3 ms·
I'd skip both of them and jump to Scheme. I've found Scheme to be far more elegant than Common Lisp or Clojure, far more modern and less clunky than CL, and yo
by pmoriarty 9y ago
I'd skip both of them and jump to Scheme. I've found Scheme to be far more elegant than Common Lisp or Clojure, far more modern and less clunky than CL, and you don't have to deal with the JVM or Java as you do with Clojure.
Chicken Scheme is my favorite, but Racket or Guile would work too. Guile is good because Emacs is slowly transitioning to using it, and Racket for its great ecosystem.
Chicken Scheme itself compiles down to C, so it's quite fast, portable, and light-weight. It also has a great ecosystem, library of eggs, and community.
Others have mentioned Elisp, which emacs uses, and that's good too if you are interested in emacs. It would just be a shame to start with it and get the impression that that's the best that a Lisp-like language can offer when it's really rather primitive compared to a modern Scheme, Clojure, or CL. I myself learned Elisp last, after already learning CL and Scheme, and found doing so to be very easy after having learned the others. So I'd really recommend starting with one of the others and learning Elisp last, if you're in to emacs.
- brudgers 9y agoWhat is interesting about eLisp is that Emacs provides an environment where there are lots of low hanging reasons to write a little code in eLisp in order to tweek the Emacs editor to do something new...there's never a shortage of small, interesting, and relevant projects/problems at hand to address in eLisp (and the tooling for tackling that class of projects/problems is very good and always at hand). For exploring Lisps in general, such 'strange loops' around building tools do not seem quite as obvious obvious or natural. Particularly at small scale programs.
- Mediterraneo10 9y ago> Guile is good because Emacs is slowly transitioning to using it Emacs on Guile is nearly a one-man project that has been progressing in fits and starts, and it's not guaranteed that that branch will be adopted as the Emacs mainline. Basically, the Emacs devs are telling the Guile folks that the onus is on them to make something worth integrating into the Emacs core, and so far it's nowhere near that. But it's important to realize that Emacs on Guile is only about using Guile as the runtime for Emacs Lisp, as multiple languages can compile to Guile bytecode, not just Scheme. Even if Emacs on Guile happened, the user would not program in Scheme, but still in Emacs Lisp.
- pmoriarty 9y ago"Even if Emacs on Guile happened, the user would not program in Scheme, but still in Emacs Lisp." Well, you could program in eLisp, and most people probably would, but you could also program in Scheme. Most emacs programs would remain in eLisp for a long time, though, as it'll take a while to convert them all over to Guile. So, yes, this is a long-term project, and most people would continue to program in eLisp for the foreseeable future, but some would program in Scheme and try to convert existing eLisp programs to Scheme. That's my understanding of the situation, anyway.