I love the spirit of Zed. From the principles to the low-level implementation details, it all screams "good taste". It's immensely interesting as an object of study (the code is great, from GPUI all the way up).
Having said that, I don't think an editor should be VC backed. It's the obvious pragmatic choice to get a team together to support a thing, but I'm concerned by it.
There was a time around ST2 when it felt like everyone was using it and it could've become The Editor, then something happened and it's been left in the dust. I wasn't even aware but apparently even fourth version of ST was released, and that was in 2021.
I lost track of what happened there (moved to Vim back then), was it VSCode that killed it?
I've been a registered user of ST for long long time, and I thought if anything hurt them in the marketplace, it was taking several years off from the end of 2013 until late 2017 with hardly anything being released that opened the door to Atom and other editors to catch up.
Yeah, as others already mentioned, I think they sat on their laurels for a bit too long and let VSCode overtake it.
For what it's worth, I went from ST3 -> VSCode -> ST4, and have been happy since. I've found that I prefer my text editor with minimal extensions, and with Sublime Text's LSP Plugin, I'm pretty content. The performance and customizable UI make it more worth it to me than VSCode.
It's the LSP plugin that finally drove me to leave ST4 for Zed. Language integration is table stakes for an editor now. The fact that ST support is behind a volunteer plugin instead of integrated directly in the editor just means it's never going to be as good as a editor that does have first class support. The ST devs need to actually improve the editor, but I haven't seen any material updates in years.
I think it's less that they sat on their laurels and more that a team of 2 had trouble keeping up with the dozens of well-paid folks working on VSCode. Which suggests that perhaps a shareware model did not work out so well for them.
I don't think development actually stopped. ST3 was in a quiet public beta for a long time, but you can see builds of ST3 from 2013 to 2019: https://www.sublimetext.com/3.
Maybe they just wanted to do something else. Sometimes people just don't want to grind on endlessly for theoretically more money when what they've got already is enough for them.
I don't get this Sublime is dead nonsense. It's still being updated and works great. It's been my editor of choice for years and I happily pay for my license. I'd probably pay more if they asked me, it's tremendous value for money in my opinion.
I dislike "$x is dead" as much if not more as you do, and I'm sure it works fine as something doesn't need to be the most popular choice to be working.
With that being said, just a quick look at, for example Stack Overflow 2025 survey tells me it doesn't have the same mindshare it once had.
Eh.. I gave ST4 a go earlier this year and moved away from it due to plugins I wanted not being updated for years, and no longer working. That really feels on the cusp of being dead to me
There needs to be a critical mass of people using it for things that aren't core to stay updated
I don't run that many plugins to be fair, but the ones I do run (of which at least a couple are no longer maintained) works fine.
The key plugins I use are some LSP servers, and they work wonders. The few languages I mainly use (yaml, json, TS/JS, python and Go) I get great language support for via the LSP servers and the editor is blissfully fast always.
I could live without even the LSP stuff, but the one feature I can't live without is Sublime's excellent recovery support. Every once in a while my system will crash, and even though I've had multiple unsaved buffers Sublime recovers them every single time. Saved my butt more times than I want to know!
Yeah Atom and then VSCode killed it. Turns out being able to use JS to extend your editor is quite valuable. Essentially every JS devs have their own Emacs without having to learn Emacs and Lisp
Plugins were kind of it's selling point, yet it was pretty easy to mess it up with Plugins to the point of it being unusable - and not knowing what plug-in caused that.
The same curse emacs suffers from. What is the best sweet spot an editor/IDE has achieved to date?
I remember the extremes of the utter unconstrained chaos of Emacs and the rigid ultra-high-boilerplate approach of the Eclipse IDE. Emacs was fun to hack on, but impractical to use as an IDE, because if you installed enough plugins to make it useful as an IDE, it was broken half the time (my experience, many years ago.) Eclipse had a robust architecture, but writing plugins for it was a dispiriting slog, even when I got paid for it (again, my experience, many years ago.)
> What is the best sweet spot an editor/IDE has achieved to date?
Unironically, maybe VS Code.
Everything simple you can do with it, either comes built-in, or within Github/Microsoft ecosystem, or has an official plugin that gets recommended and featured by the editor itself. Plugins from individual hobbyist developers I have, I can almost count in one hand. (VSCodeVim being the most important one)
Now I compare this to my Neovim setup, and that one is basically running on charity from OSS developers.
What about writing a quick ad-hoc command? Something I would have found useful today, which I would have done in emacs fifteen years ago, was writing a command to parse a file in a log, generate a curl command from it, and copy the command to the clipboard. Could I do that in VSCode without creating an entire project?
I can't even start listing the issues with your hasty generalization here - I see outdated anecdotal evidence, survivorship bias, vague metrics, false correlation, goalposts moving. While your personal experience likely genuine, presenting it as evidence that Emacs is inherently impractical as an IDE only adds to the fallacy of generalizing from a single data point to universal truth.
I have completely opposite experience with [modern] Emacs. Of course, it wasn't smooth from the day one, but neither was my ride with different IDEs. Somehow, I keep coming back to Emacs because no IDE ever provided all the machinery I need to be productive. For me (and I suppose for many other people), Emacs is far more sweeter spot of an IDE than any other alternative.
VSCode copied most of the good features of ST and it is free and open source. Just that is enough to overtake it.
I still use it, it is maintained and it is very good and fast, and that it didn't try to reinvent itself is a good thing for me. But it is not a full IDE (not Jetbrains), it didn't jump on the AI bandwagon (not Cursor), and it is not free (not VSCode), so it is not surprising that it lost some market share. But it is not dead.
good question. I think the community fell off and many plugins were left unmaintained. I was using it for over a decade up until recently. ST4 had so many plugin issues and it stopped being worth manually fixing.
For me, it was that the maintainer of a language plugin I used was, um, challenging to work with. I wanted to contribute to add some much-requested functionality and he talked to me like I was a toddler and warned me not to waste his time.
Yes, VSCode killed it, because VSCode was free. Which is kind of sad because ST2 is actually noticeably faster than VSCode. Someone mentioned Atom, but that was never really a contender, not many people used it.
Personally at that time (circa 2013 I think) I wasn't using it because it lacked integrated features like debugger or good autocomplete. I was using a specific editor, but one editor per language (java = eclipse, C++ = QTCreator, C = geany). I feel like there wasn't a true "one size fits all" editor (except maybe Vim, but it felt so... unfriendly?). Also, I'm not sure it was available on Linux (don't quote me on that)
I love ST and don't get why it stopped taking off. VSCode is way too heavyweight for just editing text and TextEdit is more like WordPad on Windows than Notepad.
People refer to it as VSCode, but do people really use VSCode or VSCodium? I personally use VSCodium. I think that is the most de-telemetry'd version of it, I think?
Not sure if VSCodium and others can use all the extensions, especially Microsoft ones like Copilot. I remember setting up something like that once and had to do a lot of fiddling to be able to use all the extensions I wanted. Gave up on it at some point.
As for people in general, safe bet in almost any topic is that most people won’t care. :) So yes most people use VSCode.
They didn’t “solve” it, otherwise it would be a thriving editor that everyone would be using.
In reality 70% of the people I see are using Cursor (Subscription), Vscode (Free) or some JetBrains products (Subscription). I only know of some people including myself that have ST for opening large files, where performance matters.
I'm using Sublime Text. I feel that most people using ST are happy with it and been happy with it for a long time, you don't see many posts about it cause most of the userbase does not make using Sublime Text into part of their digital persona like many users of other editors (not speaking as if doing this is a bad thing, but you'll see fans of other editors being a lot more vocal).
I don't have any insights into how the company is doing, I'm just going by the sample set of people around me or things I read.
I'm a fan of indie software and native apps but I know zero people in the past 10 years that switched to ST. I know plenty of people of people who switched to Vscode and all the other free or paid competitors. It's probably enough to sustain a small company, and not everyone has to strive for a monopoly. But I wouldn't call that thriving.
I don't know anyone that still uses Sublime. I haven't seen a company recommend its engineers use it either. I used to be an avid user until VSCode came out.
I mean that they solved the funding model that pays the bills of their employees, not that they solved becoming the most widely used text editor in the world.
It does, or did, use dark patterns when showing upgrade notices -- prompting you to upgrade to a version that you don't own yet, without telling you you don't own it, leaving you with an unlicensed version. I was happy to use 3 but that felt really off.
ST developer here. We aren't happy that happened either, it was a big oversight in the ST4 release that made people like yourself lose trust in us. I'm sorry and will do my best to not have something like that happen again.
> Nor was I happy about the new 3-year-of-updates license model that ST4 adopted.
I'm curious what you don't like about this model? The most common complaint with regards to updates was the long waiting period between major versions, which we've now eliminated, and without changing the perpetual nature of our licenses.
not OP but I am not thrilled by "x months/years of updates" because you pay upfront for updates that you're not really sure that will happen. Will I get bug fixes? Will I get new features or at least significant improvements? Will the two person team work on the other project?
I have had experience with quite a few projects switching to recurring billing, occasionally justifying it with "to support development of great new stuff" and then... just keeping the same rate of updates as before, resulting in a de facto steep upgrade. That said, three years of upgrades for 99$ is reasonable, even if there were only bugfixes
> not OP but I am not thrilled by "x months/years of updates" because you pay upfront for updates that you're not really sure that will happen. Will I get bug fixes? Will I get new features or at least significant improvements? Will the two person team work on the other project?
Thanks. I guess I'm just not really seeing how that is any different to what we did before? You'd buy a ST3 license with no knowledge of what improvements would be made to ST3.
It's been a while since I used my ST3 license, but I remember my license affording me all updates for ST3. Maybe that model was unrealistic for ST4, but as others have echoed in this thread, major feature updates for ST4 are not common, so if you bought a license in May of 2022, your access to new features would have expired right before the new updates of the May 2025 release.
Personally, I'm OK with using an old build so I don't mind that much about the limitation. Although if my 3 years elapses right before ST4 introduces first-class LSP support and an official Debugger, I may be very peeved. :)
> It's been a while since I used my ST3 license, but I remember my license affording me all updates for ST3
That's correct, but it could be the day after you buy the license is when ST3 stops receiving any updates and we dedicate all work to ST4. With a ST4 license you're always getting 3 years of updates, ST3 it was anywhere from 3 to zero (not counting the beta period, ST3 only got 3 years of updates).
> Although if my 3 years elapses right before ST4 introduces first-class LSP support and an official Debugger, I may be very peeved. :)
I think we can all sympathize with some buyer's remorse. Unfortunately the line needs to be drawn somewhere. Maybe you can take solace in that we probably can't put multiple huge features in a single update, at least not a dev release. :)
the difference is subtle, and comes mainly from the fact that at the time the concept of Major Release still existed.
You got everything in the current Major Release and every minor and patch update to it. If it took more than three years to release a new Major it was not a big problem, because you were still covered.
The new model might leave you without bugfixes before a new major version is out.
I think you might be misunderstanding the new licensing model. There are no major versions anymore; features & bugfixes just get released when they're ready.
This happened to me and I tried to recover the last licensed version I had used but mixed up my shortcuts or something and, after the 100th time I saw the nagware screen, I gave up and uninstalled and went with something simple and free: Notepad++.
They switched to a subscription model (3 year licenses are still subscriptions), and since the release of ST4 in 2021, there has been exactly one release with new features (May 2025). All other releases have been bug fixes and "improvements".
I get that developers need to make a living, but 4 years of fixing bugs in your products is probably not what I want to be paying for, at least not when that is the only thing I'm getting. Speaking of releases, they're also usually 6-12 months apart.
I have used ST ever since the first version replaced TextMate for my use (TM2 spent something like a decade catching up to ST2), but I've since switched to Code and Zed (mostly Zed as of late, Code on windows until Zed is ready there).
ST was great back when it was still an actively maintained product, but in recent years (ever since ST2) it has felt like it was mostly on the back burner and other editors have passed it in functionality.
As for VC funding, it has done miracles for Code to have Microsoft sponsor it (and others). Code is currently the editor to beat for anything that doesn't involve opening large files.
> They switched to a subscription model (3 year licenses are still subscriptions)
Licenses are perpetual. It is not a subscription. Don't like the work we've done? You can continue to use that version of Sublime Text until the end of time.
> there has been exactly one release with new features (May 2025)
August 2024 we added kinetic scroll and xdg-activation support for Wayland; we also added the ability to configure image extensions and allow dynamically switching between the hex-editor and image view.
November 2023 we added native font dialog support for switching fonts.
August 2023 we added webp and proper support for running as administrator on all platforms.
November 2022 we added syntax-based code folding and operating system recent file integration
December 2021 we added GB18030 support.
I'll stop there. Those are just the largest, most user facing new features, not any of the new settings, new APIs or improvements that I'd argue are new features.
> Speaking of releases, they're also usually 6-12 months apart.
We do stable releases infrequently, because they're stable. If you want more frequent releases you can switch to the development releases. You can see from the build number how many builds we've done of ST4 since the release: it's around 150.
> You can continue to use that version of Sublime Text until the end of time.
Or until some change in the underlying OS makes it no longer able to run. Not saying that happens frequently, but it does happen.
> I'll stop there. Those are just the largest, most user facing new features, not any of the new settings, new APIs or improvements that I'd argue are new features.
New font pickers, settings dialogues and other "polish" is hardly what i'd call new features. I listed those as "improvements".
I'm not trying to start a fight, and i respect the work that has been done, but VS Code releases more features in a month that ST has released in the entire ST4 lifecycle, as does Zed.
I understand the team is (a lot) smaller. My frustration is that when a new release comes out, it's mostly polish, and not fixing the real problems, such as lack of integration and plugins.
You may choose to brush it off as just an old fart rambling on the internet, but i'm not alone with this opinion, and i have ST licenses dating back to the original version, and i have chosen not to renew my ST4 license. When you start losing the "religious" users, maybe it's time to reevaluate ?
Please don’t conflate limited updates with subscriptions. The problems with subscriptions are that the company can take away your own files, the company can take away your software, the lifetime cost is extremely high, and the company can unilaterally change the deal or stop offering one. None of this applies to limited updates offers like Sublime Text’s. You pay once and keep it forever. The three year limit is on the time into the future for which the company continues to add to what you’re keeping forever. Of course this isn’t unlimited, it’s pay once for the program, not pay once for the lifetime servitude of everybody who works on it.
They also don’t bother to announce a Vim or Emacs one either. VS Code provides good default and most people don’t care about editor fluency. Which is why they keep using it.
>>They also don’t bother to announce a Vim or Emacs one either.
vim has a universal and in many ways a eternal use case. You have to edit a file at some point on a server, be it a self hosted or even on ec2. Thats kind of the only real use case for vim.
In these days of AI assisted coding, no one really 'edits' code. A lot of editor short cuts and fluency related concepts kind of in many ways are not relevant in this paradigm.
The thing is vscode just works, like just works, for nearly all the usecases. In case of emacs, learning it and mastering it takes lots of time in ones career. In case of vscode you don't have to do this, you can straight away work on the project that you want to get done.
emacs is some what like a massive distraction from the actual task you want to achieve. Instead of writing code to build a project, you have to first write code to make emacs work, then use emacs to write the project code. In vscode you just write project code.
