requiring proper tail recursion

[Morry Katz <katz@quilty.Stanford.EDU>]

     When a procedure p1 is to be called from a point within the definition of
     a procedure p2 that has been called with continuation k, and when the only
     observable actions or effects of p2 that remain to be accomplished after
     execution of p1 is to yield control directly to k, passing it exactly the
     value(s) yielded by p1, then the point of call of p1 is said to be a
     <I>tail position</I> within p2, continuation k itself should be used as the
     continuation of p1, and the call to p1 is said to be a <I>tail call</I>.

I believe that only famous authors are allowed to write paragraphs
composeed of a single sentence.  :-)  I like the content, but suggest
that we need to find a way to break it into a couple of sentences
before we publish it.
--------------------
Morry Katz
katz@cs.stanford.edu
(415)723-9510 (office)
(415)694-9121 (pager)
--------------------