My Resignation from Emacs Development | Hacker News Reader

My Resignation from Emacs Development

lists.gnu.org

77 pointsjordigh65 comments

Emacs is in the process of moving from legacy languages modes using regexps and elisp for syntax analysis to new modes using tree sitter.

In this context, what does a name like "c-mode" should mean? Options: 1) it should stick to the old mode, cc-mode here. To use the new mode, use explicitly c-ts-mode; 2) it should move to the new tree sitter mode, c-ts-mode. To use the old mode, use explicitly cc-mode; 3) it should mean the new preferred Emacs mode, with a way for the user to take back control if they have a different preference. This preferred mode will change at some point from legacy to tree sitter.

The change is (3), with a move to tree sitter in Emacs 30 (to be released soon) IIUC. It makes sense to me. Saying that anyone own a name as generic as "c-mode" in an open source project just because they're first and have a long history as a contributor (thanks by the way!) seems excessive. Change of default is normal in an evolving project, and as long as it's clearly documented with a way to override (which is the case IIUC) it's fine to me. One can dislike the change, but it's impossible to please everyone anyway. Emacs users are used to adjust configuration based on their preferences.

I understand it can be an emotional situation for the maintainer of the legacy mode. But I don't see the need to call foul play.

And indeed, if anything, a project like emacs being unable to make a decision like this results in a project that slowly dies from the weight of its own history.

Tree-sitter is fairly universally understood now to be "the future." While cc-mode will likely have its place for a long time (hard to beat regexes on speed, even if they break down when the input is too noisy), moving the default to the tree-sitter implementation aligns with the other language modes going to tree-sitter. For good or ill, consistency is almost certainly better than new users having to learn "Your code is parsed by tree-sitter. Oh, except your C and C++ code, unless you set this flag, because Mackenzie threw a fit in 2024. That's a fun bit of history you get to care about forever now as a user!"

I agree with giving users control, but unfortunately I cannot agree with the move to c-ts-mode. And I cannot disagree more with associating CC mode with "legacy" when it's objectively better than the other alternative, at least currently. I don't think Emacs developers are doing users a favor in this specific case.

CC Mode is extremely capable. Over the years it has developed to such a maturity that almost all needs can be satisfied, and performance has never been a problem for me. It contains very few, if any, bugs, that affect my use.

On the other hand, the tree-sitter major modes are not at al production-ready to be considered as default. For one thing, the whole highlighting can break for complex macros and ifdefs. (I'd be glad to be enlightened whether it's theoretically possible to fix at all -- can you correctly highlight ifdefs without doing semantic analysis with the help of a compiler?) For another, CC mode has a feature called c-guess that can quickly analyze an existing source buffer and generate a format definition which proves extremely valuable. Alas, c-ts-mode has zero support for it.

I had high hopes for tree-sitter. I turned on tree-sitter modes for all my coding when it was out, and now I have zero enabled. They still have a long way to go and I don't want to spend time debugging emacs code at work. :-)

Tree-sitter is not a panacea. Fast parsing alone is not what makes a good major mode.

As someone whose pronouns are C-programmer/vim, I feel unsafe.

My living nightmare would be to develop highly verbose Java programs in an editor with 999 gorillion different "modes" with seemingly random names.

"oh, you're making an singletonfactoryfacade in Treesat-19 mode, you'll need to be using CCC-mode, treesat-19 mode is for factoryencapsulationfactory patterns"

patch: https://lists.gnu.org/archive/html/bug-gnu-emacs/2024-02/msg...

I think it makes emacs prefer the "new" Treesitter based c/c++ modes ahead of the venerable c/c++-mode that is older than most posters here, maintained by the email author.

Is there no code review for making changes? And that patch is from feburary, why is this email drama happening only now? Genuine questions, I don't know much about emacs/gnu project development.

Yeah, a basic "code ownership" check before merge would have been enough.

It reads like this is more of a bio problem than a digital problem.

That would require the use of a 21st-century development workflow.

you've misunderstood, no one is complaining about a lack of technical controls.

Minor changes aren't reviewed very much, major changes are typically reviewed when someone posts them on the list before committing them. The author is complaining that this is a major change, it's complicated his package (a heavily used one), and the problems haven't been resolved after many months.

