Re: [stack] Why is point-free form so interesting?

On 7/9/07, John Carter <[email protected]> wrote:
> On Sat, 16 Jun 2007, Christopher Diggins wrote:
>
> > I have been struggling to formalize/rationalize my intuition that
> > point-free form is somehow better than a form with names from the
> > standpoint of tools (e.g. analyzers, compilers, translators,
> > optimizers, etc.).
>
> Probably because the pros and cons are symmetrical...
>
> ie.
>
> The Good thing about point free form is the mechanism for getting from
> parameters to using them in expressions is explicit.
>
> ie. You know exactly how the magic happens.
>
> The Bad thing about point free form is the mechanism for getting from
> parameters to using them in expressions is explicit.
>
> ie. You know exactly how the magic happens....but you're almost sure
> you don't care as any other mechanism should still give the same
> answer and it's presence is just noise not signal.
>
> As usual Christopher is asking a valuable question...

Thank you.

> ...perhaps we can rephrase it...
>
> Is there a representation of a program where the mechanism of moving
> parameters to values in expressions is...
> * Explicit (ie. Can be reasoned about / transparent on inlining)
> * Trivial (Adds almost nothing to the code / vanishes in this representation)?

I like this a lot. This does a good job of summing up what I like
about point-free form very elegantly. One point though, is "explicit"
the correct term? I would normally consider argument passing in a
stack-language to be implicit, rather than explicit. However I agree
that it is obvious and unambiguous.

Some other bullet points concerning what I like about point-free form
(I've just brainstormed these and expect that they can be refined):
- data transfer is obvious
- data transfer is automatic
- function forms are uniform (stack -> stack)
- order of evaluation is simple and unambiguous: left to right (no
precedence rules)
- subterms can be evaluated concurrently

Cheers,
Christopher