I got a lot of interesting comments from that article. Someone pointed me to this brilliantly evil implementation which can be seen at https://gist.github.com/1679908 (originally from http://www.pick.ucam.org/~ptc24/yvfc.html, which seems to be down). There's also Norvig's implementation, which is much cleaner, at http://norvig.com/lispy.html, and a Lisp to Python compiler at http://bernhardkausler.wordpress.com/2009/11/28/sinc-%e2%80%....