Subject: If then else and unevaluated conditionals
From: Robert Dodier
Date: Mon, 29 Jan 2007 21:56:54 -0700
On 1/29/07, Stavros Macrakis <macrakis at alum.mit.edu> wrote:
> On 1/29/07, Robert Dodier <robert.dodier at gmail.com> wrote:
> > The motivation for unevaluated if--then--else is that I want naive stuff like
> > plot2d(if x > 0 then 1 else 0, ...) and integrate(if x > 0 then 1 else 0, ...)
> > and (if x > 0 then 1 else 0)^3, etc, to just work.
>
> I do understand your point, and sympathize, but I just don't see how
> to make this work in general. Painful experience makes me very wary
> of solutions which solve the simple case, but make other cases much
> messier.
Well, I can guess at your objections but it would be simpler if you
would just tell me.
Maybe if you have time you can look at rtest_boolsimp.mac and
say which cases should turn out differently. Or glance at the
comment header of boolsimp.lisp and say which policies are
mistaken. I've taken your comments into account before; I'm
certainly prepared to do so again.
best
Robert