freenode/#stumpwm - IRC Chatlog
Search
12:21:58
gotwf
Hmmm.... looks like we get non parallel threads in emacs 26?? https://www.emacswiki.org/emacs/NoThreading
12:27:32
Formbi
maybe if people would start working on Guile Emacs again, we would get better multithreading
12:27:56
gotwf
Iirc, once upon a time threads were slated for emacs 23, then pushed out to 24, then seemed to have been +/- given up on/deprioritized yet further by Wiegley.
12:28:39
gotwf
Heh, and here's a thread from Ferrier going all the way back to 2003... https://lists.gnu.org/archive/html/emacs-devel/2003-11/msg00613.html
12:28:45
jasongrossman
I believe there's only a tiny number of people who currently understand the internals of Emacs well enough to work on this sort of thing. Of course that might change. But at the moment it seems very reasonable NOT to be trying to make Emacs fully parallel. After all it's got plenty of capacity for async processes.
12:29:19
jasongrossman
Maybe in the VERY long run it would be nice to make the gnarly bits of Emacs simpler.
12:29:53
jasongrossman
And there is some work on making it simpler ATM, I think. But there's a lot that's gnarly, especially I think in the display code.
12:34:43
gotwf
That said, fifteen plus years to be still talking about it equates to several eons in software development time, so it definitely is a drag that it has taken so long for such a significant project to get with the new millennium already.
12:38:41
jasongrossman
Oh, that's Emacs for you. It's VERY stable, in both bad ways and good ways. Not many platforms that are still widely used and that you can run the same macros on as you could decades ago.
12:44:19
gotwf
Here's Wiegley talking about guile and concurrency three years back. Is guile even still on the table? I just use emacs and don't even try keeping up with stuff like this https://lists.gnu.org/archive/html/emacs-devel/2015-10/msg00007.html
12:46:19
jasongrossman
I don't think it's *totally* clear what "on the table" means in this context, but I wouldn't hold your breath for guile Emacs (or for rust Emacs).
14:15:54
plathrop
Blackbeard: the code is stuffed full of my-personal-crap; if I get a chance though I'll clean it up and share.
16:45:39
Fade
it's kind of clear that elisp wants to be common lisp. guile is most definitively a scheme.
18:02:19
gotwf
As one who cannot code *lisp jack 'chit... would not common lisp be a most "portable" choice??
19:38:06
Fade
although the relationship between the scheme implementations and their spec seems a little loose.