This is very interesting to me for two reasons:

1. I'd really like to run my development environment for things under Siri for its tiling window management but for better or worse I'm deeply embedded in the Apple ecosystem for everything else, this looks like it could be a really nice way of doing it (possibly once multi monitor support is in).

2. There are still a few applications which have supported Linux builds but no support for macOS (Iridium's Niagara Workbench application for configuration of building management systems springs to mind here). Since Apple ended support for Quartz this has been a bit of a pain to deal with.

I think there are many use cases for this software.

For example, you may not want to run some graphical applications directly on your Mac for security, isolation or testing purposes.

If this software turns out to be lower latency than RDP and CRD, I could also see it being very useful for accessing a remote graphical workstation (e.g.: running heavy software on an beefy machine in a data center instead of taking up resources on my skinny laptop).

It's not necessarily something only available for Linux, but something that you want to containerise. (And then it's inherently running on Linux.)

Emacs runs much faster and better on Linux VMs. And I have a VM for each client I work with.

Try building Inkscape or GIMP from source on macOS and see how "multi-platform" those GTK apps feel in practice. Even when a Mac build exists, it is often skinned oddly or lags because somebody has to carry Mac patches against an old fork.

This is for the long tail. The compositor path dodges a pile of volunteer-port churn and runs the Linux build directly, which is a lot more appealing for niche GUI tools and dev apps that barely get maintained on Linux, never mind macOS.

In reply to then throwaway.

Anything I want sandboxed or “grouped”.

Work on a project -> open the relevant container.

Similar to parallels window integration mode.

It’s all from shortcomings to have a hierarchical view on your data and applications.

Goal: isolation. Security-wise, and focus-wise

Many of the keyboard commands are configurable in settings, complete with switching cmd and ctrl keys around. Or you can get used in a week or two when switching, this is what I did years ago and now for me Win/Linux is confusing and find the location of the command key more ergonomic on a Mac.

Here some history on how the command key came to be https://www.folklore.org/Swedish_Campground.html https://en.wikipedia.org/wiki/Command_key

As someone that switches between MacOS (dayjob) and Linux (my own PCs) workstations daily - I wish I could do the opposite for Linux. MacOS keyboard shortcuts are just way more intuitive to me, and they are way more consistent across applications.

Apart from just running Linux apps, you can use this to run graphical applications remotely on a Linux server, like X11 forwarding.

I want to use KDE Plasma instead of Mac OS ugly (in my opinion) interface

Popular apps? Probably not many.

But in the field of integrated circuit design there’s lots of apps that are Linux-only. I’ve tried to run some of them in containers on Mac. But XQuartz is awful.

If they ever transitioned to Wayland perhaps this would let us run these apps on Mac in a nice way.

On the other hand some of them have started getting ARM builds (for running simulations on certain cloud environments) so maybe native Mac GUI builds could happen someday soon.

PuTTY

But why MacOS then? If you take away the interface what differentiates Darwin from FreeBSD or GNU?

Incredibly L take. macOS keyboard commands are great for working in the terminal since system shortcuts use a different key and don't interfere with control codes

Sorry having to use ctrl+shift for in a terminal is absolutely awful. macOS keyboard shortcuts are king.

I mean, you can simply use Linux and save yourself all those hacks…

Super key for most keybinds is much nicer than windows in my opinion, where it is entirely wasted on opening the start menu. On Linux it gains a few functions based on the desktop environment but not much.

Then you want to run KDE on Linux. This is not going to replace your native mac desktop environment.

What about this plus XWayland? Would that do it?

That it will actually run on Apple Silicon.

TBH, I would love to install GNU or BSD on my M4 Max Mac Studio. What I really wanted is a modern UNIX workstation. My Studio’s price/performance was the best available, so that’s what I bought. Now, I am happy with that purchase except for the constant diminution in software quality from Apple.

Performance on apple chipsets!

Exactly. You don’t have the terminal itself fight with whatever is running on the other side of the term.

Being limited to just control and alt definitely cuts down on the options. Conversely, having MacOS command key act as “super” in Emacs opens up some possibilities.

agree. I guess it's a force of habit, but I am so used to the cmd+<whatever> (specially copy & paste) shortcuts, that I configured them into my linux desktop to behave the same way

I thought Wayland was different from X11, and didn't allow this. But I'm far from an expert on this topic so I'd like to learn more.

This is possibly the first time in human history this opinion has been stated.

Honest question: why use MacOS at all then? If you prefer KDE, why not run a system that KDE natively supports? Is it a particular MacOS application? Or is it that Linux support on Mac hardware is not good enough?

