That’s the adage oft repeated by people who had had a cursory look at Emacs, but haven’t been really using it. In reality, the opposite is true. Emacs is a great editor, but the operating system that it is [0] is lacking in many ways:
- performance is not great (although it’s getting better with the advent of native compilation in Emacs 28)
- the system language is awkward (no namespaces, dynamic scoping by default, ...)
- no preemptive multithreading
- etc.
On the plus side, it is auto-documenting and extremely malleable.
> the system language is awkward (no namespaces, dynamic scoping by default, ...)
RMS's dislike of Common Lisp really caused a colossal amount of damage. Imagine an Emacs written in Common Lisp instead of Emacs Lisp, with decades and decades of improvements. Having written a reasonable amount of software in both, Common Lisp Emacs would be very, very preferable, in part for the reasons you list and in part for others (e.g. a full-fledged, built in object system and improved extensibility).
> RMS's dislike of Common Lisp really caused a colossal amount of damage. Imagine an Emacs written in Common Lisp instead of Emacs Lisp, with decades and decades of improvements.
If Emacs were written in Common Lisp, only people on big mainframes would have been able to use it in 1985, and it probably wouldn't have become popular in the first place. Also, implementing Common Lisp would have taken much longer. RMS has said that his main goal with his LISP dialect was to keep the programming language as small and performant as possible, since at that time, machines had maybe 1MB of RAM (if you were lucky) and no virtual memory. This is why for instance he decided against lexical scoping (and still, Emacs was known back then as "eight megabytes and constantly swapping").
I'm very interested in guile-emacs, not because I think it'll replace main emacs, but I do think it'd be really neat to have my config in a 'cleaner' lisp like scheme, a clojure like language, lokke[0], or even python[1]
This is not much of a problem in practice these days. As you probably know it's trivial to enable lexical-binding for an entire source file and many (most?) modern packages do just that. It's even on by default in the scratch buffer. Emacs is just more user friendly than most other systems when when it comes to backwards compatibility.
I am in process of switching from emacs to neovim atm (doom evil user before). Whatever I do in emacs (emacs28, jit-compilation, M1) I still experience some UI lags l cannot get with profiler. I suspect slow lsp interface and suboptimal caching/gc. At some point I've tried neovim and was just surprised by how fast it feels.
Company mode is the big offender for me, it handles very large completion lists poorly. I'm not yet fully satisfied with the latency situation in my emacs config, but for me there is no going back. Evil mode let me keep almost all of the muscle memory I ever learned in vim/neovim, and the rest of the emacs ecosystem more than makes up for the latency issues in my opinion.
I've also experienced some heavy lag immediately after startup - maybe warming up the caches? Had a feeling it gets worse after each package update with compilation.
Long line performance is still a huge problem. Emacs team does an amazing job keeping up, but the age shows itself everywhere. In the end I've realized that org/magit were not life-changing for me, evil-mode it the only thing I care about. :)
Honestly this is a big thing that emacs folks don't value, multi threaded performance.
I don't even use LSP because I have my own ways of browsing source code through emacs. So I go off and tout emacs' amazingness, but then when someone tries it and attempts to use what worked in other editors, it's a slow laggy mess.
Emacs is a reflection or a rhyme of a great operating system; it has the critical feature of total inspectability/hackability, but the implementation is poor. There is no other system in use today that has that kind of total transparency all the way to the bottom, or at least, as far down as Emacs does.
I don't think people should try to use Emacs to build an OS on top of, but building an OS without understanding why Emacs survived is going to just end us up in the same world of inflexible, brittle, clunky software we're in now.
Doesn't everyone need geo-locate and map coordinates while simultaneously editing a config file? You're on hackernews, people will come out of the woodwork to defend this stuff.