I don't know if you're trolling or not, but there's one thing that VSCode and nearly all other "normal" editors don't have and I want: Non-tied Windows (pane) and buffers (opened files). One of my most used layout is one main window and two smaller ones. Layout like this are my mental frame, but what I actually want to look at may vary at any moment. It may be a test result, a git diff, or going down a reference link. It's like a moodboard instead of a stack of paper you can only look at one at a times.
Emacs and Vim has this built-in. Other editors kinda have that, but it's clunky. I can suffer IDE because they provide a lot more than editing.
> Instead of writing code to build a project, you have to first write code to make emacs work, then use emacs to write the project code
That's only done once. It's like adjusting the mirrors and seats of a car. Once it's comfortable, you don't have to touch it. Using VS Code feels like borrowing a car with a very limited range of adjustments. Why is the explorer on the left and the terminal at the bottom? Why are they always there?
Have you even ever watched someone experienced using Emacs or you're making assumptions on your (I suppose limited experience)?
The "distraction" framing assumes everyone has the same preferences and working style, I for one find VSCode (and IDEs in general) massively distracting from productively solving many tasks. No, it's not "a skill" issue - I have used InteliJ every single day for almost a decade, diving into some profoundly advanced and non-documented features, and I do open VSCode from time to time.
I feel your argument conflates initial learning curve with ongoing productivity, and assumes VSCode's approach is universally optimal rather than just different.
>>Have you even ever watched someone experienced using Emacs or you're making assumptions on your (I suppose limited experience)?
I am one of those experienced Emacs users myself. Wrote more stuff in Emacs and even vim than most devs today will even write code over their careers.
Its just vscode now does simply too many things out of the box, you obviously can recreate that in Emacs, but its a pointless exercise. Time consuming, and distracts your from your real job. My job is to write code, not build emacs to write code.
I totally stopped using Org-mode, because Google docs do it way better.
At some point you have to move on. For some people like that point arrived a little early.
Can you name Emacs packages you've authored, maintained or contributed to?
> vscode now does simply too many things out of the box
Oh yeah? Can you edit your filesystem in VSCode like a wiki? Changing directories and filenames as if you're editing plain text? Can any of thousand of its extensions provide "indirect buffer" experience? Can you bind a double tap of the comma to navigate to the last error and automatically fix it? Can you at all bind to double/tripple comma or just about any arbitrary key based on context in VSCode? Keyboard macros with counters - e.g. to transform lists into numbered lists? True buffer based (not file oriented) workflows? Occur-mode style search and replace? Comint style process interaction? Are there any extensions that allow you to move the cursor to the piece of a plain text like "rfc-3540" or "myproj#4044", intelligently recognize it and allow you to browse the RFC document or review a Pull-Request - Emacs does. Can you run a shell command and pipe it into a buffer, or pipe the content of a buffer through series of shell commands and into another buffer?
> I totally stopped using Org-mode, because Google docs do it way better.
Don't be ridiculous. They are completely different classes of products. If you are even having to compare Org-mode with GDocs, perhaps you don't know well either or both. GDocs can't manage my dotfiles. I can't use it to annotate books or academic papers. I can't have inline LaTeX formulas in GDocs. It can't let me control a videoplayer and type my notes at the same time. I can't use it to manage my flashcards or keep the log of my LLM interactions there. I can't have executable blocks of code - in org-mode I can explore APIs or send SQL queries while passing them into data transforming code blocks. There are no TODO states/keywords in GDocs, no agenda views, no scheduling/deadlines, no time tracking/clocking, no habit tracking, no pomodoro timers, no dynamic time tables, no tag-based filtering, no capture/refile system.
> At some point you have to move on
Like I told you already, you're assuming too broadly. Both VSCode and Google Docs are excellent products - no denying that, but they are not universally better. If you came to the opposite conclusion, it is yours, and yours only.
Zed only kinda works on Windows. As of today when you click to download windows version you can only sign up on beta program that I assume allows only select people to use that version.
The problem with accepting VC money is they will eventually demand a return on their investment, which means that the forces that drive enshitification will eventually come for Zed in some form. I suspect that we'll see more and more features locked behind a paid subscription and the open core of the editor will become neglected over time.
Here I am on my free-as-in-freedom operating system, making commits with my free DVCS tool in my free programmable text editor, building it with my free language toolchain, using my free terminal emulator/multiplexer with my free UNIX shell. VC backed tools like Warp and Zed that seek to innovate in this space are of zero interest to me as a developer.
I might HAVE to learn EMacs (prefer over Vim) because I think eventually everything else will be tainted by mandatory AI features and/or subscriptions.
VSCode is open source, too, but it's been pretty easy for them to keep forks from taking off by having proprietary extensions, a "markeplace", and other lock-in.
All they have to do is only permit official builds to talk to official builds (for security, of course ;-), and forking Zed becomes a lot harder.
I really dislike the dismissive "fork it" response. You do know how much time and work it takes to maintain your own fork of something, right? It's great that it's possible, but to expect most people to do something like that is absurd.
i do know how hard it is, that doesn’t change anything. nobody is forced to use something like zed and if you think it’s important enough to stress about the product decisions they make, the open source license gives you freedom to decide what to do about it. anyway you don’t even have to maintain it. you can take gpl software you like today, build it for linux and with a docker container you’ll be able run that same binary for the foreseeable future. then you can choose to extend it or not
A year or two ago I moved away from one of the neovim distros when they randomly changed all the keybinds on an upgrade (such things really anger me) and set up my own config. Funnily enough, I preferred vimscript. I still do use lua of course for various things, but those just go in lua EOF blocks in the vimscript. Vimscript is really terse and convenient for many things, I love it.
I’ve always maintained my own configs for (neo)vim. The only area where I prefer vimscript is with certain incantations for which there are no lua-based alternatives. And those are increasingly rare.
Authoring plugins is a lot more attractive in lua, imho.
You can try Helix editor, it is super underrated editor. I always wanted to go down the vim/nvim path but just couldn't stick to it, especially with nvim. Helix configuration is straightforward have some pretty nice built-ins and it is the fastest/snappiest editor I have used so far.
It's not emacs-like. But a lot of plugins wants to adopt the emacs philosophy of having it open for the duration of your login session. Instead of the quick edit and be done of standard Vim.
That’s fine as long as they don’t force AI prompts to me.
To clarify, I use AI agents, but I absolutely hate them submitting code in my editor. Chatting is fair enough and useful, but I need to turn off the auto-generating code part.
Sure, but given the existence of vim/nvim, emails, visual studio code, cursor, etc the price for editors has largely been driven to zero, or at least capped by what JetBrains charges. My concerns are more this is a big bet on a different thing, not the editor (which is quite nice, even if using typescript regularly makes it balloon to 15gb of ram), making them a giant pile of money. With the editor as a free complement.
You can pay for Zed today if you'd like - https://zed.dev/pricing - and also the editor itself is open-source under the GPLv3 license. So if at any point in the future Zed changes direction in a way you don't like, you are perpetually free to build the version you liked from source (or make a community fork and take it in a different direction).
No, it's not odd, I like this take a lot. They almost need a dropdown menu indicating where you'd like your money to go to (editor, LSP, etc).
I'd also much rather have a means of paying a single flat fee to indicate my support than yet another subscription which is misleading because I have zero interest in the AI components of Zed.
I'm enjoying using Zed, and I do pay for tools I use regularly. It's now approaching a month of daily use for me and I don't see that changing. But to echo the other replies, I'm uninterested in the LLM tools. I don't use those normally and paying for that as a way to support Zed would send the wrong signal. You have to be careful when you restrict what people can pay for, because that will become what you optimize for which may not be what your users actually care about.
Zed is fast, easy to configure (so far, maybe some hard parts I haven't run into yet), works well with the languages I care about via LSPs, and the collaboration features are compelling. I want to pay to support that, I don't want to pay for an LLM feature I don't care about that ends up distracting from the progress on the things I want to see maintained or improved.
I already pay for Zed Pro, but my fear (and likely GP's as well) is that this doesn't provide enough revenue for the team.
Since switching from Emacs last year, I have absolutely loved how this editor has evolved, and I am looking for any way to directly support the effort. I have been a Zed Pro subscriber for quite a while now, and I have started trying to contribute to the codebase, but I really wish there were monetization options beyond making a spread on Anthropic API pricing.
Big fan of Zed. I want to echo a sibling comment that I don't see that as paying for Zed, I see it as paying for LLM usage. And since I already have my own LLM keys, I just use those instead.
I can't pay for Zed Pro using my work funds because it's an unapproved AI service. Can you provide another way to pay? E.g. cross device settings sync or professional support.
I'm a big fan of Zed and having met much of the team I think it's some great people building a great product. But I do echo concerns that while the intentions are all honorable the incentives of the pricing structure, business environment, and now a funding round are concerning in the long term. I don't think anyone at Zed has a single ill intent or a secret master plan but these days anything I'm not paying for I just assume is going to be enshittified eventually. Especially for an app where the only paid features are AI-centric and there's a VC expecting to make multiples on a $60m investment.
So here's my ask: let me pay for it without paying for AI! None of my use cases will stress your servers; I have `"disable_ai": true` in my settings.json. Give me a $5/mo "support the devs tier" or a $10/mo tier with some random app quality-of-life features and I'm there. I specifically want to pay for good software without paying for AI to signify the value proposition that still exists there cause I don't think a VC would believe me otherwise.
Unless they have very unusual terms on their funding, it isn't really entirely in their control in the long term. Hopefully they find a way to make their investors whole that doesn't suck for everyone else, but if not, well, I at least appreciate that the editor is truly open source, since at least it offers a contingency plan in the worst cases.
If I'm wrong I'd love to know, but I think that we need to start talking about what funding really implies more honestly. It's traditionally met with unabashed enthusiasm and congratulations, which I totally understand, but it's a mutual exchange, not an award or a grant. I absolutely believe that everyone wants to make good on their promises, but promises made to users are not legally binding, and the track record for upholding those has not been great. Plus, as a user, I want to pay for software, but nothing feels worse than paying, then watching enshittification unfold anyways... When this happens, the investors should send you a nice postcard thanking you for paying back some of their money.
Can $20/mo sustain a text editor company with a massive multimillion dollar valuation? Well, we'll see. Good luck Zed Industries, we're all counting on you.
Please, tell me what it's like living in your free-as-in-freedom house,
feeding your free-as-in-freedom offspring? Eventually demanding a
return on your investment? The audacity!
I haven't seen any negativity only push back. Rust became too hyped. I don't think Rust has a visibility problem right now. Everyone who cares enough already knows about it.
Also, it doesn't have to be rust, it could be C# or Zig or Lua or whatever
Whilst rust is a big part of this it is not the core of their innovation. Pushing everything to the GPU is.
Stating that everyone else is building on Electron belies your lack of exposure to any editors other than vscode \ atom spin offs.
Realistically all the editors including those based on Electron could easily shift to leveraging the GPU more and that would be big leap forward for performance.
Regardless of your political stance, how is taking money from a VC company that has a single person that said a single thing you disagree with “supporting genocide”? And how is uninstalling a text editor going to help literally anyone?
This is 100% pure virtue signaling. This brigading is not helping a single Palestinian. These people just want to feel and show the world that they are “the good guys”, while not actually doing anything or helping anyone. To me, that is absolutely a joke.
Wow, some people really don't know what to give a shit about, do they? It must be endlessly exhausting to go around looking for any reason to boycott everything, instead of using these fantastic tools to make things that make the world a better place.
Counterpoint, apparently cursor's revenue is in the 300-400 million range. So it's not wildly inconceivable that you'd do 40m profit (although I too am doubtful).
I strongly doubt that Cursor makes anywhere near 40m profit. All they're revenue is spent on tokens with the LLM vendors. I'd be surprised if they are even running at positive margin and not just subsidizing usage with the VC money.
Unsure of what the end goal is, but I expect everything AI related to be a load-leader right now and then the goal being to figure out how to drive down costs or make even more money later.
If the world is moving towards open models which are "good enough" – the main winner will be the one controlling the distribution. So if you're controlling the web browser, the editor or the OS – you are the winner.
Obviously, the risk here is very high from this perspective, since nothing guarantees anything.
Zed isn’t just a text editor. It is the only working platform for code assistants.
Neither Google nor Claude nor anyone can’t at the moment get right basic operations like file edits. Zed is flawless in co-operating with most LLM models. And not just that - also switching models during conversation and more.
I am at Zed Pro at $20 but when Zed offers $200 Max plan I will sign up right away.
I switched off of Zed's agent system onto the Claude Code CLI because I was blown away by how much better CC was.
Even with "auto edit" turned on, Zed just kept asking me for confirmation. I'd be like "hey your code has this bug", and it'd be like "you're right, and this is why. here's how you can fix it" ??? just fix it man it's your code. Maybe this is fixed now, but Claude Code never has this issue and is very good at only stopping when truly stuck (and generally for good reason!)
Changing the topic a bit, Zed's collaboration features seem really good but it's quite hard to use when nobody uses it in the first place. With VSCode, I can use the LiveShare extension and everyone on the team can just join with no fuss at all. LiveShare is likely not nearly as technically great, but the simple fact that people can use it easily makes it win hard.
Honestly it would be cool if Zed can somehow become more popular thanks to this investment. As long as it can keep its speed and technical excellence. VSCode used to be super lean and cool, but now it's just another fat IDE with unlimited bells and whistles. It feels like how Eclipse felt back in the day.
No....there is value in having many developers use your tool for daily code development. You should think if it as a lot more than a text editor as well, just like Github is a lot more than just hosted Git (also VC backed at first).
I love working in Zed. It really is a delight to use, and I think the Agentic coding integration is really well done. I'm excited to see them investing in this space more.
I understand people's concerns about VC funding, but I think building quality products takes capital. The funding is still relatively small, especially when you compare it to players like Cursor, etc. And I think Zed is a much, much better product!
Zed being OSS is a gift to the community, and I suspect DeltaDB will be as well. And as others have said, Nathan (CEO) is a delightful human.
Maybe you can comment on this. I saw in the Zed demo that they're using "rewrite optimized version Claude". My first thought was wondering how big the bull is going to be.
> And as others have said, Nathan (CEO) is a delightful human.
Hate cancel culture as much as the next techbro, but since y'all point out they're "a delightful human", it'd be interesting to see how Nathan then responds to concerns raised in some quarters: https://github.com/zed-industries/zed/discussions/36604
With so many new AI editors popping up, it feels like the Zed team is in a tough “lose/lose” spot.
If they stick to their current path—focusing on craftsmanship and letting their world-class talent build the best editor—they risk falling behind giants like Cursor. Cursor seemingly came out of nowhere and is already doing $300–400M annually, rivaling JetBrains (who took 20 years to get there) in just two years. With that kind of momentum, Cursor can now buy their way into a superior product.
On the other hand, if Zed takes more VC money, it likely means doubling down on AI in ways they clearly don’t seem eager to—but at least it would give them a fighting chance.
I really feel for this team. It couldn’t have been an easy decision, and from the few interactions I’ve had, they strike me as incredibly talented, kind, and genuine folks. I truly wish them the best.
I hadn’t thought about it that way but it’s an interesting point. The cursor folks probably have 99% of the work already done for them, for free, funded by Microsoft, probably in perpetuity.
