Perhaps the biggest reason is that the existing FOSS OS world has little to offer for people who like a Mac-style desktop. It's possible to customize DEs to have a very shallow visual resemblance but the details are all wrong, with Windows, legacy Linux desktops, and even mobile operating systems (in e.g. the case of GNOME) exerting the greater bulk of design influence.
The question might then, "why not make this a DE instead?" to which the answer is that Mac users also like how the underpinnings of the OS work (or have worked in older releases) and would like to continue to use their favorite Mac apps, which isn't going to happen as long as any traditional Linux distribution is involved. Mac apps also have a design ethos that's rare among current native Linux apps, which drives demand for alternatives.
To do this properly it really needs to be a whole-OS endeavor.
EndeavorOS. I think it's actually the userspace though, which this aims to emulate. The problem is that if you're using a macbook for final cut pro, then endeavorOS doesn't solve your problem.
Hmm, isn't EndeavorOS in the same vein as CachyOS? Mainly a pre-configured Arch variant with a custom point-and-click configurator built in? I don't see where it'd be of particular interest to would-be switchers using macOS.
I haven't downloaded or played with it because I'd rather not toy with software so obviously full of landmines, but from what I've seen it also falls into the "surface level" bucket. It goes a little further than one might get by e.g. enabling KDE's global menubar and installing one of the Mac-themed Breeze forks, but not much.
I’m super confused as to what is missing in the FOSS world, how does it “have little to offer?” It really sounds like you just like the macOS style and want a clone of it, and that’s certainly fine as a desire or a request, but if you like macOS, macOS is already available. For what logical reason should a project be copying macOS?
I would absolutely agree that a logical reason is to run macOS apps on other platforms, but other than that, the macOS operating system and its design is nothing incredibly exciting, unique, or unattainable elsewhere.
You’ve established that a Linux desktop environment in macOS style alone is not sufficient, and that makes me wonder what details of the underpinnings is Linux getting wrong compared to Mac? Is launchd superior to systemd? Is Metal superior to Vulkan? Is container superior to docker/podman in Linux? I think in a lot of cases the macOS underpinnings are quite inferior to those on offer by the FOSS world. Not in every case, of course, but a whole lot of Apple’s low level tools are half-baked and weird.
I think we need to articulate what is actually wrong with Linux or other FOSS alternatives as far as their underpinnings because I am really coming up short.
Heck, macOS and FOSS alternatives share a whole lot of identical low level tools underpinnings, like how you can run your favorite shell in either OS or even install Homebrew on Linux. (Another hot take: Linux package management is vastly superior to macOS or any other operating system on the market).
This idea of the design ethos being superior on Mac…was that true with Liquid Glass too? Are we going to exclude from our analysis every time Apple rushes out the next big Ui redesign to meet marketing deadlines and fixes it in subsequent updates like they’ve done multiple times in the past?
I look at befuddling aspects of the macOS design ethos like the fact that the menu bar will just silently truncate items when you run out of space or the fact that there’s a giant notch on the screen as evidence that maybe the macOS design ethos is nothing to envy.
Maybe macOS design is merely “pretty good” and not actually something that significantly far above and beyond its competitors.
Again, I applaud the idea of making software to run macOS apps outside of the propriety ecosystem just like the tools we use to run Windows apps in FOSS operating systems. However, I don’t think users looking for macOS design ethos or platform quality are starved of options.
> You’ve established that a Linux desktop environment in macOS style alone is not sufficient, and that makes me wonder what details of the underpinnings is Linux getting wrong compared to Mac? Is launchd superior to systemd?
macOS isn't perfect, particularly in its incarnations since Ventura, though gripes have been piling up since the "great flattening" with Yosemite, which is why Mac users might be interested in a similar alternative.
As a whole however, it works really well for some people. Better than Windows and Windows-like desktops (of which most DEs are closest to) and better than the desktoppified-mobile-OS shape of GNOME/Pantheon or tiling WM sort of thing.
The value lies more in the collection of frameworks referred to as Cocoa (primarily AppKit) and the desktop and third party apps built with it. The global menubar and perks it brings (such as a standardized menu API, which then enables things like System Settings allowing any menu item in any app to have its key shortcut remapped), the application-document process model, the windowing model that's built around windows overlapping with just the relevant bits remaining visible, first and third party GUI apps all being scriptable via AppleScript, high level of adherence to a central HIG, among many other things.
