[sachdevap]

> I guess most of the core emacs devs are greybeards who started programming in the 80s or earlier, so what happens when they grow too old? Who will replace them?

What makes you say that? I don't think this is true. Not among the package maintainers anyways, and also of note here is remacs - I can't imagine greybeards programming in rust.

And popularity does not always mean more contributors, not when they are just the consumers of the product. The drop off in delta contributors / delta consumers is fast.

[serf]

>What makes you say that? I don't think this is true.

the same thing that probably made you say this :

>I can't imagine greybeards programming in rust.

emacs is an older software package, from the 'original home of hackers', extended in a (sorta) language that is notoriously connected to old-school AI and academic programming. The holy war of vi vs. emacs has raged for decades -- and emacs/vi/lisp jokes are some of the oldest computer geek jokes in existence.

I can imagine the fans skew older then say fans of something like Electron or Flutter.

The only computer-geek-joke-theme that I can think of that'd be older would be the billions of COBOL jokes -- but that's just because i'm too young to have heard the billions of punch-card and wire-routing jokes that inevitably existed before I was here.

[sachdevap]

>>What makes you say that? I don't think this is true.

>the same thing that probably made you say this :

>>I can't imagine greybeards programming in rust.

Guilty. I clearly had a blind spot there.

[erik_seaberg]

At the time Emacs was created, Lisp was one of the few languages that had automatic memory management and higher-order functions and error handling, but those have been lifted so often that they're now table stakes for modern languages.

I'm pushing 50 and excited about Rust because it's finally conceivable to have software that works reliably. I like Emacs, but it's not like I haven't seen my share of crashes in its kinda clunky C runtime. Hell, Rust even has macros.

[Torwald]

> it's finally conceivable to have software that works reliably.

Care to elaborate what you mean by that?

[erik_seaberg]

The industry has been trying to write C and then C++ for decades, but it all blows up randomly. Buffer overruns, use after free, shared mutable state. Apparently tooling exists that tries to make Rust-like guarantees, but it's so costly I've literally never seen it used.

[Torwald]

I don't get it. What about the JVM or ObjC or C#? How is it that only with Rust the things you stated are coming to us?

[erik_seaberg]

Much as Elisp is written in C, the JVM and CLR are each written in C++ and assembly, hosted on an OS written in C and assembly, and I've seen all of these crash.

[Torwald]

Thanks, I know see better what you mean. But, wasn't that achieved prior to Rust by realtime OSs like QNX?

[dmortin]

> What makes you say that? I don't think this is true. Not among the package maintainers anyways

I was talking about the "core developers" who work on the emacs core, not packages. AFAIK most of them has a long history with emacs, using it for decades, so they can't be that young.

But of course I know mostly about the maintaners and stuff. It's possible there are lots of younger contributors.

[rfrey]

> I can't imagine greybeards programming in rust.

Why not?

[Koshkin]

A lack of imagination?

[sachdevap]

Guilty. I clearly had a blind spot there.