On 16 Jul 1998 at 11:42:44 +0200, sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) wrote: Common Lisp is not inherently incompatible with tail recursion elimination. For example: - CLtL2 mentions the possibility on page 686: <URL:http://www.cs.cmu.edu/Groups/AI/html/cltl/clm/node227.html> - Harlequin's mostly-ANSI-compliant Liquid Common Lisp (formerly Lucid CL) does such elimination, as documented in their user manual: <URL:http://www.harlequin.com/education/books/lispworks/lcl/aug/aug-51.htm... - I'm also pretty certain (although not entirely positive) that the CLtL2-compliant-with-some-ANSI-extensions Macintosh Common Lisp (<URL:http://www.digitool.com/>) does tail recursion elimination on functions declared (speed 1) or better. At least I seem to remember that from my now somewhat distant MCL programming days although, admittedly, the system may have changed drastically in the 6 years since I last used it. :-) I have Allegro CL from Franz (<URL:http://www.franz.com/>) on my home machine, so I'll try compiling a tail-recursive procedure tonight and see if it applies this optimization. (BTW, those of you with Linux systems can try out Allegro CL for free. See the web page. Those with Windows systems might want to try out Harlequin's FreeLisp, <URL:http://www.harlequin.com/>;. Note that both come with an Emacs-like editor.) I once read something about why some Common Lisp implementations do not provide tail recursion elimination. I've been racking my brains to recall where I read it, but no luck so far. It was something to do with supporting catch and throw as I recall, but for the life of me, I can't imagine why that would be much of an obstacle. Surely you could just refuse to do TRE on a function containing a catch if you don't want to deal with analyzing how the stack might unwind. On the subject of call/cc, I also would like to thank Chipsy for the nice tutorial. I would like one more piece of information, though. How does call/cc relate to a lexical closure? -- Jerry James Email: jerry(a)cs.ucsb.edu WWW: http://www.cs.ucsb.edu/~jerry/