And on top of that, since it’s essentially just vscode, it costs users almost zero effort to make the switch. It’s the perfect crime!!
I hope zed does well though. I love their blog, and all of the cool open source stuff they’ve made. I recently heard they added a “helix mode” which might be enough to get me to switch from vscode…
this is where the advances in collaboration come in; it's a way out for them because existing editors are largely stuck with git and are patching ai on top of everything.
zed came at the incredibly (un?)fortunate timing where they were just able to build a solid base before the editor wars began. their only path now is to fully maximize the few advantages they do have:
* a fresh base that is far more flexible
* really good experience with performance, design, general craftmanship
* a buzzy community and fresh/boldness that attracts vcs
for zed to truly win (at least in sequioa's eyes) they will need to completely take over vscode as the new default, and that will require a big lead when it comes to collaboration and ai
> Cursor seemingly came out of nowhere and is already doing $300–400M annually, rivaling JetBrains (who took 20 years to get there
Revenue, not profit. I’d imagine cursor loses a dollar for every 50 cents of revenue and makes it up in volume. Meanwhile JetBrains is a profitable company not beholden to the whims of outside investors.
Tbh: I think git is not the century-tool that a lot of people think it is, and we're in an era now where its decreasingly serving the needs of its users. I'm ready for deeper ground-up innovation in the space of code collaboration.
Developers don't think in terms of commits; they think in terms of Tickets and PRs. Git doesn't have any native representation of either of these things. Getting tickets into the repository is something the community has talked about for years, and Linus himself has said he wants, but it hasn't happened yet. Branches work fine enough to start a PR, but then Github, again, has to take over for comments, CI execution, even the decision on how you want the commits on that branch represented on the main branch (squash? merge? rebase? why should I care?). LFS has been years in the making. Monorepos are still weird.
Idk, I think there's capacity in the industry to take the concept of a repository back to square one and think more holistically about how we all interface with the repository. The way they talk about DeltaDB gives me hope that they'll start thinking more about this with this funding; commits just aren't a useful primitive anymore, and a repository, locally-downloaded, open source, ultratight connection between diffs, merges, communication, and automated tooling is how I want to be coding in 2030.
Totally agree. git notes might help a bit with some of those points, but what most shops need / want is something like GitLab. git itself is usually too low level.
Is IntelliJ "bad"? Aren't the reactions here overly negative?
This means the company is funded, development will continue, zed will continue to improve. An IntelliJ style license (for example) is an acceptable trade-off from my point-of-view
The thing with taking VC funding is that your intention usually is not to steadily grow a sustainable product.
You take the funding so that you can outgrow the competitors and get the market faster.
All you need is the small promise of innovation in an area which is somewhat new.
At the beginning, the product is good enough and you have money to keep marketing and developing slightly faster than others. This will get you the users.
In the end, is your product at that point truly the best among competitors? It matters less, since you already have the users.
I think Zed didn’t need this one since they had a great product. Many would have been ready to pay at least a little. They could have grown slowly and see what works. With VC money take can go to completely wrong direction with giant steps and they are not noticing that unless it is too late. And then investors want returns.
Another big point was they implemented their own parsers for everything which allowed them to make nifty things - the refactor features way back in the early 10s was miles ahead of everyone else - but then LSP happened and that advantage is diminishing and becoming a liability
>IntelliJ lost the plot at the inception of CLion etc.
I was a customer for so many years. "One IDE to rule them all" and then they started cashing on more.
What are you talking about?
ReSharper came out 21 years ago 3 years after Intellij.
RubyMine came out 15 years ago. 7 years before CLion.
I don't write Ruby, but I write Go and C, and C++ and I was left facing a new license. For no reason at all. It's the same debugger and the same code base, you just need to hook into gdb or lldb instead of all the other ones.
Like I said it's only one of the problems, read the rest.
The paid version of Intellij has never lost anything. Pretty much everything the specialized IDEs can do, Intellij can do too, though maybe some features lag. CLion, Rubymine etc are just less expensive specialized versions.
I'm sure the free version has lost some things.
Its been a long time since I used CLion but it was the best C++ IDE by a huge margin.
I agree CLion is the best as C IDE, but there is no reason that intelliJ couldn't do C/C++/Go, other than cashing in with new product lines and licenses.
Only ones that it can't is C/C++ (probably because that started as Apple only IDE) and .NET (that started as extension for Visual Studio).
They started Fleet that can do all of them, as a response for VS Code. But then came AI and Cursor probably took a large chunk of their users. So they redirected resources (like Zed that paused their collaborative features to focus on AI). There is currently a closed EAP for Cursor like ide based on Fleet. And their Git client as separate app.
The C++ plugin is included in Android Studio which is essentially a distribution of IntelliJ. There seems to be no technical reason for not allowing it in full IntelliJ.
The days of using a separate IDE for each language are kind of over.
These very paradigms are outdated these days. vscode got it, very early. vscode works for everything. Most projects use Python/Go and JS, and out of the box vscode just works for all these languages and their tools.
DeltaDB sounds of being a >git innovation for coding itself, and would fulfill Zed's promises in Nathan Sobo's debate/discussion with Steve Yegge recently.
Seems to solve a real problem which is growing rapidly, both in the old way and in the new way ... if it can overcome _slop_ in LLM chats, and the sheer enormity of code/data ahead. Trying to picture how coherence will survive.
With claims/hype/concern floating around that >90% of code will be LLM-generated within 3-6 months, with the insinuation/tone [1] that the same amount of code will be written by humans as now ( at least at first ) but LLM code will radically grow to dilute the space ( as is happening ) ... seems like DeltaDB being done right/well is going to be do-or-die on whether coherence remains possible!
For anyone not clicking through the link: it's from 5 months ago, predicting 90% of today's code will be from LLMs.
>I think we will be there in three to six months, where AI is writing 90% of the code.
They're going to keep saying it because it's a juicy sound bite and they're sales people. That doesn't make it any more true than "9 out of 10 dentists recommend our socks" or how we surely have all had flying cars for decades now.
I had a week where I embraced and went deep on ai first coding (not vibe, that'd be crazy)
The thing wrote at least 80%, so we aren't far off in this anecdotal instance. There are citizen devs who are building fun things for themselves where the AI does 100%
It made me realize these things are more capable than I knew, though they still do dumb stuff reliably. But, it is easy to undo those changes, so the productivity boost remains
for those who just caught the headline, the real gem here isn't the investment, but what zed's doing different: rather than using git for version control, they're announcing deltadb which incrementally stores all changes as crdt [1] while being interopable with git.
this feels pretty important; git is definitely not the right primitive for version control with ai and that pain is obvious with existing solutions. zed seems to be going all in on collaboration with realtime, git, and now something in between and it'll be interesting to see where they end up-to me three solutions feels overcomplicated but that may be necessary given how teams work now.
Ehh Zed tried to push the collaborative editing before the AI hype was thing (that blog post is from 2022), and seemingly everyone outside of Zed developers promptly ignored the idea.
The Zed developers quickly refocused on milking the AI cow.
I'll argue git is plenty good, and there are a lot of people who don't understand it very well. The ones talking about Github PRs as if they were related to Git in any way most definitely don't.
It's funny how often people buy into the marketing and say "blazing fast" without actually questioning it. FWIW, I still prefer Zed because its LSP integration and vim mode are better than Sublime's.
I really love Sublime. If it had a solid "remote-ssh" type integration and a more straightforward way to manage plugins/extensions that isn't a rats nest of random directories with similar names - I would return to using it. I purchased every upgrade from like 2012 to 2022 or so.
I have tried it out and by default it was so slow as to be unusable. After discovering it required some customization in /etc (because it's the only GUI application that fails to recognize my GPU on a very popular distro with next to zero customization, because I game a lot on Linux - weird how that's a me problem and not a Zed problem) it got better, but still noticeably slower than VS Code.
The modern Sublime Text is Sublime Text. There is way too much "extra" in Zed to compare it. If anything, it's a new IntelliJ.
Slower than VS Code?? I guess it's just poorly optimized for Linux. On MacOS, I find it to be significantly faster than VS Code, and the only alternative I've found that's in the ballpark of Sublime (from the performance aspect).
I do not understand how one says this with a straight face unless you’ve never used either product. Do you also believe Neovim to be a “new IntelliJ”? That’s the level of functionality out the box, though fortunately Zed does not require lots of screwing with config files to get basic things working.
This is accurate. I came over from Sublime Text because it had become laggy over >5 running instances, and native LLM integration. Even VS Code doesn't actually have that... where everything is an extension versus seamlessly/perfectly fitting
As mentioned in other comments, it actually outperforms window management in general in many/most cases. Radically flexible and almost never gets in the way
Just to clarify: I'm not saying I doubt they are the fastest, I don't know and I don't care. What I do care is that a company (a VC backed one specially) that markets their product like that follows the statement with proof, otherwise we should call it BS
I don't know what's DeltaDB, but if it will be able to show when a commit is simultaneously a refactoring and a change of code being refactored (moved into another module) – I would love to see that! At the moment if you refactor you punish the people reviewing the code, which reverses the motivation for cleaning things up.
Not sure what this guess is based on. Would that be a guess for git also, if mentioned by a company versus an individual?
My read was that they are pulling a Linus Torvalds with the Linux->Git move where both are innovations on their own, but work great together ( without dystopian universe instantiation )
zed is just on the hype train, obviously very talented people, they are thinking hard about LLMs but I'm not really sure where they are going, their pivot is probably going to be more interesting...
Sounds like a noble goal.. but I'm skeptical. I normally don't care about every single edit I've done, and don't want to store them and certainly don't want to see other people's character level edits.
Also how on earth do you handle conflicts?
This sounds most similar to IntelliJ's local history tracking, but that is local only which makes sense.
Still, more IDEs is always good so I wish them luck and will wait and see. (As with Zed - I keep trying it but it's still very alpha quality so I always end up back in VSCode.)
although I tried the collaborative features I unfortunately cannot say that I used these extensively for now. Although I found them quite nice, well integrated, seamless, and straightforward!
The issue is again people, they don't wanna change their _archaic_ workflow, stuck with inefficient -copy/paste- loop to the chat (ie. Slack) and back.
The story in the article went a bit too far that I agree, but I guess that is their north-star vision. Current implementation allows you to "join" a workspace session shared by someone, edit the same or different file, follow/watch a certain person, as well as have a chat (without requiring copy-paste) about certain piece of code. (both written or via voice)
If something, large enterprises generally don't support smaller and ambiguous licenses. Therefore, if Zed will allow enterprise licensing (ie. via on-prem license server or volume ordering, SSO, whatever) that would increase their adoption quite well...
For the record: I have never used the collaboration aspects of Zed
What I also have not used is vim emulation, though I have a vim background
As mentioned elsewhere, Zed is still very configuration-dependent to get the full power of it, and a lot of its functionality is never discovered for that reason
What pushed me to try it was Ollama integration which is not an afterthought, then I realized I loved it _way_ more than SublimeText, especially on performance, at first, then everything else once that won me over
I have ~10 running instances at any given moment, and >99% of the time never feel any lag, whatsoever
Another unexpected benefit is that terminals, code editor panels, and assistant chats, get to be sized and fit wherever you want, so it is also kind of a window manager... I often have more terminals open in Zed than in the Window Manager of the OS itself
> then I realized I loved it _way_ more than SublimeText, especially on performance
I currently have 19 instances of Sublime Text open, each to a separate folder containing a mix of C++ and Python code bases (some tiny some huge). Like ~8 of those have the clangd LSP plugin enabled. I don't think I've ever experienced lag in Sublime. KDE System Monitor is reporting 2.0 GiB of ram being using by sublime currently.
The clangd LSP plugin in Sublime isn't perfect, and it does occasionally break, and rarely spikes in CPU usage for no reason (although the editor always remains responsive). But, if I ever switch away from Sublime Text, I cannot imagine it ever being due to performance reasons.
I do all those same things in VS Code, especially the vim bindings, wouldn't give those up, but did recently leave the vim ecosystem because I had to spend too much time making the IDE work or enable features that are out-of-the-box in a code-oss based IDE
I give lots of feedback to Copilot in the hopes it makes the agents better in the long-run. I want them to read my code and train on it, along with the interactions with copilot, which is the next frontier in (post) training
Nothing since then really recaptured what I personally liked about GWave or let me use their tool in similar ways to how I used it. YMMV, of course, more so than most of my comments.
The vision sounds interesting but I think if they have two challenges they will need to overcome to compete in the IDE space
1) Their LLM integration needs to be at the quality of Cursor and VSCode to pull people away from those
2) Reduced friction to move over (keyboard shortcuts, common plugins, etc)
I think the Zed team is perfectly capable of winning. The bigger risk would be them trying to tackle fancy stuff before making sure the basics are good enough to get developers to switch.
They need to build up a deep understanding of why folks are sticking to Cursor/VSCode and not swapping over.
P.S. I would love for Zed to win in the market because I'm sick of slow software and it's refreshing to finally see an intense focus on performance.
I customized the hell out of my keybindings. I might be in the minority though! I was trying to give examples of (stupid) little points of friction that might prevent someone from taking the leap.
I think they're tackling a real problem that is revealing itself via agentic coding, and I think they've positioned themselves in an interesting spot.
I'd be interested to know how much data DeltaDB accumulates over time - because the level of granularity is so high - and are they going to want to use that data as training data?
Well done I suppose to the Zed founders as they're on the Sequoia gravy train now. But as others have noted this puts an inevitable clock on the enshittification no matter how hard the team crosses their heart and truly believes otherwise (or not, I mean maybe this IS the gameplan).
Hopefully Zedless [0] gets some community steam behind it.
While i am sure this sounds insane i had to drop zed due to lack of the “last file dif” button gitlens vscode plugin provides.
It is a godsend on quickly debugging the why of things. If anyone knows how to replicate the same functionality with the same number of clicks in zed, id happily switch back to it.
I am not familiar with gitlens so not sure how close this gets you but you should be able to replicate the functionality you need from the git CLI and some light scripting. This can be a jumping off point maybe. If you want to view the diff using the zed diff viewer, you can do so using `zed --diff`, as demonstrated in this GitHub discussion: https://github.com/zed-industries/zed/discussions/33503#disc...
I love Zed and I am glad they are getting funded. This allows them to truly polish both the remote collab and the AI features and compete with Cursor and friends.
The product has improved so much in the last year, it's been a pleasure to use and get excited about new features.
VC backing apparently has the power to move the standard from "hodge-podge 800-line commit must have an issue key" to "track every damn interaction and back and forth".
That and the ever-trailing "especially now with AI".
Is anyone actually using the collaboration features? Zed is my daily editor and I love it, but I've never touched the collaboration features, which is their "USP" when raising cash.
Zed defaults don't seem to capture the full value of what it offers. For instance, edit predictions via tab completion are documented yet I've never experienced them. I need better settings, I guess.
On the bottom right, there is a button looks like magic emoji (). Click that and it will show general AI settings. You can choose _Eager-Completion_ rather than trigger-based one.
As a user, most IDEs/Editors currently show in _eager_ way. For example, VSCode by default shows ghost-text as well as Amazon Q extension too. Usually disabling those disables the AI completion completely.
