Little Lisp interpreter

    var tokenize = function(input) {
         return replace(/\(/g, ' ( ')
               .replace(/\)/g, ' ) ')
               .trim()
               .split(/\s+/);
       };

Code like this makes me a little upset. It'll completely break on simple cases: `(print "To close the REPL, type (quit).")` being an obvious example.

I get that it's a toy example, but all too often it slides into tricking people that hard* problems are trivial.

* - I'm aware that tokenization has been a solved problem for decades—what I'm trying to say is that it takes quite a bit more thought than the four-line implementation suggests.

I thought the article did very well in providing an introductory resource to students hoping to learn more about the structure of an interpreter. On the other hand, if one wanted a nearly identical, and possibly more rigorous approach to learning about implementing a "tiny lisp" Peter Norvig's lispy implementation is a wonderful resource: http://norvig.com/lispy.html ... of course any computer scientist studying languages should also have read Mcarthy - http://www.cse.sc.edu/~mgv/csce330f13/micromanualLISP.pdf

> (first (1 2))

How does the interpreter know that the (1 2) part is a list, and not an application of 1 to 2?

    var tokenize = function(input) {
         return replace(/\(/g, ' ( ')
               .replace(/\)/g, ' ) ')
               .trim()
               .split(/\s+/);
       };

shouldn't it say return input.replace( ...

If anyone is looking for something a little more full-featured, BiwaScheme is a pretty decent R6RS Scheme interpreter in Javascript.

http://www.biwascheme.org/