News for hackers. Stuff that matters.
Re:A curmudgeon tries a language server by dieggsy on Wednesday August 26, 2026 @11:59AM (#49451384)
I don't object to any of this. "Restart this system from initial state" is quite a bit more common than people imagine in CL. I was hinting at it perhaps being even more common than image-based development, though don't quote me on that. I personally don't even really create Lisp images in dev (save for creating executables that I don't then further modify, which are technically images in SBCL).

Incremental runtime modification is great for development, but I very frequently restart (from sources) every so often, and certainly at least before deployment to verify everything is working as expected. Sometimes because it's completely necessary, sometimes for my own sanity.

For production, I think it's a very "neat" thing to be able to do and I've heard some great stories (no personal experience), but it's definitely a foot gun if not used very carefully.

Re:A curmudgeon tries a language server by sroerick on Thursday August 27, 2026 @04:04PM (#49470456)
Can you explain how you sync back to code, in practice?

In other words - if you modify the run time, are you then able to mirror those changes in your codebase and then reboot?

Re:A curmudgeon tries a language server by dieggsy on Thursday August 27, 2026 @08:26PM (#49472967)
tl;dr you just don't really do that.

99% of the time, you simply write the code or change in the sources first, then compile/run it in the runtime.

If you're modifying the runtime directly at the REPL without having sources, good luck? or you don't really care about persisting those changes. CL has pretty good introspection, so you could look at the source expression of a function defined at the REPL for example, if you really wanted to recover something (and happened to put the change in a function first), but this isn't always reliable.