Meanwhile I like Zed's approach that you can trigger completions with Alt+Space, not burning through your "tokens" in free-tiers. They also provide a free-tier completions, as well as _next-edit suggestions_.
*) Next-edit suggestions: When you edit some piece of code repeatedly, it suggest to do similar on the next few instances, with context awareness, quite nice feature saving several keystrokes every single time.
Not a good move. Evernote went that path and then turned an otherwise very useful app into a VC bet that forced the founders to pivot the app into something nobody wanted. Same happened to Soundcloud.
I wanted to try out Zed. It looked fine on my laptop screen but as soon as I tried it on a 1440p monitor, the font and icon rendering fall completely flat and are a blurry mess. If your product is a text editor focused on code, your font rendering better be *top of class*.
All of this data is valuable for the Zed team and
could easily be compiled into reports for management
about developer's individual progress and productivity.
I am certain the team will say that is not on the roadmap,
but for me, it still sounds like a possible leak in the
future.
Or maybe a semi draconian "mandatory" extension your
bosses will demand installed.
If you need a simple text editor without DeltaDB bloat, see CudaText. No so fast drawing speed because it does not use GPU. But drawing speed is not the decision factor for many tasks of text work.
Zed has better drawing speed. But DeltaDB bloat, AI bloat, telemetry, need of modern videocard.
From the description it seems my deveopment experience would
become a lot more crowded and furstraiting.
All the different actors sharing a live coding experience.
I feel my focus will become fragmented like someone poking
my shoulder frequently.
Hopefullly intended use is to schedule a time for it.
deltadb sounds interesting. AI editors should look into letting you operate on multiple git branches simultaneously, like isolated browser tabs where the AI gen doesn't contaminate the other.
One of the reasons I find LLMs don't increase productivity much is that I can't switch branches to multitask while it's processing. Context switching isn't always useful but there's still lots of opportunities for rapidly experimenting or ticking off a couple small bugs quickly while AI takes the first pass on something more complicated.
Each "branch tab" could have a sort of TODO list or plan.
I have been running multiple instances of the Codex CLI tool in several terminals to do different stuff. You can even checkout the repo several times, write the todo to a file (or ask the agent to do so), if you need.
I have trouble imagining a world where I will ever pay for this. Unless it becomes as feature rich and powerful as an IDE like PyCharm, it’s not compelling.
Exciting ideas! I'll take a look at this again in several years when they aren't trying to rebuild everything from the ground up from first principles.
I do like Zed very much. Fast, good UI, decent set of features.
Still missing some key features to get parity for python with PyCharm, and git features need more love, but pace of development has been blistering. Kudos to the team for building something really solid.
I don't use the agentic stuff though -- I have a ClaudeMax subscription and use CC or more recently opencode (the only non-CC that works with a ClaudeMax sub); I don't want to pay another sub for Zed and the free model use that comes with Zed runs out very quickly (understandably so, no complaints there). If I could connect my ClaudeMax sub to it then I'd maybe use it, though CC and opencode are pretty good.
The code completion is decent and I especially like the "subtle" mode saves a lot of backspacing (no, dammit, that's not what I wanted!).
I understand Zed needs to make money, but VC backed, especially Sequoia, doesn't inspire any love, tbh. I don't care what platitudes VCs say about independence their love for the product etc., they need their 10x return and if it means enshittification, so be it, they don't really care.
That brings them to 120x the amount of money I've invested in the same problem, and still they're only just announcing that they're intending to start trying to do something that I've already done better
It's apparently this but I can't really say that I get it: https://bablr.org
He seems to be saying he spent $350k making this. I guess it's some tooling for writing parsers.
He has this to say about Zed:
> Zed: Founded by Atom’s dev team, Zed was the rewrite that Atom always wanted to be able to do but couldn’t when Microsoft bought Github and made the executive decision to kill a product it might otherwise have had to compete with. Unfortunately Zed decided to do that rewrite in Rust. This has slowed their iteration speed, caused much of their dev effort to go to cross-platform support instead of innovation, cut them off from being able to offer their experience on the web, severely limited their hackability, and generally made theirs a niche tool for enthusiasts. What’s worse, their reliance on LSP — a product which believes that the presentation layer should be the primary abstraction layer — means their product is forever doomed to look like a VSCode knock-off. [1]
IMO in the 12 months or so I've been aware of the BABLR work and observed the author's comment contributions, they've never really substantiated why BABLR would be preferable to tree-sitter, or how a JS-based implementation of parser tech can fulfill any of the same niches. Most consumers of tree-sitter leverage it via FFI or native code, not an embedded or external JS runtime.
It's not clear to me how you could substantially replace the capabilities/benefits of what LSP provides with BABLR either.
We have some meaningful advantages that help us gain a practical edge on Tree-sitter:
- Our core and grammars are relatively tiny and can easily be loaded into any web page as a syntax highlighter, except actually a bit more: more like being able to embed ASTExplorer directly into your docs page or blog post to help people understand code examples.
- We support runtime extensibility of languages, e.g. TS can extend from JS at runtime. Tree-sitter only supports static linking, so every shipped language extended from their JS grammar contains a complete copy of the JS grammar.
- Our grammars are much easier to debug. They're written as plain scripts, and they can be run and debugged in exactly the form they're written in.
- We can parse inputs with embedding gaps. An example of such an input would be the content of a template tag before interpolations have been applied. Parsing after interpolation opens the door to injection attacks, but parsing before interpolation allow safe composition of code fragments using template tags.
- We emit streaming parse results on the fly, and can parse infinitely long data streams with ease or syntax highlight. within large single-line files without freezing up
- Tree-sitter is half an IDE's state solution: "just add text buffer". Our solution is the whole thing. One stop shop. IDE in a box.
- CSTML lets us do round trip serialization of any tree and also gives our trees stable hashes. Tree-sitter could trivially represent its parse results as CSTML if it cared to, giving it competitive compatibility with BABLR. A rising tide lifts all boats.
- While they're setting out to make version control for the first time now, we're already basically as powerful as git thanks to the combination of hashed trees, immutable data, and btree amortization within nodes for maximum structural reuse of data.
- We can probably literally just run the Javascript source code for Tree-sitter grammars on our runtime. The only problem is the C lexers, but C lexers are one of the great annoyances of tree-sitter anyway since any context in the grammar requires you to hand-write the lexer
I don't think they are doomed to be a VSCode knock-off due to their technical decisions. I think they decided to be a VSCode knock-off by design. I mean they follow the same project oriented GUI design that VSCode and Sublime helped make popular. There is nothing inherent to their tech that required that.
We're approaching the problem by drawing from browser design. We want to see an editor with a DOM API for code documents. BABLR is a parser framework meant as a direct answer to Tree-sitter.
Myself and most people I know that switch to Zed (many from Sublime) did it because we DON'T want a DOM in our text editor. Webtech makes for crappy text editors.
Sorry, I didn't make that very clear. I'm not talking about the HTML DOM, which exists as an API to control the 2D pixel rendering layer.
A standardized DOM for code would provide a universal way of changing code documents using scripts and would allow many different code-oriented tools to interoperate naturally where they never could before.
I think I know what you're saying, and it sounds like an interesting thing to attempt. Some sort of universal API that feels really good to make structural edits to code documents would be excellent.
I love the spirit of Zed. From the principles to the low-level implementation details, it all screams "good taste". It's immensely interesting as an object of study (the code is great, from GPUI all the way up).
Having said that, I don't think an editor should be VC backed. It's the obvious pragmatic choice to get a team together to support a thing, but I'm concerned by it.
Sublime Text solved this 17 years ago with the 40-year-old shareware model.
It's also faster than Zed, works on Linux/Win/MacOS, and is decently customizable.
There was a time around ST2 when it felt like everyone was using it and it could've become The Editor, then something happened and it's been left in the dust. I wasn't even aware but apparently even fourth version of ST was released, and that was in 2021.
I lost track of what happened there (moved to Vim back then), was it VSCode that killed it?
I've been a registered user of ST for long long time, and I thought if anything hurt them in the marketplace, it was taking several years off from the end of 2013 until late 2017 with hardly anything being released that opened the door to Atom and other editors to catch up.
Yeah, as others already mentioned, I think they sat on their laurels for a bit too long and let VSCode overtake it.
For what it's worth, I went from ST3 -> VSCode -> ST4, and have been happy since. I've found that I prefer my text editor with minimal extensions, and with Sublime Text's LSP Plugin, I'm pretty content. The performance and customizable UI make it more worth it to me than VSCode.
It's the LSP plugin that finally drove me to leave ST4 for Zed. Language integration is table stakes for an editor now. The fact that ST support is behind a volunteer plugin instead of integrated directly in the editor just means it's never going to be as good as a editor that does have first class support. The ST devs need to actually improve the editor, but I haven't seen any material updates in years.
I think it's less that they sat on their laurels and more that a team of 2 had trouble keeping up with the dozens of well-paid folks working on VSCode. Which suggests that perhaps a shareware model did not work out so well for them.
They literally stopped developing for about 5 years, it wasn't just about the team not keeping up
I don't think development actually stopped. ST3 was in a quiet public beta for a long time, but you can see builds of ST3 from 2013 to 2019: https://www.sublimetext.com/3.
Why would you stop developing for 5 years?
Maybe they just wanted to do something else. Sometimes people just don't want to grind on endlessly for theoretically more money when what they've got already is enough for them.
Probably to write sublime merge
Agreed, LSP has replaced Linter extensions and the TabNine LLM which is nice (and snappy). Even if some of the lsp servers are clunky to use.
even before VSCode Atom had started to eat their lunch
I don't get this Sublime is dead nonsense. It's still being updated and works great. It's been my editor of choice for years and I happily pay for my license. I'd probably pay more if they asked me, it's tremendous value for money in my opinion.
I dislike "$x is dead" as much if not more as you do, and I'm sure it works fine as something doesn't need to be the most popular choice to be working.
With that being said, just a quick look at, for example Stack Overflow 2025 survey tells me it doesn't have the same mindshare it once had.
Eh.. I gave ST4 a go earlier this year and moved away from it due to plugins I wanted not being updated for years, and no longer working. That really feels on the cusp of being dead to me
There needs to be a critical mass of people using it for things that aren't core to stay updated
I don't run that many plugins to be fair, but the ones I do run (of which at least a couple are no longer maintained) works fine.
The key plugins I use are some LSP servers, and they work wonders. The few languages I mainly use (yaml, json, TS/JS, python and Go) I get great language support for via the LSP servers and the editor is blissfully fast always.
I could live without even the LSP stuff, but the one feature I can't live without is Sublime's excellent recovery support. Every once in a while my system will crash, and even though I've had multiple unsaved buffers Sublime recovers them every single time. Saved my butt more times than I want to know!
Out of curiosity which plugins didn't work for you?
wish I could remember, this was a good 6 months ago
Yeah Atom and then VSCode killed it. Turns out being able to use JS to extend your editor is quite valuable. Essentially every JS devs have their own Emacs without having to learn Emacs and Lisp
Plugins were kind of it's selling point, yet it was pretty easy to mess it up with Plugins to the point of it being unusable - and not knowing what plug-in caused that.
The same curse emacs suffers from. What is the best sweet spot an editor/IDE has achieved to date?
I remember the extremes of the utter unconstrained chaos of Emacs and the rigid ultra-high-boilerplate approach of the Eclipse IDE. Emacs was fun to hack on, but impractical to use as an IDE, because if you installed enough plugins to make it useful as an IDE, it was broken half the time (my experience, many years ago.) Eclipse had a robust architecture, but writing plugins for it was a dispiriting slog, even when I got paid for it (again, my experience, many years ago.)
> What is the best sweet spot an editor/IDE has achieved to date?
Unironically, maybe VS Code.
Everything simple you can do with it, either comes built-in, or within Github/Microsoft ecosystem, or has an official plugin that gets recommended and featured by the editor itself. Plugins from individual hobbyist developers I have, I can almost count in one hand. (VSCodeVim being the most important one)
Now I compare this to my Neovim setup, and that one is basically running on charity from OSS developers.
What about writing a quick ad-hoc command? Something I would have found useful today, which I would have done in emacs fifteen years ago, was writing a command to parse a file in a log, generate a curl command from it, and copy the command to the clipboard. Could I do that in VSCode without creating an entire project?
I can't even start listing the issues with your hasty generalization here - I see outdated anecdotal evidence, survivorship bias, vague metrics, false correlation, goalposts moving. While your personal experience likely genuine, presenting it as evidence that Emacs is inherently impractical as an IDE only adds to the fallacy of generalizing from a single data point to universal truth.
I have completely opposite experience with [modern] Emacs. Of course, it wasn't smooth from the day one, but neither was my ride with different IDEs. Somehow, I keep coming back to Emacs because no IDE ever provided all the machinery I need to be productive. For me (and I suppose for many other people), Emacs is far more sweeter spot of an IDE than any other alternative.
Emacs has eglot built in these days and it works quite well as an IDE
An emacs "distribution" like e.g. doom-emacs has worked to be quite stable for me.
VSCode copied most of the good features of ST and it is free and open source. Just that is enough to overtake it.
I still use it, it is maintained and it is very good and fast, and that it didn't try to reinvent itself is a good thing for me. But it is not a full IDE (not Jetbrains), it didn't jump on the AI bandwagon (not Cursor), and it is not free (not VSCode), so it is not surprising that it lost some market share. But it is not dead.
NB: Not all parts of VSCode are open source, especially not many popular extensions.
good question. I think the community fell off and many plugins were left unmaintained. I was using it for over a decade up until recently. ST4 had so many plugin issues and it stopped being worth manually fixing.
For me, it was that the maintainer of a language plugin I used was, um, challenging to work with. I wanted to contribute to add some much-requested functionality and he talked to me like I was a toddler and warned me not to waste his time.
Well, OK then. Back to Emacs I went.
Yes, VSCode killed it, because VSCode was free. Which is kind of sad because ST2 is actually noticeably faster than VSCode. Someone mentioned Atom, but that was never really a contender, not many people used it.
Personally at that time (circa 2013 I think) I wasn't using it because it lacked integrated features like debugger or good autocomplete. I was using a specific editor, but one editor per language (java = eclipse, C++ = QTCreator, C = geany). I feel like there wasn't a true "one size fits all" editor (except maybe Vim, but it felt so... unfriendly?). Also, I'm not sure it was available on Linux (don't quote me on that)
I love ST and don't get why it stopped taking off. VSCode is way too heavyweight for just editing text and TextEdit is more like WordPad on Windows than Notepad.
I think it stopped taking off when they stopped developing on it for something like four years
Anecdotally a lot of people I know went from Sublime to Atom to VSCode. I think it mostly was about scale of community and momentum of updates.
Basically yes, vscode was free, almost as fast and had more features, like out of the box intellisense.
People refer to it as VSCode, but do people really use VSCode or VSCodium? I personally use VSCodium. I think that is the most de-telemetry'd version of it, I think?
Not sure if VSCodium and others can use all the extensions, especially Microsoft ones like Copilot. I remember setting up something like that once and had to do a lot of fiddling to be able to use all the extensions I wanted. Gave up on it at some point.
As for people in general, safe bet in almost any topic is that most people won’t care. :) So yes most people use VSCode.
and before that was TextMate.
vscode came along with a thriving extension ecosystem. That made up for any pitfalls really.
