Hacker News Reader

Comment by BeetleB | Hacker News Reader

I speak only for myself and not the Emacs team, but ... IMO it shouldn't do this by default - unless there is a clear "winner" among all the choices.

IIRC, for eglot I had to install a Python language server. There are a bunch of them. Why would I want it to pick one by default?

I don't know if Emacs will ever have an IDE with TS and LSP working "out of the box" without some configuration. From my experience with Eglot, if it were enabled by default, it would confuse the heck out of some users who just want to, say, edit a Python script.

There are always opinionated releases like Doom Emacs or Spacemacs that can set whatever nice defaults you think should be there. Leave vanilla Emacs to be, well, vanilla. Emacs, while it can be a decent IDE, is not an IDE and shouldn't default to being one.

View on HN

I recently switched from full IDE to Emacs (wrote up about the things I learned at https://github.com/KallDrexx/emacs-zero-to-ide-journey/blob/...)

Typescript is a perfect example of why Emacs shouldn't own that setup. Typescript version 6 uses the "typescript-language-server" which is a totally stand alone nom package for lsp support.

Typescript 7, is incompatible with typescript-language-server and the compiler itself now acts as a lsp.

If Emacs handles that all by default it then needs to handle versioning of each language server itself across all operating systems, handle cases when different machines have different npm version managers, upgrades, etc...

It's not impossible but it's a lot to ask, and it requires more frequent version upgrades just to keep things in sync.

I couldn't even use the built in version of eglot and had to use ELPA version for bug fixes that affect c# LSP support that were committed to the package back in December.

Emacs makes many opinionated choices about defaults; if it didn't, it would launch to little more than an elisp REPL by default.

What's important is that _you can change it_. Good defaults ensure that Emacs doesn't repel new users and casual users and users who are not interested in fiddling.

And this isn't a matter of color theme default, or font default; this is a shipped feature that _does not function_ without changes to the defaults! By default, it does not work.

> if it didn't, it would launch to little more than an elisp REPL by default.

I'm confused. Have things changed? The scratch buffer it starts with is for writing elisp.

> this is a shipped feature that _does not function_ without changes to the defaults!

Do you know how many shipped features come with Emacs that don't function without changes to defaults?

A huge number of them! Stuff like hippie-expand, ibuffer, etc. This isn't something new - it's been that way for decades.

> Good defaults ensure that Emacs doesn't repel new users and casual users and users who are not interested in fiddling.

This has been argued to death forever. While even I think they can and should change some defaults, the reality is that Emacs as an ecosystem is doing just fine without so many new/casual users, and will continue to do so. They don't want to be something for everyone.

And while I don't have the latest Emacs, in my experience with eglot - it's simply not ready for the masses. I frequently get timeouts, and would often have to shut down and restart eglot. It's been a while since I had to, so perhaps they've fixed all the issues, but it's simply not as robust as what you'd get with other IDEs. I'd rather they address those issues than the ease of use.

It launches with a scratch buffer, menu bar with populated menus, a functional mini buffer, and so on. All of this is opinionated defaults and configurable with elisp.

And yes, much of Emacs is broken out of the box and it's been that way a long while. That's not a good thing.

Emacs isn't ready for the masses. Broken defaults, and a brittle experience once features are made working is standard.

I've used Emacs for thirty years now. I don't recommend it to new users.

> Emacs isn't ready for the masses.

It never will be. It's simply not a goal they have. It's tiring to keep hearing people point it out. Emacs is also not ready to be your web server. Or your Jira replacement. It's not a lot of things - we don't need to itemize them all.

I've written a lot of stuff. Some of it I released publicly. None of it is for the masses. There are better tools for the masses - let them use those!

> I don't recommend it to new users.

"New" to what? In general, I don't recommend Emacs unless someone shows genuine curiosity. I always tell people "Give me any task, and I'll find a better tool than Emacs." And "If you want to get up and running quickly for your task, Emacs is not a good choice." Followed by "What makes Emacs worthwhile is you get it all in one package where they can interact with one another." And "It's a lifelong journey."

Nevertheless, some not technically savvy users do use Emacs today because they needed help editing files on a remote machine and I couldn't teach vim because Emacs was all I knew :-)

> I always tell people "Give me any task, and I'll find a better tool than Emacs." And "If you want to get up and running quickly for your task, Emacs is not a good choice." Followed by "What makes Emacs worthwhile is you get it all in one package where they can interact with one another." And "It's a lifelong journey."

That is the proper response. These tools are designed for someone that does need them as they are, not for someone that just need some tools. Software like vim and emacs are in fact a whole workshop that you need to make your own. If you just want a tool, there are better software out there, but people that usually go for emacs wants a workspace, not just one tool.

The best tool is the one that makes completing tasks better: faster, easier, cheaper, or superior.

It takes significant usage and training to overcome the overhead that Emacs imposes. There are other tools that are fast, free, and capable which do not have this problem.

Even tools which are reasonably extensible. Like VSCode. There's good reason that all the new agentic development environments aren't forks of Emacs or NeoVim.

> There are other tools that are fast, free, and capable which do not have this problem.

You are merely mirroring what I said.

> There's good reason that all the new agentic development environments aren't forks of Emacs or NeoVim.

And there's nothing wrong with the fact that Emacs will never be the tool people fork for agentic (or IDE, or for any other purpose). That's not a mark against Emacs.

Emacs isn't the best code editor. It isn't the best note taker. It isn't the best document writer. It isn't even a good lisp.

What is Emacs the best at these days?

Why are you asking the person who said Emacs isn't the best at any particular thing?

If you want to engage, at least read the comments.

Because it's the question that I'm interested in. I use Emacs still because it's astoundingly mediocre at all things that I use it for; but always mediocre and never terrible. I would like to know if there's something it's the best at, besides being predictably mediocre.

I think eglot often just works now. I recently started doing c++ stuff, and it just detected clangd and I was off to the races with no config. Same with rust-analyzer for rust projects.

Reply on news.ycombinator.com