Re: [stack] Why! a concatenative language ?

chris glur <[email protected]> wrote:
> I've been lurking and waiting/expecting the answer to this question
> to become obvious  -- so far nothing.

To the best of my knowledge, the answer to that question is not
obvious to any of us. I think that if it can be answered, it will take
some real work.

> Via a forgotten tortuous search for means which could reduce software
> errors via some formal methods, perhaps transforming conventional
> imperative style syntax, I came upon 'Joy'.

Welcome in!

> Q -Can concatenative languages substantially help in building
>    'program correctness' ?   If not, what do they offer that's special ?

They can because they offer (at least) one thing special: they allow
you to build small units that are correct programs, then join those
units together and know that the result is also a correct program.

This is not a magic formula, though -- "a correct program" is not
automatically "the program you meant to write". Other methods are
needed, and because the idea of a concatenative language is so new, we
don't yet know what they'll be.

> I'm also investigating:
>   http://www.rebelscience.org/Cosas/components.htm
> # COSA is based on the premise that the
> # primary  reason  that  computer programs are unreliable is the
> # age-old  practice of using the algorithm as the basis of software
> # construction.   Switch  to  a  synchronous,  signal-based  model
> # and the problem will  disappear.

> Perhaps COSA  too is a futile investigation ?

I doubt it's futile; but I'm pretty sure that the description they're
giving is overblown. I don't see how you can get around the fact that
computers are extremely literal creatures, and programmers can't keep
track of everything the computer needs to be told at the same time. As
long as those things are true, I doubt that "the problem will
disappear."

Perhaps COSA will help, though. I'll read up on it; thank you for the link.

> == Chris Glur.

-Billy