They didn’t “solve” it, otherwise it would be a thriving editor that everyone would be using.
In reality 70% of the people I see are using Cursor (Subscription), Vscode (Free) or some JetBrains products (Subscription). I only know of some people including myself that have ST for opening large files, where performance matters.
I'm using Sublime Text. I feel that most people using ST are happy with it and been happy with it for a long time, you don't see many posts about it cause most of the userbase does not make using Sublime Text into part of their digital persona like many users of other editors (not speaking as if doing this is a bad thing, but you'll see fans of other editors being a lot more vocal).
That makes sense. The rest of us left ST a long while ago, and the rest remain because they're happy with it and have been for a long time.
Why do you think that it is not thriving? Is the company struggling? Not everyone needs to use the thing to be thriving.
I don't have any insights into how the company is doing, I'm just going by the sample set of people around me or things I read.
I'm a fan of indie software and native apps but I know zero people in the past 10 years that switched to ST. I know plenty of people of people who switched to Vscode and all the other free or paid competitors. It's probably enough to sustain a small company, and not everyone has to strive for a monopoly. But I wouldn't call that thriving.
I don't know anyone that still uses Sublime. I haven't seen a company recommend its engineers use it either. I used to be an avid user until VSCode came out.
It has users but the number of users is dwarfed by the big dogs like VS Code, Visual Studio and IntelliJ Idea - https://survey.stackoverflow.co/2024/technology#most-popular... (using 2024 because 2025 onwards is AI slop).
People are 7x more likely to be using VS Code, which means that a niche tool is far more likely to have a VA Code plugin than an ST plugin.
Other than that, if the 11% of people using it are happy then there’s no issue.
That’s true. It’s all relative.
I mean that they solved the funding model that pays the bills of their employees, not that they solved becoming the most widely used text editor in the world.
It does, or did, use dark patterns when showing upgrade notices -- prompting you to upgrade to a version that you don't own yet, without telling you you don't own it, leaving you with an unlicensed version. I was happy to use 3 but that felt really off.
Yeah, I wasn't happy about that. Nor was I happy about the new 3-year-of-updates license model that ST4 adopted.
Although at least to me, Sublime Text 4 feels like a "finished" product.
ST developer here. We aren't happy that happened either, it was a big oversight in the ST4 release that made people like yourself lose trust in us. I'm sorry and will do my best to not have something like that happen again.
> Nor was I happy about the new 3-year-of-updates license model that ST4 adopted.
I'm curious what you don't like about this model? The most common complaint with regards to updates was the long waiting period between major versions, which we've now eliminated, and without changing the perpetual nature of our licenses.
not OP but I am not thrilled by "x months/years of updates" because you pay upfront for updates that you're not really sure that will happen. Will I get bug fixes? Will I get new features or at least significant improvements? Will the two person team work on the other project?
I have had experience with quite a few projects switching to recurring billing, occasionally justifying it with "to support development of great new stuff" and then... just keeping the same rate of updates as before, resulting in a de facto steep upgrade. That said, three years of upgrades for 99$ is reasonable, even if there were only bugfixes
> not OP but I am not thrilled by "x months/years of updates" because you pay upfront for updates that you're not really sure that will happen. Will I get bug fixes? Will I get new features or at least significant improvements? Will the two person team work on the other project?
Thanks. I guess I'm just not really seeing how that is any different to what we did before? You'd buy a ST3 license with no knowledge of what improvements would be made to ST3.
It's been a while since I used my ST3 license, but I remember my license affording me all updates for ST3. Maybe that model was unrealistic for ST4, but as others have echoed in this thread, major feature updates for ST4 are not common, so if you bought a license in May of 2022, your access to new features would have expired right before the new updates of the May 2025 release.
Personally, I'm OK with using an old build so I don't mind that much about the limitation. Although if my 3 years elapses right before ST4 introduces first-class LSP support and an official Debugger, I may be very peeved. :)
> It's been a while since I used my ST3 license, but I remember my license affording me all updates for ST3
That's correct, but it could be the day after you buy the license is when ST3 stops receiving any updates and we dedicate all work to ST4. With a ST4 license you're always getting 3 years of updates, ST3 it was anywhere from 3 to zero (not counting the beta period, ST3 only got 3 years of updates).
> Although if my 3 years elapses right before ST4 introduces first-class LSP support and an official Debugger, I may be very peeved. :)
I think we can all sympathize with some buyer's remorse. Unfortunately the line needs to be drawn somewhere. Maybe you can take solace in that we probably can't put multiple huge features in a single update, at least not a dev release. :)
the difference is subtle, and comes mainly from the fact that at the time the concept of Major Release still existed.
You got everything in the current Major Release and every minor and patch update to it. If it took more than three years to release a new Major it was not a big problem, because you were still covered.
The new model might leave you without bugfixes before a new major version is out.
I think you might be misunderstanding the new licensing model. There are no major versions anymore; features & bugfixes just get released when they're ready.
This happened to me and I tried to recover the last licensed version I had used but mixed up my shortcuts or something and, after the 100th time I saw the nagware screen, I gave up and uninstalled and went with something simple and free: Notepad++.
How is Sublime faster?
ST is also all but dead.
They switched to a subscription model (3 year licenses are still subscriptions), and since the release of ST4 in 2021, there has been exactly one release with new features (May 2025). All other releases have been bug fixes and "improvements".
I get that developers need to make a living, but 4 years of fixing bugs in your products is probably not what I want to be paying for, at least not when that is the only thing I'm getting. Speaking of releases, they're also usually 6-12 months apart.
I have used ST ever since the first version replaced TextMate for my use (TM2 spent something like a decade catching up to ST2), but I've since switched to Code and Zed (mostly Zed as of late, Code on windows until Zed is ready there).
ST was great back when it was still an actively maintained product, but in recent years (ever since ST2) it has felt like it was mostly on the back burner and other editors have passed it in functionality.
As for VC funding, it has done miracles for Code to have Microsoft sponsor it (and others). Code is currently the editor to beat for anything that doesn't involve opening large files.
> They switched to a subscription model (3 year licenses are still subscriptions)
Licenses are perpetual. It is not a subscription. Don't like the work we've done? You can continue to use that version of Sublime Text until the end of time.
> there has been exactly one release with new features (May 2025)
August 2024 we added kinetic scroll and xdg-activation support for Wayland; we also added the ability to configure image extensions and allow dynamically switching between the hex-editor and image view.
November 2023 we added native font dialog support for switching fonts.
August 2023 we added webp and proper support for running as administrator on all platforms.
November 2022 we added syntax-based code folding and operating system recent file integration
December 2021 we added GB18030 support.
I'll stop there. Those are just the largest, most user facing new features, not any of the new settings, new APIs or improvements that I'd argue are new features.
You can read the full changelogs here: https://www.sublimetext.com/download
> Speaking of releases, they're also usually 6-12 months apart.
We do stable releases infrequently, because they're stable. If you want more frequent releases you can switch to the development releases. You can see from the build number how many builds we've done of ST4 since the release: it's around 150.
> You can continue to use that version of Sublime Text until the end of time.
Or until some change in the underlying OS makes it no longer able to run. Not saying that happens frequently, but it does happen.
> I'll stop there. Those are just the largest, most user facing new features, not any of the new settings, new APIs or improvements that I'd argue are new features.
New font pickers, settings dialogues and other "polish" is hardly what i'd call new features. I listed those as "improvements".
I'm not trying to start a fight, and i respect the work that has been done, but VS Code releases more features in a month that ST has released in the entire ST4 lifecycle, as does Zed. I understand the team is (a lot) smaller. My frustration is that when a new release comes out, it's mostly polish, and not fixing the real problems, such as lack of integration and plugins.
You may choose to brush it off as just an old fart rambling on the internet, but i'm not alone with this opinion, and i have ST licenses dating back to the original version, and i have chosen not to renew my ST4 license. When you start losing the "religious" users, maybe it's time to reevaluate ?
Please don’t conflate limited updates with subscriptions. The problems with subscriptions are that the company can take away your own files, the company can take away your software, the lifetime cost is extremely high, and the company can unilaterally change the deal or stop offering one. None of this applies to limited updates offers like Sublime Text’s. You pay once and keep it forever. The three year limit is on the time into the future for which the company continues to add to what you’re keeping forever. Of course this isn’t unlimited, it’s pay once for the program, not pay once for the lifetime servitude of everybody who works on it.
It does look like ST is lost here. They don't know where to go next. But I do like their Sublime Merge product. It's really good.
That no major AI players even bother to announce a ST plugin/package is proof enough, its now just a tool to manage random text copy/paste snippets.
vscode seems to have totally taken over dev mindshare these days.
They also don’t bother to announce a Vim or Emacs one either. VS Code provides good default and most people don’t care about editor fluency. Which is why they keep using it.
>>They also don’t bother to announce a Vim or Emacs one either.
vim has a universal and in many ways a eternal use case. You have to edit a file at some point on a server, be it a self hosted or even on ec2. Thats kind of the only real use case for vim.
In these days of AI assisted coding, no one really 'edits' code. A lot of editor short cuts and fluency related concepts kind of in many ways are not relevant in this paradigm.
The thing is vscode just works, like just works, for nearly all the usecases. In case of emacs, learning it and mastering it takes lots of time in ones career. In case of vscode you don't have to do this, you can straight away work on the project that you want to get done.
emacs is some what like a massive distraction from the actual task you want to achieve. Instead of writing code to build a project, you have to first write code to make emacs work, then use emacs to write the project code. In vscode you just write project code.
I don't know if you're trolling or not, but there's one thing that VSCode and nearly all other "normal" editors don't have and I want: Non-tied Windows (pane) and buffers (opened files). One of my most used layout is one main window and two smaller ones. Layout like this are my mental frame, but what I actually want to look at may vary at any moment. It may be a test result, a git diff, or going down a reference link. It's like a moodboard instead of a stack of paper you can only look at one at a times.
Emacs and Vim has this built-in. Other editors kinda have that, but it's clunky. I can suffer IDE because they provide a lot more than editing.
> Instead of writing code to build a project, you have to first write code to make emacs work, then use emacs to write the project code
That's only done once. It's like adjusting the mirrors and seats of a car. Once it's comfortable, you don't have to touch it. Using VS Code feels like borrowing a car with a very limited range of adjustments. Why is the explorer on the left and the terminal at the bottom? Why are they always there?
Have you even ever watched someone experienced using Emacs or you're making assumptions on your (I suppose limited experience)?
The "distraction" framing assumes everyone has the same preferences and working style, I for one find VSCode (and IDEs in general) massively distracting from productively solving many tasks. No, it's not "a skill" issue - I have used InteliJ every single day for almost a decade, diving into some profoundly advanced and non-documented features, and I do open VSCode from time to time.
I feel your argument conflates initial learning curve with ongoing productivity, and assumes VSCode's approach is universally optimal rather than just different.
>>Have you even ever watched someone experienced using Emacs or you're making assumptions on your (I suppose limited experience)?
I am one of those experienced Emacs users myself. Wrote more stuff in Emacs and even vim than most devs today will even write code over their careers.
Its just vscode now does simply too many things out of the box, you obviously can recreate that in Emacs, but its a pointless exercise. Time consuming, and distracts your from your real job. My job is to write code, not build emacs to write code.
I totally stopped using Org-mode, because Google docs do it way better.
At some point you have to move on. For some people like that point arrived a little early.
> Wrote more stuff in Emacs
Can you name Emacs packages you've authored, maintained or contributed to?
> vscode now does simply too many things out of the box
Oh yeah? Can you edit your filesystem in VSCode like a wiki? Changing directories and filenames as if you're editing plain text? Can any of thousand of its extensions provide "indirect buffer" experience? Can you bind a double tap of the comma to navigate to the last error and automatically fix it? Can you at all bind to double/tripple comma or just about any arbitrary key based on context in VSCode? Keyboard macros with counters - e.g. to transform lists into numbered lists? True buffer based (not file oriented) workflows? Occur-mode style search and replace? Comint style process interaction? Are there any extensions that allow you to move the cursor to the piece of a plain text like "rfc-3540" or "myproj#4044", intelligently recognize it and allow you to browse the RFC document or review a Pull-Request - Emacs does. Can you run a shell command and pipe it into a buffer, or pipe the content of a buffer through series of shell commands and into another buffer?
> I totally stopped using Org-mode, because Google docs do it way better.
Don't be ridiculous. They are completely different classes of products. If you are even having to compare Org-mode with GDocs, perhaps you don't know well either or both. GDocs can't manage my dotfiles. I can't use it to annotate books or academic papers. I can't have inline LaTeX formulas in GDocs. It can't let me control a videoplayer and type my notes at the same time. I can't use it to manage my flashcards or keep the log of my LLM interactions there. I can't have executable blocks of code - in org-mode I can explore APIs or send SQL queries while passing them into data transforming code blocks. There are no TODO states/keywords in GDocs, no agenda views, no scheduling/deadlines, no time tracking/clocking, no habit tracking, no pomodoro timers, no dynamic time tables, no tag-based filtering, no capture/refile system.
> At some point you have to move on
Like I told you already, you're assuming too broadly. Both VSCode and Google Docs are excellent products - no denying that, but they are not universally better. If you came to the opposite conclusion, it is yours, and yours only.
Zed works on Linux/Win/MacOS. I'm also frankly skeptical that ST feels that much faster, Zed is pretty darn fast, far faster than any Electron app.
Zed only kinda works on Windows. As of today when you click to download windows version you can only sign up on beta program that I assume allows only select people to use that version.
ST is not electron..
The problem with accepting VC money is they will eventually demand a return on their investment, which means that the forces that drive enshitification will eventually come for Zed in some form. I suspect that we'll see more and more features locked behind a paid subscription and the open core of the editor will become neglected over time.
Here I am on my free-as-in-freedom operating system, making commits with my free DVCS tool in my free programmable text editor, building it with my free language toolchain, using my free terminal emulator/multiplexer with my free UNIX shell. VC backed tools like Warp and Zed that seek to innovate in this space are of zero interest to me as a developer.
I might HAVE to learn EMacs (prefer over Vim) because I think eventually everything else will be tainted by mandatory AI features and/or subscriptions.
And Zeds multiplayer features might make it so your workplace mandates Zed if you're unlucky and Zed succeeds with their plan.
Zed is fully open source. Fork it. The code is pretty nice, too, easy to understand.
VSCode is open source, too, but it's been pretty easy for them to keep forks from taking off by having proprietary extensions, a "markeplace", and other lock-in.
All they have to do is only permit official builds to talk to official builds (for security, of course ;-), and forking Zed becomes a lot harder.
I really dislike the dismissive "fork it" response. You do know how much time and work it takes to maintain your own fork of something, right? It's great that it's possible, but to expect most people to do something like that is absurd.
I would be willing to use a very old version of neovim or sublime text with only bugfixes.
Not sure how many people would use that though
i do know how hard it is, that doesn’t change anything. nobody is forced to use something like zed and if you think it’s important enough to stress about the product decisions they make, the open source license gives you freedom to decide what to do about it. anyway you don’t even have to maintain it. you can take gpl software you like today, build it for linux and with a docker container you’ll be able run that same binary for the foreseeable future. then you can choose to extend it or not
if you haven't checked neovim out, the Lua based config is really nice and easy to grok these days. 10x better than classic vimscript!
