Re: [stack] sweetening concatenative syntax

On Wed, Mar 19, 2008 at 3:09 PM, John Cowan <[email protected]> wrote:
> Christopher Diggins scripsit:
> > > > In other words you can't write code (in R5RS) that distinguishes
> > > > whether a given value is a closure, a function or a procedure.
> > >
> > > Closures satisfy PROCEDURE?, procedures satisfy PAIR? (and more specific
> > > predicates). Functions are abstracta, and don't have representations
> > > in Scheme.
> >
> > That is interesting. I can't find in the specification
> > (http://schemers.org/Documents/Standards/R5RS/r5rs.pdf) that closures
> > must not satisfy "PAIR?". Can you point it out?
>
> 3.2 says that PROCEDURE? and PAIR? are disjoint. It's true that in R5RS
> "procedure" is used instead of "closure", but in talking about Scheme
> I usually hear "closure" for the object, and "procedure" for the text.

And 4.1.4 is says that a lambda expression evaluates to a procedure.

- Christopher