Many of these aren't unique but the particular combination is and doesn't really exist in the Linux desktop world… the best you can do is light mimicry and even that takes considerable manual configuration on the user's behalf, not to mention fighting against the grain and the awkwardness that brings.
That's what this is. Darwin isn't installable today. Apple just dumps a bunch of source code. Previous attempts at making Darwin bootable didn't go far and had no GUI.
Four years and still pre-alpha. Stick a fork in it.
There's also PureDarwin (?) which had been around forever but last I recall the lone developer was seriously ill and the project was not receiving help. I don't see how this would be any different.
To everyone here cheering this on: when will you be contributing? Or are you just waiting to download it, run it once for the lulz, and drag it to the Trash?
why do anything? why Linux? why FreeBSD? why Mac at all? Why build your own wind turbine when corporations make large ones? Why build a cyclekart when you own a car? Why workout when everyone's going to die?
With that mindset, everything is useless and nothing has a point. Have I made my point?
I regret to actually say this, and mean no controversy - but, I have been actually asking myself just that - and, not in the "workout or not" sense, alone. In the "does anything have a point", way. Apologies.-
PS. To indeed contribute something: I am a big fan of the "doing things because of the joy of doing things" itself theory of it all, which is, seemingly, an antidote.-
In general, things have meaning because people assign meanings to them. Nothing inherently has meaning as meaning exists only within the human mind/soul. If you are searching for meaning, you will find nothing.
"This is a huge effort. Y'all must be crazy!
Yes. Here's to the crazy ones... because the people who are crazy enough to think they can change the world are the ones that do.
In all seriousness, though, it's not my first rodeo. I have no illusions about the scale of building an OS or how long projects like Haiku, ReactOS or GNUStep took to get usable. This is just something I like doing in my Copious Spare Time™ and that I hope will be useful someday.
I've always liked Apple's design and elegance, but am not thrilled with the direction of recent macOS® or how closed the hardware has become. And I'm a big believer in open source. So ... let's build something that feels as elegant, without all the lock-in, and maybe (just maybe) I can get it to a point that reasonably useful Mac apps build & run on it. And if not, we'll build a lot of cool pieces along the way and learn a ton. I'm OK with that."
Please don't fulminate on HN. The guidelines make it clear we're trying for something better here. https://news.ycombinator.com/newsguidelines.html
Perhaps the biggest reason is that the existing FOSS OS world has little to offer for people who like a Mac-style desktop. It's possible to customize DEs to have a very shallow visual resemblance but the details are all wrong, with Windows, legacy Linux desktops, and even mobile operating systems (in e.g. the case of GNOME) exerting the greater bulk of design influence.
The question might then, "why not make this a DE instead?" to which the answer is that Mac users also like how the underpinnings of the OS work (or have worked in older releases) and would like to continue to use their favorite Mac apps, which isn't going to happen as long as any traditional Linux distribution is involved. Mac apps also have a design ethos that's rare among current native Linux apps, which drives demand for alternatives.
To do this properly it really needs to be a whole-OS endeavor.
EndeavorOS. I think it's actually the userspace though, which this aims to emulate. The problem is that if you're using a macbook for final cut pro, then endeavorOS doesn't solve your problem.
Hmm, isn't EndeavorOS in the same vein as CachyOS? Mainly a pre-configured Arch variant with a custom point-and-click configurator built in? I don't see where it'd be of particular interest to would-be switchers using macOS.
2 replies →
Didn't the Best Koreans rip off the Mac GUI in their home-grown spyware-laden Linux distro? Red Star Linux or whatever it's called.
I wonder if they have a GitHub...
I haven't downloaded or played with it because I'd rather not toy with software so obviously full of landmines, but from what I've seen it also falls into the "surface level" bucket. It goes a little further than one might get by e.g. enabling KDE's global menubar and installing one of the Mac-themed Breeze forks, but not much.
I’m super confused as to what is missing in the FOSS world, how does it “have little to offer?” It really sounds like you just like the macOS style and want a clone of it, and that’s certainly fine as a desire or a request, but if you like macOS, macOS is already available. For what logical reason should a project be copying macOS?
I would absolutely agree that a logical reason is to run macOS apps on other platforms, but other than that, the macOS operating system and its design is nothing incredibly exciting, unique, or unattainable elsewhere.