A year or two ago I moved away from one of the neovim distros when they randomly changed all the keybinds on an upgrade (such things really anger me) and set up my own config. Funnily enough, I preferred vimscript. I still do use lua of course for various things, but those just go in lua EOF blocks in the vimscript. Vimscript is really terse and convenient for many things, I love it.
I’ve always maintained my own configs for (neo)vim. The only area where I prefer vimscript is with certain incantations for which there are no lua-based alternatives. And those are increasingly rare.
Authoring plugins is a lot more attractive in lua, imho.
You can try Helix editor, it is super underrated editor. I always wanted to go down the vim/nvim path but just couldn't stick to it, especially with nvim. Helix configuration is straightforward have some pretty nice built-ins and it is the fastest/snappiest editor I have used so far.
Nothing against emacs, but check out NeoVIM. If you like Emacs, you might like NeoVIM and its powerful extensibility features.
What makes NeoVIM emacs-like?
It's not emacs-like. But a lot of plugins wants to adopt the emacs philosophy of having it open for the duration of your login session. Instead of the quick edit and be done of standard Vim.
Eventually you will need a text editor with your emacs :)
I hate to break it to you, but emacs was a product of the MIT AI lab.(prep.ai.mit.edu anyone?).
classical AI and modern generative AI are VERY different beasts. also, there isn't any AI in emacs itself. It was a tool built to make a job easier.
That’s fine as long as they don’t force AI prompts to me.
To clarify, I use AI agents, but I absolutely hate them submitting code in my editor. Chatting is fair enough and useful, but I need to turn off the auto-generating code part.
Sure, but given the existence of vim/nvim, emails, visual studio code, cursor, etc the price for editors has largely been driven to zero, or at least capped by what JetBrains charges. My concerns are more this is a big bet on a different thing, not the editor (which is quite nice, even if using typescript regularly makes it balloon to 15gb of ram), making them a giant pile of money. With the editor as a free complement.
Note to Zed: I prefer paid products to enshittened ones.
Please please please, get paid rather than holding on too tightly to making things free forcing future enshittening.
We're working on it! :)
You can pay for Zed today if you'd like - https://zed.dev/pricing - and also the editor itself is open-source under the GPLv3 license. So if at any point in the future Zed changes direction in a way you don't like, you are perpetually free to build the version you liked from source (or make a community fork and take it in a different direction).
This will sound a bit odd, but I don't necessarily want to pay in a way that makes it look like I'm paying for the AI rather than the editor.
No, it's not odd, I like this take a lot. They almost need a dropdown menu indicating where you'd like your money to go to (editor, LSP, etc).
I'd also much rather have a means of paying a single flat fee to indicate my support than yet another subscription which is misleading because I have zero interest in the AI components of Zed.
Yeah, same here, the least interesting aspect for me.
I'm enjoying using Zed, and I do pay for tools I use regularly. It's now approaching a month of daily use for me and I don't see that changing. But to echo the other replies, I'm uninterested in the LLM tools. I don't use those normally and paying for that as a way to support Zed would send the wrong signal. You have to be careful when you restrict what people can pay for, because that will become what you optimize for which may not be what your users actually care about.
Zed is fast, easy to configure (so far, maybe some hard parts I haven't run into yet), works well with the languages I care about via LSPs, and the collaboration features are compelling. I want to pay to support that, I don't want to pay for an LLM feature I don't care about that ends up distracting from the progress on the things I want to see maintained or improved.
I already pay for Zed Pro, but my fear (and likely GP's as well) is that this doesn't provide enough revenue for the team.
Since switching from Emacs last year, I have absolutely loved how this editor has evolved, and I am looking for any way to directly support the effort. I have been a Zed Pro subscriber for quite a while now, and I have started trying to contribute to the codebase, but I really wish there were monetization options beyond making a spread on Anthropic API pricing.
I would also pay for Zed (the editor) and not the AI. But I also don't want to pay just make Zed more attractive for VC acquisition.
You don't have to give me any more features than what's in the free editor. I would gladly pay up to $300 just to have a "license".
Big fan of Zed. I want to echo a sibling comment that I don't see that as paying for Zed, I see it as paying for LLM usage. And since I already have my own LLM keys, I just use those instead.
I can't pay for Zed Pro using my work funds because it's an unapproved AI service. Can you provide another way to pay? E.g. cross device settings sync or professional support.
I'm a big fan of Zed and having met much of the team I think it's some great people building a great product. But I do echo concerns that while the intentions are all honorable the incentives of the pricing structure, business environment, and now a funding round are concerning in the long term. I don't think anyone at Zed has a single ill intent or a secret master plan but these days anything I'm not paying for I just assume is going to be enshittified eventually. Especially for an app where the only paid features are AI-centric and there's a VC expecting to make multiples on a $60m investment.
So here's my ask: let me pay for it without paying for AI! None of my use cases will stress your servers; I have `"disable_ai": true` in my settings.json. Give me a $5/mo "support the devs tier" or a $10/mo tier with some random app quality-of-life features and I'm there. I specifically want to pay for good software without paying for AI to signify the value proposition that still exists there cause I don't think a VC would believe me otherwise.
With Shaun Maguire's recent behavior associating with Sequoia is gross
Unless they have very unusual terms on their funding, it isn't really entirely in their control in the long term. Hopefully they find a way to make their investors whole that doesn't suck for everyone else, but if not, well, I at least appreciate that the editor is truly open source, since at least it offers a contingency plan in the worst cases.
If I'm wrong I'd love to know, but I think that we need to start talking about what funding really implies more honestly. It's traditionally met with unabashed enthusiasm and congratulations, which I totally understand, but it's a mutual exchange, not an award or a grant. I absolutely believe that everyone wants to make good on their promises, but promises made to users are not legally binding, and the track record for upholding those has not been great. Plus, as a user, I want to pay for software, but nothing feels worse than paying, then watching enshittification unfold anyways... When this happens, the investors should send you a nice postcard thanking you for paying back some of their money.
Can $20/mo sustain a text editor company with a massive multimillion dollar valuation? Well, we'll see. Good luck Zed Industries, we're all counting on you.
you don't have to choose. paid things get enshittified just as easily as free things
I mean, eventually, sure. It took Uber around 15 years to get to profitability. ChatGPT came out in 2022, so get your predictions for 2037 in now.
For 2038 I predict Epochalypse and since there will only be vibe coders by then, good luck to y'all.
Please, tell me what it's like living in your free-as-in-freedom house, feeding your free-as-in-freedom offspring? Eventually demanding a return on your investment? The audacity!
Sure but does it mean that every business needs to be unicorn?
Yup, I was playing around with Zed and kind of liked it, and even debated switching over. But this kills it for me.
It was a cool idea while it lasted, I hope other editors embrace the learnings once zed is gone.
I doubt anyone will follow this.
1. Everyone else is building on Electron.
2. People still sleep on or dunk on Rust. There's a great deal of negativity here on HN for the language.
3. There's only so much Rust talent out there.
I haven't seen any negativity only push back. Rust became too hyped. I don't think Rust has a visibility problem right now. Everyone who cares enough already knows about it.
Also, it doesn't have to be rust, it could be C# or Zig or Lua or whatever
Whilst rust is a big part of this it is not the core of their innovation. Pushing everything to the GPU is.
Stating that everyone else is building on Electron belies your lack of exposure to any editors other than vscode \ atom spin offs.
Realistically all the editors including those based on Electron could easily shift to leveraging the GPU more and that would be big leap forward for performance.
> Stating that everyone else is building on Electron belies your lack of exposure to any editors other than vscode \ atom spin offs.
Enlighten me.
Same. Especially not having been familiar with who Sequoia is. Altman, Huang, Musk, etc.
What do you mean by that list? They pretty much have every single big tech company in their portfolio.
They have some house cleaning to do in their leadership before I am willing to use a product they back.
https://github.com/zed-industries/zed/discussions/36604
I have never seen so much virtue signaling and moral superiority in a GitHub issue. What a joke.
There is an active genocide going on as we speak. I am sorry if you find boycotting those who support it is virtue signalling and a joke.
Regardless of your political stance, how is taking money from a VC company that has a single person that said a single thing you disagree with “supporting genocide”? And how is uninstalling a text editor going to help literally anyone?
This is 100% pure virtue signaling. This brigading is not helping a single Palestinian. These people just want to feel and show the world that they are “the good guys”, while not actually doing anything or helping anyone. To me, that is absolutely a joke.
There is no indication that zed supports this genocide (accepting money from an entity who might != Supporting)
This is far fetched and that's why it cannot be taken seriously.
Wow, some people really don't know what to give a shit about, do they? It must be endlessly exhausting to go around looking for any reason to boycott everything, instead of using these fantastic tools to make things that make the world a better place.
You’re complaining that the people giving enough of a shit to boycott a thing don’t know what to give a shit about?
Point was they didn't have Zed in their portfolio until this announcement.
It does not support Monokai because of reasons. That does not scream good taste to me. It screams something different.
> "I don't think an editor should be VC backed"
it's a software company. they sell software.
Then sell software, not equity
This quote slaps!
I'm willing to accept this claim:
- You can make money when your product is a text editor.
I am very skeptical of these claims:
- When your product is a text editor, $42 million in capital can be effectively deployed to make meaningful improvements to your product.
- When your product is a text editor, your lifetime inflation-adjusted profit will eventually exceed $42 million.
Sequoia is apparently not so skeptical, and willing to put the money on the table. That must have been a truly extraordinary pitch deck...
> Sequoia is apparently not so skeptical, and willing to put the money on the table. That must have been a truly extraordinary pitch deck...
They invested like $200 million in FTX and had a glowing article about SBF about mile long on their website.
The big VC firms are by no means immune from just doing plainly stupid things.
Right now all the hype around anything even at the periphery of "AI" is enabling a lot of similar stupidity.
Counterpoint, apparently cursor's revenue is in the 300-400 million range. So it's not wildly inconceivable that you'd do 40m profit (although I too am doubtful).
I strongly doubt that Cursor makes anywhere near 40m profit. All they're revenue is spent on tokens with the LLM vendors. I'd be surprised if they are even running at positive margin and not just subsidizing usage with the VC money.
Unsure of what the end goal is, but I expect everything AI related to be a load-leader right now and then the goal being to figure out how to drive down costs or make even more money later.
Maybe that's what Sequioa thinks too...
how much profit does Cursor generate on that revenue?
If I had to guess ... I would say they spend more on Ai tokens than they take in in revenue.
If the world is moving towards open models which are "good enough" – the main winner will be the one controlling the distribution. So if you're controlling the web browser, the editor or the OS – you are the winner.
Obviously, the risk here is very high from this perspective, since nothing guarantees anything.
Zed isn’t just a text editor. It is the only working platform for code assistants.
Neither Google nor Claude nor anyone can’t at the moment get right basic operations like file edits. Zed is flawless in co-operating with most LLM models. And not just that - also switching models during conversation and more.
I am at Zed Pro at $20 but when Zed offers $200 Max plan I will sign up right away.
I switched off of Zed's agent system onto the Claude Code CLI because I was blown away by how much better CC was.
Even with "auto edit" turned on, Zed just kept asking me for confirmation. I'd be like "hey your code has this bug", and it'd be like "you're right, and this is why. here's how you can fix it" ??? just fix it man it's your code. Maybe this is fixed now, but Claude Code never has this issue and is very good at only stopping when truly stuck (and generally for good reason!)
Changing the topic a bit, Zed's collaboration features seem really good but it's quite hard to use when nobody uses it in the first place. With VSCode, I can use the LiveShare extension and everyone on the team can just join with no fuss at all. LiveShare is likely not nearly as technically great, but the simple fact that people can use it easily makes it win hard.
Honestly it would be cool if Zed can somehow become more popular thanks to this investment. As long as it can keep its speed and technical excellence. VSCode used to be super lean and cool, but now it's just another fat IDE with unlimited bells and whistles. It feels like how Eclipse felt back in the day.
How does this compare to the rust rover with Junie? it is definitely well integrated for code edits and iterations with the IDE
No....there is value in having many developers use your tool for daily code development. You should think if it as a lot more than a text editor as well, just like Github is a lot more than just hosted Git (also VC backed at first).
They don’t sell an editor, they sell editing.
What they are really selling is an AI service subscription.
They are giving out the actual text editor for free.
AI text editor
I love working in Zed. It really is a delight to use, and I think the Agentic coding integration is really well done. I'm excited to see them investing in this space more.
I understand people's concerns about VC funding, but I think building quality products takes capital. The funding is still relatively small, especially when you compare it to players like Cursor, etc. And I think Zed is a much, much better product!
Zed being OSS is a gift to the community, and I suspect DeltaDB will be as well. And as others have said, Nathan (CEO) is a delightful human.
Congrats, Zed!
> I understand people's concerns about VC funding, but I think building quality products takes capital.
When VC funding comes in it stops being about building any products, because the company itself becomes a product.
Maybe you can comment on this. I saw in the Zed demo that they're using "rewrite optimized version Claude". My first thought was wondering how big the bull is going to be.
> And as others have said, Nathan (CEO) is a delightful human.
Hate cancel culture as much as the next techbro, but since y'all point out they're "a delightful human", it'd be interesting to see how Nathan then responds to concerns raised in some quarters: https://github.com/zed-industries/zed/discussions/36604
With so many new AI editors popping up, it feels like the Zed team is in a tough “lose/lose” spot.
If they stick to their current path—focusing on craftsmanship and letting their world-class talent build the best editor—they risk falling behind giants like Cursor. Cursor seemingly came out of nowhere and is already doing $300–400M annually, rivaling JetBrains (who took 20 years to get there) in just two years. With that kind of momentum, Cursor can now buy their way into a superior product.
On the other hand, if Zed takes more VC money, it likely means doubling down on AI in ways they clearly don’t seem eager to—but at least it would give them a fighting chance.
I really feel for this team. It couldn’t have been an easy decision, and from the few interactions I’ve had, they strike me as incredibly talented, kind, and genuine folks. I truly wish them the best.
I hadn’t thought about it that way but it’s an interesting point. The cursor folks probably have 99% of the work already done for them, for free, funded by Microsoft, probably in perpetuity.
And on top of that, since it’s essentially just vscode, it costs users almost zero effort to make the switch. It’s the perfect crime!!
I hope zed does well though. I love their blog, and all of the cool open source stuff they’ve made. I recently heard they added a “helix mode” which might be enough to get me to switch from vscode…
this is where the advances in collaboration come in; it's a way out for them because existing editors are largely stuck with git and are patching ai on top of everything.
zed came at the incredibly (un?)fortunate timing where they were just able to build a solid base before the editor wars began. their only path now is to fully maximize the few advantages they do have:
* a fresh base that is far more flexible
* really good experience with performance, design, general craftmanship
* a buzzy community and fresh/boldness that attracts vcs
for zed to truly win (at least in sequioa's eyes) they will need to completely take over vscode as the new default, and that will require a big lead when it comes to collaboration and ai
> Cursor seemingly came out of nowhere and is already doing $300–400M annually, rivaling JetBrains (who took 20 years to get there
Revenue, not profit. I’d imagine cursor loses a dollar for every 50 cents of revenue and makes it up in volume. Meanwhile JetBrains is a profitable company not beholden to the whims of outside investors.
