Hacker News new | ask | show | jobs
by nagvx 3350 days ago
Has there been any update to the status of GuileEmacs? I have read that it aims to be "the future of Emacs", but I rarely see mention of Guile from the Emacs community.
4 comments

It isn't the Emacs devs who are really asking for GuileEmacs. Rather, it’s Guile fans who are proposing to integrate a software they love into Emacs for mutual benefit. While the Emacs devs don't a priori reject such a thing, they naturally insist that the Guile-based Emacs be a mature, stable and faster piece of software before it is considered for the mainline, and that just hasn’t happened because only a handful of people are working on it.
This is the branch to watch for guile-elisp: http://git.savannah.gnu.org/cgit/guile.git/log/?h=wip-elisp

Last thing I heard was that Guile-Elisp (and by consequence Guile-Emacs) is working, but very slow. That can be fixed but requires work. Also, Emacs internally uses a different string representation from Guile. This makes the transition between languages tough and some more glue is required.

IMHO Guile-Emacs is "nearly there". I.e. with just a little more work it could have enough momentum and attract developers to be on the trajectory of becoming the default elisp implementation in the long term.

Having the new Guile 2.2 VM drive Emacs would definitely improve things. And the C-codebase could become considerably smaller and cleaner.

Without portability there will be no GuileEmacs to replace the official Emacs implementation. AFAIK, Guile doesn't work on Windows.

Until Guile will work natively on all major operating systems, I see no future for GuileEmacs.

Guile community probably wants guileemacs to be the future, but I think it would be only natural for emacs community to think otherwise.
I'm part of the emacs community, and not part of the Guile community (though I am a fan of Scheme, just not Guile in particular, Chicken Scheme is more my speed) and I think a Guile-based emacs would be a huge step forward, and can't wait for it to Guile to be fully integrated and all the outstanding issues to be ironed out.

I really can't understand the objections from some people in Emacs land, except for those that would prefer a Common Lisp-based Emacs, but that's not what we've got. We've got Guile, and while that might not be as great as Common Lisp (in their eyes, not mine, for me Scheme is preferable), it's still a lot better than elisp. Rejecting Guile and sticking with elisp just makes no sense to me at all.

Keep in mind that having emacs be based on Guile does not mean that all the elisp emacs packages have to be jettisoned. They will still run as elisp under Guile, and you can continue writing scripts in elisp and have them continue to run under a Guile-based emacs.