This looks fantastic! Read through the homepage, clicked the "Download" button in the footer, page 404s. Joined the waitlist! We'll see.

Care to share a binary or some other project that is actually beyond a closed alpha? Otherwise not a lot of point in sharing it as an alternative, it's not available to the public.

Edit: signed up for the waiting list and received "Zed currently only targets macOS" so sounds like they only target ~10% of the actual market. Interesting strategy.

The strategy is that native macOS UIs are generally easier to develop than native Windows UIs, so they're trying to iterate on that platform first.

If they're building a native macOS UI, how is it ever possible it'll be worthwhile to port that into native Windows UI or "native" Linux UI? If you wanna go cross-platform, you usually consider that from the start, not once you're 50% done, as you're gonna have to redo a lot of work you could have gotten "for free".

The “engine” itself is likely cross-platform, with vendor-specific UIs.

It’s likely quicker to develop for a single OS and then build the additional UIs with a more feature-complete app.

Arc browser is also doing this, and they're planning on debuting the Windows UI in Fall.

Isn’t the whole selling point behind their thing that they don’t use the native macOS UIs but instead send everything to the GPU?

Rendering a platform native GUI and what you render it on are two different things. In the case of Zed, its a combination of - metal is quite easy to work with - apple has clear design guidelines and nice toolkits

The Xi editor team quickly found out how much of a mess "native" gui's were on windows. Zed will hopefully add support for vulkan and linux/windows later, but it makes sense to start off on macOS

Metal is not a platform native GUI, it is Apple's graphics API. The native UI on the platform is Cocoa (which calls Metal for some things, but isn't being used here).

I've got invites (and am not otherwise affiliated with the project) if you want to test it out. Username @ Gmail works.

If it's still Mac only, I don't have any real use for it, can only use editors that works across Windows, Linux and macOS, otherwise there is no real point in me trying them out :/ But I thank you for the offer, might reach out in the future if I hear it's available on more platforms.

Send one my way - [email protected]

If you have another invite to spare, I'd be very grateful to accept one! Email is: [email protected]

the big question: is it going to be open source, tho?

It's kind of maddening that both VSCode's remote dev extension AND Zed (which aims to be better than VSCode) AND JetBrains's remote dev AND repl.it AND <some other stuff> are all closed source.

What's the state of art open source editor/extension/protocol/whatever that enables two or more persons work on the same open files in real time? Something better than the ssh access which forces one person to work at time, then save, then the other person reloads.

ssh then attach to a tmux session

Okay this works, but how's the latency?

Can this made to work with a GUI editor? Like neovide or emacs but in X11/wayland not tty

Emacs with crdt.el, or neovim with instant.nvim are both open source and free implementations of collaborative editing with crdt's. Its going to be a bit of a struggle getting your coworkers to use emacs/neovim though

My dream is a 'universal' protocol for things like this, so that you can use whatever text editor you like to join a session.

Is there some inherent difficulty is this, or is it just "there is no standard" -> "nobody implements one"?

a combination of "there is no standard", "selling collaboration to businesses is way to make money", and that how editors handle files changes from editor to editor, so its a bit hard to sync

- zed.dev has crdt's/collaborative editing as one of its selling points, but they're planning to offer this collaboration as part a paid service to businesses. - vscode has its own sycronous collaboration plugin, but its proprietary and they declined to release the source code or offer to port it to other editors - you could probably build a plugin that bridges emacs and neovim, but it might be a bit slow as you'd have to "translate" the buffer from one editor to the other

There is an inherent difficulty in implementing it, getting asycronous collaborative editing right is hard. Combine that with how few people on emacs/neovim actually use collaborative editing (emacs still has a mailing list and does code via patches) and the chance for a "universal" version is practically nil.

Getting Teletype working properly again is one of our tangible goals for Pulsar - which is of course entirely open source. The question is of hosting so hopefully we can make it really easy to host your own instance or go peer to peer rather than just relying on a server.

Speaking with some of the devs, yes (asterisk). From what I understand, they are working out what they can open source while also retaining the ability to make money from the editor (mostly from the collaboration features)

I’ve been using Zed for a couple of months. It has replaced eMacs and VSCode for writing Rust. The integration is so well done, it’s like a constant conversation with the compiler. I’ve got ten invites by the way.

One thing I worry about a little is the business model. They want to target teams and charge for the cooperative features. But as a single user I’d love a way to just buy a lifetime license and feel that I can expect at least five years of use. I don’t like when software I love treats me like a casual fling.

Any chance I could get one of those invites? My username @ gmail.com

Thanks in advance! Looks cool