[racket] member et al.

On Thu, Nov 11, 2010 at 10:57 AM, Jay McCarthy <jay.mccarthy at gmail.com> wrote:
> This 'if' introduces a stack frame that is not otherwise needed.
>
> Jay

OK, I see your point now.  I was thinking of it as one more test at
the end so I wasn't thinking of it as having a real cost, but I see
how this makes the fold-like implementation of andmap into something
that grows the stack.

Is Typed Racket able to observe that if the arguments to and/or are
all Booleans, then the result will definitely be a Boolean?  If so,
then and/or's flexible behavior in no way diminishes the value of
having "true predicates" available to use when you want to express
that clear intent.