Enacs-lisp relies on disciplined use of a single global namespace, with extremely long names. Using names from a part of the namespace that "belongs" to someone else is rude and IMO bad.

Oh, so that's why this was so ragequit-worthy: the global elisp namespace is delicate, and unilateral changes like this without consulting the maintainer of the stomped-over package is a big no-no.

Here is some insight as to the timing:

https://lists.gnu.org/archive/html/bug-gnu-emacs/2024-11/msg...

The timing and also the root issue: Alan wants the `c-mode` symbol to mean "The mode I maintain, named 'CC-mode". The (dare I say) community is moving in the direction of it meaning "The mode users want to use for C (and C++ and C-like) files" (note: IIUC, this has outsized impact on the entire emacs ecosystem because a lot of modes for c-like languages are built atop the user's c-mode to save a lot of repetition, so the c-mode is both a major mode in its own right and the support framework for several other modes).

Ultimately, "name ownership" isn't really a thing, it's going to be a lot more convenient to change one symbol's semantic meaning than to require an unknown number of additional dependent projects to remap to some other symbol, and while I respect Mr. Mackenzie's frustration, one's "ownership" is always limited when one works on free software collaboratively with others.

These names are not internal names, like identifiers in the C source code of vim. The names in question appear in lot's of users Emacs configuration files. And Emacs has been extremely reluctant to change the meaning of those names in the past. Doing it here contradicts an important expectations of long time users. This is not only about "name ownership" of some mode's maintainer.

Unless I misunderstand, the idea is that treesitter C-mode is supposed to be pretty-much a replacement for traditional c-mode, right? As in "In circumstances where I want to invoke the C-major-mode, I now want to invoke the one that uses treesitter?"

If so, shifting the interpretation of the `c-mode` symbol seems the right avenue to accomplish that goal.

No, Emacs has lots and lots of config variables and functions which depend on the exact mode used. So, if you switch the major mode, you also have to switch to other config variables and functions. And as far as I see it, c-ts-mode is far from being a drop in replacement for everyone that currently uses c-mode.

> Unless I misunderstand, the idea is that treesitter C-mode is supposed to be pretty-much a replacement for traditional c-mode, right?

you're very much mistaken.

Head of that (very long) thread containing an objection from third party Eli to code committed by the person resigning:

https://lists.gnu.org/archive/html/bug-gnu-emacs/2024-11/msg...

I have briefly tried c++-ts-mode and it majorly clashes with clang driven indentation as provided by lsp modes (no, I'm not writing a TS config to match clang-format). I'm back to c++-mode.

I'd say the more sensible approach is to use `aphelia` and have it invoke the real life code formatter.

Seems interesting, but I like to run the formatter after (almost) every keystroke.

I hate seeing emacs and gnu people fighting like this.

A flame war written in unformatted plaintext on a public list serve actually makes me nostalgic for simpler times.

>A flame war written in unformatted plaintext on a public list serve actually makes me nostalgic for simpler times.

As a vi user I'll avoid taking sides on an issue I know nothing about, but when the author stated "...there is also a convention of treating eachother with respect on the mailing lists. Sadly this convention is superficial, and seems only to mean things like not using swear words..." it hit me right in the gut.

It's an issue in many OS projects. Heck, it's a problem in the larger policy world -- a similar sentiment is why I stopped doing technology policy, a childhood dream of mine.

There are two kinds of respect: "Treat me like an equal" and "Treat me like I have twenty years of experience and you don't."

These are not the same concept, and it's easy for people to use the same word and fail to realize they aren't the same semantics.

I wouldn’t say those are mutually exclusive.

All humans deserve respect, but a lack of respect for expertise - particularly scientific - is a major problem globally and why our “intelligent” species isn’t doing anything about climate change.

Oh, policy has always been a blood sport. You need a taste for conflict to enter the arena in the first place.

(Been working on tech policy for a very long time, with scars to prove it!)

Long-time contributors are so valuable to a project. I know that in some cases progress can be impeded by entrenched thinking from entrenched project members, but I think that for a FOSS project like Emacs they're the most valuable asset, above users, code, mindshare, features.

You can't let someone dictate direction just because they've been around a long time, or because they've sent a lot of patches. You also can't let someone have Wikipedia-editor-like feudal authority over some area just because they feel like they own it. I get that. Outside these cases though I think whatever can be done to keep core maintainers happy and on-board is probably best for the project and best for the ecosystem.