I know which I’d rather work for…
Related ongoing threads:
Zedless: Zed fork focused on privacy and being local-first - https://news.ycombinator.com/item?id=44964916
Zed for Windows: What's Taking So Long? - https://news.ycombinator.com/item?id=44964366
Tbh: I think git is not the century-tool that a lot of people think it is, and we're in an era now where its decreasingly serving the needs of its users. I'm ready for deeper ground-up innovation in the space of code collaboration.
Developers don't think in terms of commits; they think in terms of Tickets and PRs. Git doesn't have any native representation of either of these things. Getting tickets into the repository is something the community has talked about for years, and Linus himself has said he wants, but it hasn't happened yet. Branches work fine enough to start a PR, but then Github, again, has to take over for comments, CI execution, even the decision on how you want the commits on that branch represented on the main branch (squash? merge? rebase? why should I care?). LFS has been years in the making. Monorepos are still weird.
Idk, I think there's capacity in the industry to take the concept of a repository back to square one and think more holistically about how we all interface with the repository. The way they talk about DeltaDB gives me hope that they'll start thinking more about this with this funding; commits just aren't a useful primitive anymore, and a repository, locally-downloaded, open source, ultratight connection between diffs, merges, communication, and automated tooling is how I want to be coding in 2030.
Totally agree. git notes might help a bit with some of those points, but what most shops need / want is something like GitLab. git itself is usually too low level.
Whether this is good or bad for Zed remains to be seen.
I wonder though what the play is here for Sequoia - like all VCs they’re looking for the possibility of a huge return.
I don’t see how “just” an editor (even with paying users) can generate a 10x return. So what is the larger vision here?
JetBrains would be a good case study. They also arguably have a lot more IP.
Is IntelliJ "bad"? Aren't the reactions here overly negative?
This means the company is funded, development will continue, zed will continue to improve. An IntelliJ style license (for example) is an acceptable trade-off from my point-of-view
Not many people are willing to pay ~$300/year for an IDE. And Intellij didn't take any VC funding.
So at some point Zed will likely need to pursue monetization more aggressively than IntelliJ does now.
The thing with taking VC funding is that your intention usually is not to steadily grow a sustainable product.
You take the funding so that you can outgrow the competitors and get the market faster. All you need is the small promise of innovation in an area which is somewhat new. At the beginning, the product is good enough and you have money to keep marketing and developing slightly faster than others. This will get you the users.
In the end, is your product at that point truly the best among competitors? It matters less, since you already have the users.
I think Zed didn’t need this one since they had a great product. Many would have been ready to pay at least a little. They could have grown slowly and see what works. With VC money take can go to completely wrong direction with giant steps and they are not noticing that unless it is too late. And then investors want returns.
Cursor costs $240 per year and loads of people are paying for it
I doubt there is much overlap between people happily paying for cursor and people upset Zed took VC funding.
IntelliJ has always been extremely slow for me, even on my beefed up mobile workstation for work.
It’s refreshing to see an editor that’s built with performance as a priority.
IntelliJ lost the plot at the inception of CLion etc.
I was a customer for so many years. "One IDE to rule them all" and then they started cashing on more.
Progress was down to a crawl, performance down the shitter and bug reports go unnoticed for 2+ years.
VSCode poops on IntelliJ these days for everything but the UX; but with enough modding, it can be very close.
Another big point was they implemented their own parsers for everything which allowed them to make nifty things - the refactor features way back in the early 10s was miles ahead of everyone else - but then LSP happened and that advantage is diminishing and becoming a liability
>IntelliJ lost the plot at the inception of CLion etc. I was a customer for so many years. "One IDE to rule them all" and then they started cashing on more.
What are you talking about?
ReSharper came out 21 years ago 3 years after Intellij. RubyMine came out 15 years ago. 7 years before CLion.
I don't write Ruby, but I write Go and C, and C++ and I was left facing a new license. For no reason at all. It's the same debugger and the same code base, you just need to hook into gdb or lldb instead of all the other ones.
Like I said it's only one of the problems, read the rest.
The paid version of Intellij has never lost anything. Pretty much everything the specialized IDEs can do, Intellij can do too, though maybe some features lag. CLion, Rubymine etc are just less expensive specialized versions.
I'm sure the free version has lost some things.
Its been a long time since I used CLion but it was the best C++ IDE by a huge margin.
I agree CLion is the best as C IDE, but there is no reason that intelliJ couldn't do C/C++/Go, other than cashing in with new product lines and licenses.
It can do Go.
Only ones that it can't is C/C++ (probably because that started as Apple only IDE) and .NET (that started as extension for Visual Studio).
They started Fleet that can do all of them, as a response for VS Code. But then came AI and Cursor probably took a large chunk of their users. So they redirected resources (like Zed that paused their collaborative features to focus on AI). There is currently a closed EAP for Cursor like ide based on Fleet. And their Git client as separate app.
The C++ plugin is included in Android Studio which is essentially a distribution of IntelliJ. There seems to be no technical reason for not allowing it in full IntelliJ.
>>Is IntelliJ "bad"?
The days of using a separate IDE for each language are kind of over.
These very paradigms are outdated these days. vscode got it, very early. vscode works for everything. Most projects use Python/Go and JS, and out of the box vscode just works for all these languages and their tools.
> vscode works for everything.
IntelliJ did that before Atom even had ‘git init’ run on it.
VC funding and self funded are totally different beast. Self funded are organic and they can follow their own vision , not VC's vision.
Ever since the UI redesign they've lost the plot a bit.
Counterpoint: the updated UI looks and works great, and their software has never felt higher quality.
DeltaDB sounds of being a >git innovation for coding itself, and would fulfill Zed's promises in Nathan Sobo's debate/discussion with Steve Yegge recently.
Seems to solve a real problem which is growing rapidly, both in the old way and in the new way ... if it can overcome _slop_ in LLM chats, and the sheer enormity of code/data ahead. Trying to picture how coherence will survive.
With claims/hype/concern floating around that >90% of code will be LLM-generated within 3-6 months, with the insinuation/tone [1] that the same amount of code will be written by humans as now ( at least at first ) but LLM code will radically grow to dilute the space ( as is happening ) ... seems like DeltaDB being done right/well is going to be do-or-die on whether coherence remains possible!
[1] https://www.businessinsider.com/anthropic-ceo-ai-90-percent-...
For anyone not clicking through the link: it's from 5 months ago, predicting 90% of today's code will be from LLMs.
>I think we will be there in three to six months, where AI is writing 90% of the code.
They're going to keep saying it because it's a juicy sound bite and they're sales people. That doesn't make it any more true than "9 out of 10 dentists recommend our socks" or how we surely have all had flying cars for decades now.
https://allthatsinteresting.com/wordpress/wp-content/uploads...
https://allthatsinteresting.com/wordpress/wp-content/uploads...
By sheer volume, that 90% number might be right.
I agree. The level of dilution is becoming obvious and a big problem... code is becoming predominantly disposable
Despite the article being salesmanship hype ( at WEF no less ) we are now in the time mentioned, and can feel this
The idea that the code is GOOD or even being used is not necessary to be saying that it exists, strewn everywhere
I had a week where I embraced and went deep on ai first coding (not vibe, that'd be crazy)
The thing wrote at least 80%, so we aren't far off in this anecdotal instance. There are citizen devs who are building fun things for themselves where the AI does 100%
It made me realize these things are more capable than I knew, though they still do dumb stuff reliably. But, it is easy to undo those changes, so the productivity boost remains
Could you link to the Sobo/Yegge debate? Is it this? https://youtu.be/j2goZBL156Q?si=S1A2XO_HxL7fpeVU
The problem I see is the VC involvement...
VCs operate from the goal of xtreme high user market share as the problem..
As frameworks get better, the audience using the IDE changes...
Both are misaligned before even meeting and it will get worse once VC money control is added.
My bias, I am a flutter framework user and MS VSCode user.
for those who just caught the headline, the real gem here isn't the investment, but what zed's doing different: rather than using git for version control, they're announcing deltadb which incrementally stores all changes as crdt [1] while being interopable with git.
this feels pretty important; git is definitely not the right primitive for version control with ai and that pain is obvious with existing solutions. zed seems to be going all in on collaboration with realtime, git, and now something in between and it'll be interesting to see where they end up-to me three solutions feels overcomplicated but that may be necessary given how teams work now.
- [1]: https://zed.dev/blog/crdts
Ehh Zed tried to push the collaborative editing before the AI hype was thing (that blog post is from 2022), and seemingly everyone outside of Zed developers promptly ignored the idea.
The Zed developers quickly refocused on milking the AI cow.
I'll argue git is plenty good, and there are a lot of people who don't understand it very well. The ones talking about Github PRs as if they were related to Git in any way most definitely don't.
Have they fixed the big/feature where it insists on downloading entire distributions of Node.js for running it's language server functionality?
This is basically my main gripe with Zed atm — it's very keen to autodownload and execute binaries.
I have a light fork that tries to nullify this, but I don't think I've managed to catch all the instances.
Other than that, it's a very nice editor in my opinion.
Easy fix:
{ "server_url": "", }
I comment out that JSONC line periodically when I feel like cherry-picking updates
> it's very keen to autodownload and execute binaries.
I hate this pattern in software so much.
That is the great thing about Zed. Stuff just works.
That's not great at all. It's not acceptable to "just work" by downloading and running third party software without asking the user.
> we've been building the world's fastest IDE
Any data to backup this?
Well, last year, Nikita Tonsky measured it and it did worse than Sublime: https://mastodon.online/@nikitonsky/112146684329230663
It's funny how often people buy into the marketing and say "blazing fast" without actually questioning it. FWIW, I still prefer Zed because its LSP integration and vim mode are better than Sublime's.
It's commendable to try and challenge Sublime, the undisputed performance champion in GUI text editing, but making false claims is a massive red flag.
I know it probably didn't, but I wonder if part of Sequoia's decision to invest had anything to do with these false claims?
I really love Sublime. If it had a solid "remote-ssh" type integration and a more straightforward way to manage plugins/extensions that isn't a rats nest of random directories with similar names - I would return to using it. I purchased every upgrade from like 2012 to 2022 or so.
Yes if a software markets itself like this without backing up I smell BS.
Like any company now is "global leader in X market".
You can try it out. I would say it’s aiming to be a more modern Sublime Text, which is a win to be considered in the same category imo.
I have tried it out and by default it was so slow as to be unusable. After discovering it required some customization in /etc (because it's the only GUI application that fails to recognize my GPU on a very popular distro with next to zero customization, because I game a lot on Linux - weird how that's a me problem and not a Zed problem) it got better, but still noticeably slower than VS Code.
The modern Sublime Text is Sublime Text. There is way too much "extra" in Zed to compare it. If anything, it's a new IntelliJ.
Slower than VS Code?? I guess it's just poorly optimized for Linux. On MacOS, I find it to be significantly faster than VS Code, and the only alternative I've found that's in the ballpark of Sublime (from the performance aspect).
I'm on Linux and performance is the reason I switched from VSCode to Zed. It works great for me.
Perhaps the same issue that had Zed fail to recognize their GPU also tanked their performance.
> If anything, it's a new IntelliJ.
I do not understand how one says this with a straight face unless you’ve never used either product. Do you also believe Neovim to be a “new IntelliJ”? That’s the level of functionality out the box, though fortunately Zed does not require lots of screwing with config files to get basic things working.
This is accurate. I came over from Sublime Text because it had become laggy over >5 running instances, and native LLM integration. Even VS Code doesn't actually have that... where everything is an extension versus seamlessly/perfectly fitting
As mentioned in other comments, it actually outperforms window management in general in many/most cases. Radically flexible and almost never gets in the way
Just to clarify: I'm not saying I doubt they are the fastest, I don't know and I don't care. What I do care is that a company (a VC backed one specially) that markets their product like that follows the statement with proof, otherwise we should call it BS
Oh well, back to Sublime Text.
RIP Zed, you had a good run.
Met the CEO of Zed. Very humble and deeply technical. Glad to see they're doing well!
I don't know what's DeltaDB, but if it will be able to show when a commit is simultaneously a refactoring and a change of code being refactored (moved into another module) – I would love to see that! At the moment if you refactor you punish the people reviewing the code, which reverses the motivation for cleaning things up.
> To make this possible, we're building DeltaDB: a new kind of version control that tracks every operation, not just commits.
Let me guess: DeltaDB is free to use as long as we host your data and have free range on training AI based on your editor interactions.
Not sure what this guess is based on. Would that be a guess for git also, if mentioned by a company versus an individual?
My read was that they are pulling a Linus Torvalds with the Linux->Git move where both are innovations on their own, but work great together ( without dystopian universe instantiation )
CRDTs mentioned: https://zed.dev/blog/crdts
Definitely sounds very eerie. Luckily there are open source solutions that do just this with no AI integration:
https://github.com/atuinsh/atuin
how is atuin doing what deltaDB does?
ah not much at all, did too much selective reading on my end. Should have read the entire blog and not the quoted selection.
this doesn't seem related to the above post
that said atuin is excellent
zed is just on the hype train, obviously very talented people, they are thinking hard about LLMs but I'm not really sure where they are going, their pivot is probably going to be more interesting...
Data harvesting and non-ownership.
The beginning of the end.
You know where this goes.
Sounds like a noble goal.. but I'm skeptical. I normally don't care about every single edit I've done, and don't want to store them and certainly don't want to see other people's character level edits.
Also how on earth do you handle conflicts?
This sounds most similar to IntelliJ's local history tracking, but that is local only which makes sense.
Still, more IDEs is always good so I wish them luck and will wait and see. (As with Zed - I keep trying it but it's still very alpha quality so I always end up back in VSCode.)
What on earth do they need this kind of money for?
feels like we are giving up privacy for productivity.
wouldn’t be the first time a massive investment from someone like sequoia sparks the death of a previously great tool/service
Yea, this announcement makes it less likely I will try Zed
I don't want chat with coworkers in my IDE, nor do I feel the pains they describe with conversations spread between tools. It's not a top 5 problem
although I tried the collaborative features I unfortunately cannot say that I used these extensively for now. Although I found them quite nice, well integrated, seamless, and straightforward!
The issue is again people, they don't wanna change their _archaic_ workflow, stuck with inefficient -copy/paste- loop to the chat (ie. Slack) and back.
The story in the article went a bit too far that I agree, but I guess that is their north-star vision. Current implementation allows you to "join" a workspace session shared by someone, edit the same or different file, follow/watch a certain person, as well as have a chat (without requiring copy-paste) about certain piece of code. (both written or via voice)
If something, large enterprises generally don't support smaller and ambiguous licenses. Therefore, if Zed will allow enterprise licensing (ie. via on-prem license server or volume ordering, SSO, whatever) that would increase their adoption quite well...
For the record: I have never used the collaboration aspects of Zed
What I also have not used is vim emulation, though I have a vim background
As mentioned elsewhere, Zed is still very configuration-dependent to get the full power of it, and a lot of its functionality is never discovered for that reason
What pushed me to try it was Ollama integration which is not an afterthought, then I realized I loved it _way_ more than SublimeText, especially on performance, at first, then everything else once that won me over
I have ~10 running instances at any given moment, and >99% of the time never feel any lag, whatsoever
Another unexpected benefit is that terminals, code editor panels, and assistant chats, get to be sized and fit wherever you want, so it is also kind of a window manager... I often have more terminals open in Zed than in the Window Manager of the OS itself
> then I realized I loved it _way_ more than SublimeText, especially on performance
I currently have 19 instances of Sublime Text open, each to a separate folder containing a mix of C++ and Python code bases (some tiny some huge). Like ~8 of those have the clangd LSP plugin enabled. I don't think I've ever experienced lag in Sublime. KDE System Monitor is reporting 2.0 GiB of ram being using by sublime currently.
The clangd LSP plugin in Sublime isn't perfect, and it does occasionally break, and rarely spikes in CPU usage for no reason (although the editor always remains responsive). But, if I ever switch away from Sublime Text, I cannot imagine it ever being due to performance reasons.
I do all those same things in VS Code, especially the vim bindings, wouldn't give those up, but did recently leave the vim ecosystem because I had to spend too much time making the IDE work or enable features that are out-of-the-box in a code-oss based IDE
You probably give up just as much privacy with VS Code as you do with Zed, no? Just sent to different overlords?
what? I didn't comment on privacy
I give lots of feedback to Copilot in the hopes it makes the agents better in the long-run. I want them to read my code and train on it, along with the interactions with copilot, which is the next frontier in (post) training
That's true. Sequoia is often a death sentence for power users. And a huge gamble for the founders
If you're using it at work, the company might decide the risks are outweighed by the increased productivity.
It’s sounds very similar to Google wave, if anyone remembers that, but for code.
That came back as Slack and other similar products
Nothing since then really recaptured what I personally liked about GWave or let me use their tool in similar ways to how I used it. YMMV, of course, more so than most of my comments.
This sounds interesting. Which aspects?
“ never-ending stream of conversations”
VC backed? Zed's dead.
The vision sounds interesting but I think if they have two challenges they will need to overcome to compete in the IDE space
1) Their LLM integration needs to be at the quality of Cursor and VSCode to pull people away from those
2) Reduced friction to move over (keyboard shortcuts, common plugins, etc)
I think the Zed team is perfectly capable of winning. The bigger risk would be them trying to tackle fancy stuff before making sure the basics are good enough to get developers to switch.
They need to build up a deep understanding of why folks are sticking to Cursor/VSCode and not swapping over.
P.S. I would love for Zed to win in the market because I'm sick of slow software and it's refreshing to finally see an intense focus on performance.
FWIW their default keybinds mirror VS Code nearly 1:1 with some (IMO) improvements
I customized the hell out of my keybindings. I might be in the minority though! I was trying to give examples of (stupid) little points of friction that might prevent someone from taking the leap.
I think they're tackling a real problem that is revealing itself via agentic coding, and I think they've positioned themselves in an interesting spot.
I'd be interested to know how much data DeltaDB accumulates over time - because the level of granularity is so high - and are they going to want to use that data as training data?
Aw man :( I too loved the promise of Zed.
Well done I suppose to the Zed founders as they're on the Sequoia gravy train now. But as others have noted this puts an inevitable clock on the enshittification no matter how hard the team crosses their heart and truly believes otherwise (or not, I mean maybe this IS the gameplan).
Hopefully Zedless [0] gets some community steam behind it.
[0]: https://news.ycombinator.com/item?id=44964916
While i am sure this sounds insane i had to drop zed due to lack of the “last file dif” button gitlens vscode plugin provides.
It is a godsend on quickly debugging the why of things. If anyone knows how to replicate the same functionality with the same number of clicks in zed, id happily switch back to it.
Seems like a minor thing to change IDEs over. Would a Zed task that runs the relevant git command work for you? e.g.
You can even throw a keybind on it if you'd like: I am not familiar with gitlens so not sure how close this gets you but you should be able to replicate the functionality you need from the git CLI and some light scripting. This can be a jumping off point maybe. If you want to view the diff using the zed diff viewer, you can do so using `zed --diff`, as demonstrated in this GitHub discussion: https://github.com/zed-industries/zed/discussions/33503#disc...i use the gutter markers, where a click shows the chunk-diff.
there is also inline-blame, both are native (ie. no plugins required)
Nice. I dont like the gutters taking up space. I like the ux of click, see dif of last change, click see the next last change.
It seems such a silly thing but it made me not use zed at least until vscode eats all my ram again.
I love Zed and I am glad they are getting funded. This allows them to truly polish both the remote collab and the AI features and compete with Cursor and friends.
The product has improved so much in the last year, it's been a pleasure to use and get excited about new features.
VC backing apparently has the power to move the standard from "hodge-podge 800-line commit must have an issue key" to "track every damn interaction and back and forth".
That and the ever-trailing "especially now with AI".
Is anyone actually using the collaboration features? Zed is my daily editor and I love it, but I've never touched the collaboration features, which is their "USP" when raising cash.
I enjoy the responsiveness of the Zed editor compared to VSCode, but I am concerned that it is a VC backed enterprise.
For example, I don't like that I am forced to look at its "Sign In" UI, and that they have refused attempts to remove it. [0]
Zed has some much more annoying bugs, and I am not excited to help fix them given the position of the code owners.
[0] https://github.com/zed-industries/zed/issues/12325
Zed defaults don't seem to capture the full value of what it offers. For instance, edit predictions via tab completion are documented yet I've never experienced them. I need better settings, I guess.
On the bottom right, there is a button looks like magic emoji (). Click that and it will show general AI settings. You can choose _Eager-Completion_ rather than trigger-based one.
As a user, most IDEs/Editors currently show in _eager_ way. For example, VSCode by default shows ghost-text as well as Amazon Q extension too. Usually disabling those disables the AI completion completely.
Meanwhile I like Zed's approach that you can trigger completions with Alt+Space, not burning through your "tokens" in free-tiers. They also provide a free-tier completions, as well as _next-edit suggestions_.
*) Next-edit suggestions: When you edit some piece of code repeatedly, it suggest to do similar on the next few instances, with context awareness, quite nice feature saving several keystrokes every single time.
Not a good move. Evernote went that path and then turned an otherwise very useful app into a VC bet that forced the founders to pivot the app into something nobody wanted. Same happened to Soundcloud.
I wanted to try out Zed. It looked fine on my laptop screen but as soon as I tried it on a 1440p monitor, the font and icon rendering fall completely flat and are a blurry mess. If your product is a text editor focused on code, your font rendering better be *top of class*.
https://github.com/zed-industries/zed/issues/7992
All of this data is valuable for the Zed team and could easily be compiled into reports for management about developer's individual progress and productivity.
I am certain the team will say that is not on the roadmap, but for me, it still sounds like a possible leak in the future.
Or maybe a semi draconian "mandatory" extension your bosses will demand installed.
If you need a simple text editor without DeltaDB bloat, see CudaText. No so fast drawing speed because it does not use GPU. But drawing speed is not the decision factor for many tasks of text work.
Zed has better drawing speed. But DeltaDB bloat, AI bloat, telemetry, need of modern videocard.
From the description it seems my deveopment experience would become a lot more crowded and furstraiting. All the different actors sharing a live coding experience. I feel my focus will become fragmented like someone poking my shoulder frequently.
Hopefullly intended use is to schedule a time for it.
Cool, I basically get the same effect with "code snapshots" with using jj so I wonder if this will be different.
As an aside, I really like the Zed website. Easy to read and pleasant to look at.
I love Zed, but this makes me worry for its future. VCs are not the end user, and thus don't understand the perils of enshittification.
deltadb sounds interesting. AI editors should look into letting you operate on multiple git branches simultaneously, like isolated browser tabs where the AI gen doesn't contaminate the other.
One of the reasons I find LLMs don't increase productivity much is that I can't switch branches to multitask while it's processing. Context switching isn't always useful but there's still lots of opportunities for rapidly experimenting or ticking off a couple small bugs quickly while AI takes the first pass on something more complicated.
Each "branch tab" could have a sort of TODO list or plan.
I have been running multiple instances of the Codex CLI tool in several terminals to do different stuff. You can even checkout the repo several times, write the todo to a file (or ask the agent to do so), if you need.
> AI editors should look into letting you operate on multiple git branches simultaneously
Git worktrees are great for this. I built a little tool to make them more ergonomic: https://steveasleep.com/autowt/
You really don't need every LLM vendor to build their own version of worktrees.
I have trouble imagining a world where I will ever pay for this. Unless it becomes as feature rich and powerful as an IDE like PyCharm, it’s not compelling.
OK they touched Unity and now Zed. I’m sure it’s going to be good.
Exciting ideas! I'll take a look at this again in several years when they aren't trying to rebuild everything from the ground up from first principles.
> Having said that, I don't think an editor should be VC backed.
It just means that we will now see AI crapola being stuffed into this editor in the next few days.
Have you heard of Zed before just now?
Great, maybe they can finally make Zed start maximized instead of this small window and add a picker for recent files (not projects)! ;)
bad news
"total funding to over $42M"
the enshittification it'll take to extract $400m of value out of a text editor will be dismal
Should bring some interesting new features to Zed quickly. Glad to see it.
Yeesh. With every announcement, I get less and less interested in Zed. I've yet to use it because Windows has been on the "soon" list forever.
"Fast" is what got me interested. All this other stuff sounds like "we want more money"
I do like Zed very much. Fast, good UI, decent set of features.
Still missing some key features to get parity for python with PyCharm, and git features need more love, but pace of development has been blistering. Kudos to the team for building something really solid.
I don't use the agentic stuff though -- I have a ClaudeMax subscription and use CC or more recently opencode (the only non-CC that works with a ClaudeMax sub); I don't want to pay another sub for Zed and the free model use that comes with Zed runs out very quickly (understandably so, no complaints there). If I could connect my ClaudeMax sub to it then I'd maybe use it, though CC and opencode are pretty good.
The code completion is decent and I especially like the "subtle" mode saves a lot of backspacing (no, dammit, that's not what I wanted!).
I understand Zed needs to make money, but VC backed, especially Sequoia, doesn't inspire any love, tbh. I don't care what platitudes VCs say about independence their love for the product etc., they need their 10x return and if it means enshittification, so be it, they don't really care.
Ugh, ok. I just want an editor.
Is DeltaDB not open-source yet?
Butch: Zed’s dead, baby, Zed’s dead.
How is this any better than VSCode?
Getting developers to switch is going to be hard.
Personally I don't think they can do it.
Latency, responsiveness...
Never had any issues with those.
Congratulations!!
I miss Atom.
Enshitenfication gonna follow.I like the idea that Opensource software should be funed like public infrastructure - not by loanshark VCs.
I LOVE THIS FOUNDER
I am a 10 out of 10
YES!!!
Let the enshitification begin.
Am I the only one that associates "Zed" and "development" with Zed Shaw? I don't think he has anything to do with zed.dev, unfortunately.
I really appreciate Zed Shaw for walking his own path and sticking to his guns.
I miss some of his old posts that he took down from his website, in particular the one on learning statistics, that was a great one.
You’re not. Everytime I see him on Twitch I have to do a double take.
Did not know he was a streamer now, thanks!
That brings them to 120x the amount of money I've invested in the same problem, and still they're only just announcing that they're intending to start trying to do something that I've already done better
Can we see?
It's apparently this but I can't really say that I get it: https://bablr.org
He seems to be saying he spent $350k making this. I guess it's some tooling for writing parsers.
He has this to say about Zed:
> Zed: Founded by Atom’s dev team, Zed was the rewrite that Atom always wanted to be able to do but couldn’t when Microsoft bought Github and made the executive decision to kill a product it might otherwise have had to compete with. Unfortunately Zed decided to do that rewrite in Rust. This has slowed their iteration speed, caused much of their dev effort to go to cross-platform support instead of innovation, cut them off from being able to offer their experience on the web, severely limited their hackability, and generally made theirs a niche tool for enthusiasts. What’s worse, their reliance on LSP — a product which believes that the presentation layer should be the primary abstraction layer — means their product is forever doomed to look like a VSCode knock-off. [1]
1. https://docs.bablr.org/architecture/prior-art/#ides
IMO in the 12 months or so I've been aware of the BABLR work and observed the author's comment contributions, they've never really substantiated why BABLR would be preferable to tree-sitter, or how a JS-based implementation of parser tech can fulfill any of the same niches. Most consumers of tree-sitter leverage it via FFI or native code, not an embedded or external JS runtime.
It's not clear to me how you could substantially replace the capabilities/benefits of what LSP provides with BABLR either.
We have some meaningful advantages that help us gain a practical edge on Tree-sitter:
- Our core and grammars are relatively tiny and can easily be loaded into any web page as a syntax highlighter, except actually a bit more: more like being able to embed ASTExplorer directly into your docs page or blog post to help people understand code examples.
- We support runtime extensibility of languages, e.g. TS can extend from JS at runtime. Tree-sitter only supports static linking, so every shipped language extended from their JS grammar contains a complete copy of the JS grammar.
- Our grammars are much easier to debug. They're written as plain scripts, and they can be run and debugged in exactly the form they're written in.
- We can parse inputs with embedding gaps. An example of such an input would be the content of a template tag before interpolations have been applied. Parsing after interpolation opens the door to injection attacks, but parsing before interpolation allow safe composition of code fragments using template tags.
- We emit streaming parse results on the fly, and can parse infinitely long data streams with ease or syntax highlight. within large single-line files without freezing up
- Tree-sitter is half an IDE's state solution: "just add text buffer". Our solution is the whole thing. One stop shop. IDE in a box.
- CSTML lets us do round trip serialization of any tree and also gives our trees stable hashes. Tree-sitter could trivially represent its parse results as CSTML if it cared to, giving it competitive compatibility with BABLR. A rising tide lifts all boats.
- While they're setting out to make version control for the first time now, we're already basically as powerful as git thanks to the combination of hashed trees, immutable data, and btree amortization within nodes for maximum structural reuse of data.
- Did I mention you don't have to deal with this? https://github.com/tree-sitter/tree-sitter-typescript/blob/m...
- We can probably literally just run the Javascript source code for Tree-sitter grammars on our runtime. The only problem is the C lexers, but C lexers are one of the great annoyances of tree-sitter anyway since any context in the grammar requires you to hand-write the lexer
I don't think they are doomed to be a VSCode knock-off due to their technical decisions. I think they decided to be a VSCode knock-off by design. I mean they follow the same project oriented GUI design that VSCode and Sublime helped make popular. There is nothing inherent to their tech that required that.
Haha yes I wrote that, thank you for sharing.
We're approaching the problem by drawing from browser design. We want to see an editor with a DOM API for code documents. BABLR is a parser framework meant as a direct answer to Tree-sitter.
Myself and most people I know that switch to Zed (many from Sublime) did it because we DON'T want a DOM in our text editor. Webtech makes for crappy text editors.
Sorry, I didn't make that very clear. I'm not talking about the HTML DOM, which exists as an API to control the 2D pixel rendering layer.
A standardized DOM for code would provide a universal way of changing code documents using scripts and would allow many different code-oriented tools to interoperate naturally where they never could before.
I think I know what you're saying, and it sounds like an interesting thing to attempt. Some sort of universal API that feels really good to make structural edits to code documents would be excellent.
won’t be using the Zed IDF anymore… whoops i mean IDE