You’ve established that a Linux desktop environment in macOS style alone is not sufficient, and that makes me wonder what details of the underpinnings is Linux getting wrong compared to Mac? Is launchd superior to systemd? Is Metal superior to Vulkan? Is container superior to docker/podman in Linux? I think in a lot of cases the macOS underpinnings are quite inferior to those on offer by the FOSS world. Not in every case, of course, but a whole lot of Apple’s low level tools are half-baked and weird.
I think we need to articulate what is actually wrong with Linux or other FOSS alternatives as far as their underpinnings because I am really coming up short.
Heck, macOS and FOSS alternatives share a whole lot of identical low level tools underpinnings, like how you can run your favorite shell in either OS or even install Homebrew on Linux. (Another hot take: Linux package management is vastly superior to macOS or any other operating system on the market).
This idea of the design ethos being superior on Mac…was that true with Liquid Glass too? Are we going to exclude from our analysis every time Apple rushes out the next big Ui redesign to meet marketing deadlines and fixes it in subsequent updates like they’ve done multiple times in the past?
I look at befuddling aspects of the macOS design ethos like the fact that the menu bar will just silently truncate items when you run out of space or the fact that there’s a giant notch on the screen as evidence that maybe the macOS design ethos is nothing to envy.
Maybe macOS design is merely “pretty good” and not actually something that significantly far above and beyond its competitors.
Again, I applaud the idea of making software to run macOS apps outside of the propriety ecosystem just like the tools we use to run Windows apps in FOSS operating systems. However, I don’t think users looking for macOS design ethos or platform quality are starved of options.
> You’ve established that a Linux desktop environment in macOS style alone is not sufficient, and that makes me wonder what details of the underpinnings is Linux getting wrong compared to Mac? Is launchd superior to systemd?
Pretty much anything is better than systemd.
CoreAudio is far superior to anything Linux has.
WindowServer is better than Wayland.
macOS isn't perfect, particularly in its incarnations since Ventura, though gripes have been piling up since the "great flattening" with Yosemite, which is why Mac users might be interested in a similar alternative.
As a whole however, it works really well for some people. Better than Windows and Windows-like desktops (of which most DEs are closest to) and better than the desktoppified-mobile-OS shape of GNOME/Pantheon or tiling WM sort of thing.
The value lies more in the collection of frameworks referred to as Cocoa (primarily AppKit) and the desktop and third party apps built with it. The global menubar and perks it brings (such as a standardized menu API, which then enables things like System Settings allowing any menu item in any app to have its key shortcut remapped), the application-document process model, the windowing model that's built around windows overlapping with just the relevant bits remaining visible, first and third party GUI apps all being scriptable via AppleScript, high level of adherence to a central HIG, among many other things.
Many of these aren't unique but the particular combination is and doesn't really exist in the Linux desktop world… the best you can do is light mimicry and even that takes considerable manual configuration on the user's behalf, not to mention fighting against the grain and the awkwardness that brings.
1 reply →
> when you could run Darwin
That's what this is. Darwin isn't installable today. Apple just dumps a bunch of source code. Previous attempts at making Darwin bootable didn't go far and had no GUI.
Four years and still pre-alpha. Stick a fork in it.
There's also PureDarwin (?) which had been around forever but last I recall the lone developer was seriously ill and the project was not receiving help. I don't see how this would be any different.
To everyone here cheering this on: when will you be contributing? Or are you just waiting to download it, run it once for the lulz, and drag it to the Trash?
> Four years and still pre-alpha. Stick a fork in it.
Who jammed a toe in your cheerios?
7 replies →
I dunno you could definitely do a "wine but for macos" kind of implementation, e.g., https://www.darlinghq.org/ without breaking IP laws
I've always been shocked by the lengths of time people will commit to proving wrong the folks who tell them "you can't do that."
why do anything? why Linux? why FreeBSD? why Mac at all? Why build your own wind turbine when corporations make large ones? Why build a cyclekart when you own a car? Why workout when everyone's going to die?
With that mindset, everything is useless and nothing has a point. Have I made my point?
> Why workout when everyone's going to die?
I regret to actually say this, and mean no controversy - but, I have been actually asking myself just that - and, not in the "workout or not" sense, alone. In the "does anything have a point", way. Apologies.-
PS. To indeed contribute something: I am a big fan of the "doing things because of the joy of doing things" itself theory of it all, which is, seemingly, an antidote.-
In general, things have meaning because people assign meanings to them. Nothing inherently has meaning as meaning exists only within the human mind/soul. If you are searching for meaning, you will find nothing.
... or flowers for the grave? :)
PS: The FAQ speaks to this somewhat: