' than the 'statically-typed' part. It is kind of like saying basic is an interpreted version of C. The two languages are more or less unrelated outside of the fact that both are impure functional programming languages.

I'm familiar with type inference, thanks. I have no idea what it has to do with Lisp.

I just interpreted the article an imprecise way of saying " they're both functional, but one is statically typed," by assuming Lisp as the canonical example of FP (rightly or not).

Not the most apt comparison, to be sure.

Wouldn't that be Typed Scheme/Typed Racket?

Without the macro-friendly syntax.