Re: Guile Steel: a proposal for a systems lisp

[Damien Mattei <damien.mattei@gmail.com>]

On Sun, Aug 7, 2022 at 10:21 PM Christine Lemmer-Webber <
cwebber@dustycloud.org> wrote:

>
> Damien Mattei <damien.mattei@gmail.com> writes:
>
> > On Sun, Aug 7, 2022 at 2:44 PM Christine Lemmer-Webber <
> cwebber@dustycloud.org> wrote:
> >
> >  Bigloo is cool.  I think it falls under the same category of Chicken,
>
>
> > perhaps i'm wrong, but Chicken seems to have strict types as options:
> > http://wiki.call-cc.org/man/5/Types
>
> Interesting.  Any examples of it in action, personal use experience,
> etc?  I wonder how it handles the famous challenges of keeping that
> safety between both statically typed and dynamically typed parts, which
> turned out to be extremely expensive, far more than dynamic scoping,
> in Racket: https://www2.ccs.neu.edu/racket/pubs/popl16-tfgnvf.pdf
> (though there is work being done to make it better, hopefully...
> it may have progressed a lot since I last looked, probably has.)
>

i'm sorry but i almost never used Chicken...

>
> > about Bigloo the best answer would be to ask to author himself but i
> > used Bigloo a little to produce JVM class files only, not C obj files,
> > and it can be declared some statically typed variable i think in both
> > C and Java targets:
> > see:
> > https://www-sop.inria.fr/mimosa/fp/Bigloo/manual.html#G82
> > and mostly:
> > https://www-sop.inria.fr/mimosa/fp/Bigloo/manual-chapter25.html
> > Regards,
> > Damien
>
> Hm... "Warning: All type annotations are ignored by the interpreter."
>
> i suppose in the interpreter, code being not compiled, Bigloo act as a
> normal Scheme REPL and typing is ignored, but in compiled modules it works
> well.
>