callmecosmas

Haskell seems pretty left brained to me!

Actually at the bottom of page 2 of Maynard's preprint he states: "We emphasize that the above result does not incorporate any of the technology used by Zhang to establish the existence of bounded gaps between primes."

So it seems like it is actually a coincidence, though they were both building off of the same previous work (GPY) so maybe it's not that surprising.

Why not F#?

I think the main difference is that SLIME is for Common Lisp and Geiser is for Guile/Racket. More than that, I don't know enough about SLIME to say.

Haskell certainly has a big learning curve, but the experience will be much more rewarding than learning Clojure or F#. Haskell is a (relatively) uncompromising language and Clojure and F# are really just trying to take some of the features of Haskell and bring them into the mainstream. So if you learn F# or Clojure it may help you if you ever try to learn Haskell, but if you learn Haskell, F# and Clojure will be trivial to learn.

You don't actually need the second lambda since all Haskell functions take one argument, so you can just do

  f = \x y -> x + y

Python implementations do not have tail call optimization but that doesn't really have anything to do with functional programming.

Common Lisp has mutable data structures by default and actually I think one of the most elegant parts of CL is the setf macro http://www.lispworks.com/documentation/HyperSpec/Body/m_setf..., which is a very clean interface to changing state.

The best functional lisps to use right now would be Clojure and Racket. They both use immutable data structures by default, though neither is "pure". Clojure if you want all the good JVM bits and maybe actually convincing your manager to use it and Racket if you just want to learn using a language that's more beginner-friendly (http://htdp.org/ and the DrRacket IDE) or (mostly) compatible with Scheme (SICP, Little Schemer, etc.).

I've found there aren't really any widely used purely functional languages out there except Haskell (which needs it due to laziness).

I'm a grad student that works in Racket and fwiw I find that I prefer emacs for editing but DrRacket for testing and debugging. It has great tools for documentation, locating errors and macro expansion and of course it all works out of the box. I also use Geiser, but I've found it buggy and unreliable (including instantly shutting down my OSX machine if I tell emacs to kill it when I quit, plus I still haven't figured out how to limit its memory consumption).

If you're going to try Scheme and you're scared of emacs/vim, I'd suggest Racket. It's got a nice IDE (DrRacket) and a lot of the community is focused on teaching so there's a lot of resources for newbies http://racket-lang.org/learning.html

I'm an emacs lover myself, but for that use-case it seems like all you need is rlwrap: http://utopia.knoware.nl/~hlub/rlwrap/#rlwrap

Racket allows for (), [] or {} in code and it seems that in practice [] is used to distinguish some syntactic forms that have nested parens from normal function calls such as the let and cond forms. So instead of

  (cond ((...) (...))
        ((...) (...)) 
        (else ...)) 

  (cond [(...) (...)]
        [(...) (...)]
        [else ...]) 

I mean I can see how that would be a sexual joke given a certain tone of voice.

Pretty much: https://github.com/imkevinxu/faithinhumanity/blob/master/fai...

Cool work but doesn't mix well with text selection. I'm not into js but I wonder if it would be easy to distinguish between clicking on one spot and clicking and dragging for say copy-pasting.