[plt-scheme] 301.11

--- Greg Woodhouse <gregory.woodhouse at sbcglobal.net> wrote:

> 
> In a way, this is part of the motivation for my little evaluator (the
> big motivation, of course, is my own education). In a normal order
> language, where identity is \x . x, then this is a reduction you
> ought
> to be able to carry our right away (or so I'd think). But applicative
> order seems to wreak havoc with this kind of optimization, since in
> (\x . x) y you don't even look at \x . x until you've evaluated y.
> 

Come to think of it, though, you really may "only" care about an
s-expression being defined. In

(define x 'y)
((lambda (x) x) x)

there is no need to apply (lambda (x) x), just look up x to get y, so
maybe there is something that can be done.



===
Gregory Woodhouse  <gregory.woodhouse at sbcglobal.net>

"It is foolish to answer a question that
you do not understand."
--G. Polya ("How to Solve It")