Re: FOSDEM 2019

[Amirouche Boubekki <amirouche.boubekki@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 2275 bytes --]

Hello all,

Le mer. 6 févr. 2019 à 18:47, Ludovic Courtès <ludo@gnu.org> a écrit :

> Nala Ginrut <nalaginrut@gmail.com> skribis:
>
> > Does it mean someone is working on writing Racket specific dialect
> > front-end on Guile? I appreciate!
>
> Nope!  Actually Chris Webber raised the issue of collaboration between
> Racket and Guile in their talk, which is why we’re discussing it.
>
> But Guile is still Guile and there’s also lots of Guile hacking waiting
> for us.
>

I actually meant to work on the racket front-end.

Skimmed through their reference API and
nothing stands out. I known that I don't know.
Perfection is in the detail. Thad said some particular
topic attracted my attention:

- Phases and compilation guarantees [1]
- Continuations Frames and Mark [2]
- Custodians [3]

[1]
https://docs.racket-lang.org/reference/eval-model.html#%28part._module-eval-model%29
[2]
https://docs.racket-lang.org/reference/eval-model.html#%28part._mark-model%29
[3]
https://docs.racket-lang.org/reference/eval-model.html#%28part._custodian-model%29

TIL it is possible for me to create a front-end in guile.
There is various example in guile tree.
But again perfection is the detail.

Also, I would like to point out that there are already
racket things available in GNU Guile:

- for-loops https://bitbucket.org/bjoli/guile-for-loops
- syntax-parse https://gitlab.com/guile-syntax-parse/guile-syntax-parse

Last but not least guile fibers https://github.com/wingo/fibers
which rather unique in scheme universe at least.

I was interested in this racket-on-guile because it seems difficult,
also there was potential to make a contribution that is useful,
and learn a few things about Guile inner working.

At the end of the day it's not because we can run racket
programs that GNU Guile will suddenly become the place
to be. I don't say it is a useless project, working code is always
great. But it is not enough. See for instance, PyPy has pratically
catched up CPython and has better performance, it still not
mainstream.

Enough said ;-)

I will look at r7rs branch [a] and pdf report [b]

[a] http://git.savannah.gnu.org/cgit/guile.git/log/?h=r7rs-wip
[b] https://small.r7rs.org/attachment/r7rs.pdf

[-- Attachment #2: Type: text/html, Size: 4017 bytes --]