> Try using "Emacs" or even "free software" to mean something different, and see just how quickly you would hear back from Richard Stallman.

What would Stallman do is not the rubric I use for, well, anything really... But I can see how somebody who feels that way would be willing to cease to provide free labor because they believe they have lost control of something on account of it becoming too popular for them to maintain fiat authority over its use.

Fights between individual contributors can be so strange. I find that they frequently don't actually address the underlying issue because the underlying issue can be arbitrary and somewhat petty, and engineers don't want to think of themselves as either of those things. I once got into a long argument about folder nesting depth with a co-worker only to finally discover that the underlying issue was that I was using vs code (which tacitly penalizes you the way the UI works for shallow nesting) And he was using vim (and his configuration worked better if all of the relevant files were at the same level in the file system).

Once I realized the issue, I deferred to him on seniority and went about retooling my process to match to the existing process. But it took us a while to realize that was the underlying issue!

An unrelated project that has dealt with the problem at issue is Debian.

https://wiki.debian.org/DebianAlternatives

I understand the purpose of `major-mode-remap-alist' to be setting up a similar alternatives mechanism. If you have defaults, you have to have a winner.

The legitimate complaint is that there is a lot of code that expects `c-mode' to bring up cc-mode and builds on that, that might get loaded as a hook on `c-mode', and this will break if that function changes. It is also legitimate to expect a little discussion when something that might make your code appear obsolete is merged. When cc-mode was the default, c-ts-mode was the alternative, and you probably want a little bit of ceremony when making the switch.

The annoying part is viewing ownership of the `c-mode' function as somewhat absolute because you got there first. But when you have file variables that say, "mode: c", for c-like code, but you use c-ts-mode for editing c-like stuff, you really want it to load c-ts-mode and not cc-mode.

Also annoying is passive-aggressively putting in a change that messes up the mechanism "to spur discussion". That doesn't seem very collegial.

I'm just remarking on my understanding of the discussion. I'm not an Emacs maintainer by any means, just a user.

seems like a human issue (based on https://lists.gnu.org/archive/html/emacs-devel/2024-11/msg00...) and a bit of process one (too easy to step on someone's shoe in emacs-devel ?)

wonder what's a good course of action to restore peace and good spirit here

This is a little, uh, I don't know a polite way to say it, but is the author ok?

If you contribute code to a free software project, people are going to use, extend, or modify it. They don't have an obligation to ask permission or even inform you. That is kind of the point.

I agree that the change is maybe not the most well thought out and introducing ambiguity in the meaning of a symbol is probably bad software engineering in the long term. But to call it "an act of aggression" is unhinged.

In meme form:

> contributes code to a free software project

> someone else touches that code

> suprised pikachu face

> people are going to use, extend, or modify it. They don't have an obligation to ask permission or even inform you. That is kind of the point.

That's all well and good, but the objection doesn't relate to someone exercising their software freedoms, it relates to the effects on the repository. If this happened in a fork then nobody could say anything, but if it happens upstream then it messes with the maintenance and development responsibilities of previously borne by Mackenzie.

It also messes with config files of lot's of users.

You can't just send a pull request to Linus replacing ioctl with some function of you own writing, certainly not when it breaks other functionality. In meme form: one simply does not walk into Mordor.

And the author seems ok to me, just pissed off.

Agreed. This read to me like the person had a pre-existing axe to grind.

They clearly had pre-existing beef, but also, you don't just break a 20-year-old important plugin without even telling the author. Especially if the guy who wrote it is on the short-list of core contributors with you. It makes sense that Alan feels totally disrespected.

I'd say it's the "straw that broke the camel's back," but this is less a straw and more a cinder block.

> Contributes code for 20 years.

> Somebody else makes breaking change to 20 years of work

> Project loses experienced maintainer of core feature.

> Surprised pikachu

Yes, with the added detail that the change was accepted to main and the author of c-mode feels like everyone was consulted but him, when the change is "replace c-mode."

I can definitely see how one feels pushed-out in that context.

what a weird comment to make. the author is upset that the project that they have worked on, for free, for decades, and has been a cornerstone of emacs as a programmer's editor, was unilaterally deprecated by an Emacs maintainer without any discussion.

Reply on news.ycombinator.com