That's a Windows app.

The use of the Windows key extends far beyond the start menu. Builtin functions include window management, invoking programs on the taskbar, locking the computer, invoking Explorer and Settings, invoking and controlling accessibility functions like Magnifier. The Microsoft Power Toys add a lot of functions using the Windows key by default as well, like screen snipping, screen OCR, color picking, enhanced clipboard, and many more.

Absolutely. I went through great lengths to install Asahi on my work M1, only to have most things not work (RTFM). So when one is forced to use MacOS, may it round corners in hell, for work…

here is Wayland vs X11 visualised.. it might help - https://vectree.io/c/compositing-window-management-architect...

I don't want KDE but I would much prefer gnome to macos desktop, I think it's both prettier and more functional, and that's not a tahoe thing. I hate finder too, and don't see any way to properly use an alternative.

It really isn’t. There are a great many people who use macs for work but who do not like Apples design choices. And that number has skyrocketed even further since Liquid Glass was pushed onto people.

In fact one of the front page articles today is literally calling macOS “ugly” in the title.

Similar opinions are voiced in about every recent macOS UI thread, and even occasionally in Windows threads.

not sure about the looks but I also find KDE much better in usability

To be honest, I agree a little bit because I remember from my time at customizing KDE that everyone wanted it to make it look like Mac OS

but it feels a bit of peer-pressure/cool-factor, people used to like how Mac OS look but after Tahoe, I feel like most people don't.

To be honest, I am on mac right now but I really like Niri/Hyprland and to a degree KDE as well. I definitely feel like those were immensely more customizable and I miss that customizability, even if some people might use that customizability to make it look like MacOS default.

Linux has good support for m1 and m2, so for newer devices running a custom desktop would be neat.

PuTTY is absolutely available for Linux. On Debian-based distros it’s just a `sudo apt install putty` away. But why?

My problem is that I don’t use the majority of these functions at all. Command I can use for almost everything no matter how frequent or infrequent. It also replaces most “ctrl+shift” binds which is a great plus for me.

Aside: the new, large radius Liquid Ass corners that make some parts of the window basically unusable are really annoying me.

https://github.com/neonkore/waypipe proxies Wayland over a network. It’s straightforward enough in theory: Wayland core is just a communications protocol plus shared memory; so you just need to forward the messages, and detect and send changes in the shared memory. Not the cheapest thing, but perfectly tractable. Of course, there are also more difficult extensions, like GPU integration, but that sort of thing was a problem for X as well.

I mean Apple hasn't done itself any favours with Macos Tahoe.

That’s still like 10 uses vs unlimited uses on macOS

Maybe try this: https://ke-complex-modifications.pqrs.org/#pc_shortcuts

This is how modern x11 worked too since nobody uses software rendering with x primitives anyway.

“Liquid Ass” as some people say.

Even just the window management category is more than ten uses. And it is unlimited uses, as you can assign additional shortcuts however you like.

You must not use MacOS. Command gets used all over the place, even during editing. And in Emacs it gets used as Super, which opens up some options.

Yeah I've used Karabiner to get windows-style shortcuts (home/end, etc.) and it works very well.

It may be ten uses, but it's ten uses I use constantly throughout the day.

This is definitely not worth using. It doesn't even say what hypervisor its using. Is it using QEMU? Docker? Podman? Lima? Colima?

And also this chart is super weird:

    Solution  Latency  HiDPI    Native Integration  Setup Complexity
    Cocoa-Way Low      Yes      Native windows      Easy
    XQuartz   High     Partial  X11 quirks          Medium
    VNC       High     No       Full screen         Medium
    VM GUI    High     Partial  Separate window     Complex

A standard VM will always be the easiest to set up by far. And latency should be the same across all 4. I mean after all it's a VM running on your local machine. Honestly I don't even know what it means when it says "Latency".

I also looked at some of the code and it's using OpenGL 3.3 Core which is... super old. But it makes sense in the context of this being LLM-generated since most of its training data is probably OpenGL 3.3 Core code....

Overall this project is very strange. It makes me feel more confident in my skills, AI isn't all that great. It's all hype. You can get to the frontpage of HN. And if you're Peter Steinberger you can get acquired by OpenAI for a billion dollars. But that's about it. The code isn't getting any better.

This reminds me of that C-compiler-in-Rust publicity stunt by Anthropic. There's no substance. It's just a headline.

While I agree with the rest of your comment, they do mention they use OrbStack as their hypervisor in their demo video.

Gotcha thanks for that info. Yeah that's insane. You have to read the description of a YouTube video to understand what a project on Github is doing. There is no architecture here.