> Fedora Linux 45 was released on 3 November, and we shipped Fedora Asahi Remix on the same day. This is the first time we have been able to get a new version of Fedora Asahi Remix out on the same day as upstream!
This is a wonderful thing to read. I've been a proud financial backer of Asahi since the first announcement of their Patreon account (now OpenCollective). I've only installed and booted it once on an M1 MacBook Air to kick the tires. I liked what I saw. It worked well enough to validate the work being done. I then reinstalled MacOS so that my partner had dedicated modern laptop. I also doubled my donation.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
Oh man, I just doubled my donation again. I'm now at 4x my original donation based on reading this because hell yeah. Congrats to the team doing this stellar work. Now that my partner has a MacBook Neo, I'm gonna wipe that MacBook Air and test out the latest release. I'm so happy folks are doing this.
I support Asahi due to ideological conviction and self-interest (in case I ever want to have an alternative on Apple hardware) rather than because I want to replace MacOS. But who knows. If Apple's rent-seeking behavior with the App Store grows and further disrupts my opinion of them, maybe I'll end up on Asahi full time.
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
Unfortunately, the license allows this. It’s also a social issue - why shouldn’t a faster mover be able to slop a feature faster than the project owners? Who cares if it’s disruptive to them?
I’ve seen this happen on other projects and beyond rejecting the changes, I’m not sure what to do. It’s damaging morale when someone has worked weeks or months on a feature, worked with countless individuals on planning and implementation, and then someone releases a vibe coded slop fork with the last 5-10% and then takes credit as if they’re some genius doing something useful or helpful.
Ultimately the issue is manners, but when there is opportunity and generosity, the generous are going to be taken advantage of.
LLM haters are the worst type of toxic; they will not stop until everyone else stops using it, too.
Punitively changing a license with an intent to further restrict what someone does with the code is childish. If the concern is "someone will feed it into an LLM to obfuscate it" — what stops them from doing that with GPL code?
GPL code is, by the way, ripped off on a daily basis and routinely ignored because it's bordering on impossible to enforce.
In some cases LLMs are a legal issue. In other cases they point to low effort contributions. Those changes require the same amount of effort to review, but are no effort to produce. When there is no mechanism to control a resource sink, everyone suffers.
GPL will possibly scare off serious, risk averse companies with lawyers. It's not going to scare of Chad the AI Bro who wants to move fast and break things, crush it in the market, ride the VC rocket, and has no qualms about monetizing the work of others.
Who are these large commercial players worried about Asahi Fucking Linux of all things?
Oracle? HP?
Is EMC building a SAN out of recycled MacBooks?
Not to discredit any of the work they've done but this is at its core a hobby project, and no large commercial player is touching this with a 10m pole.
People who oppose the use of LLMs have the right to do what they want with their software just like anyone else does. It’s not “childish” or “toxic” for them to express their opinion by changing the license on their own software or writing about it on their own blog.
Furthermore, if the GPL is really as unenforceable as you say, then there’s no reason why their license change is “punitive”. Either the GPL works, in which case the Asahi Linux maintainers have the right to restrict use of their software, or it doesn’t and there’s no reason why anyone should be upset by a gesture of free speech.
The Asahi Linux maintainers have the opinion that LLMs are harming their ecosystem, and if they want to prevent people from using LLMs on their software, then they should be able to do so without being unfairly attacked.
It's not that I am against LLM because I'm using them, but I get sick of reading that mentality of AI-must-learn-from-everything-and-if-you're-against-you're-against-progress.
It's an incredible yet immature technology, and (too much IMO) people want to use it for everything without asking themselves any question. There was and still is a whole world of technology and knowledge before AI/LLM that left us teachings, laws, technologies that we must live with. Some of these things are here for good and maybe others for bad reasons.
In any case you should take your time to understand why something is there before blindly disagreeing with it. That's what moves discussion forward.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
I appreciate all of the hard work that's gone into Asahi linux, but this statement is basically nonsensical. Making the license copyleft does not prevent people from monetizing it, using AI to develop on it, improve it or wreck it. All it does is make sure that anyone who does will share their source code modifications.
It doesn't sound like they want their code to be open source. It sounds like they want it to be closed source or at least restrictively licensed enough that they can tell users of the source how they're allowed to interact with it and modify it.
The entire point of open source / free software is that it lets you do what you want with it. The original author doesn't need to accept your AI patches or write AI code themselves, but downstream users are not restricted.
I am just baffled by how confused even very smart people seem to be about how open source and permissive licenses work and what they are for.
(Now, should they use trademark law to ensure others can't use the name Asahi Linux confusingly? Absolutely. But that's got nothing to do with copyright.)
You are confusing why they choose MIT (to further adoption by other existing distributions and allow other developers to reuse pieces) and how it is being used (to build cheap distributions on it slapping a different name on front), they understand well the implications as anyone else.
> I appreciate all of the hard work that's gone into Asahi linux
Actually no, you don't seem to. Regardless of the license, remind yourself that they didn't lose ownership of what they did. They can still be disgusted at how people are building lazy derivatives just because they wanted to share what they did with a different kind of developers. I know I would.
Frankly, I'm not sure if permissive licenses have a future. No one wants dev marketeer to save tokens reusing their work.
They’re switching to GPL and you think that’s anti-open source? MIT has basically zero restrictions and assumes a good-faith collaboration environment. GPL is more realistic about the motives of much of the tech world, and in practice is much more encouraging of collaborative, open, and free software development. Kudos to them on making the switch.
The only blocking point for me before I would use Asahi Linux is the support of the USB-C displays. I was considering and reconsidering to install Asahi so many times, but I just cannot give up the ability to plug an external display into my MacBook Air.
I am just surprised that other features like support of the newer CPUs still get a higher priority than this one. I am not complaining or anything, I am not even related to the project in any way, so not in the right place to ask for anything, but I really am surprised that this seems to not to bother most of the users?
I am always looking at this "Device Support" page [1] and it's still red there. I even have a calendar reminder to check it out every once in a while, but if it's not being updated I might drop my calendar reminder then :)
Sometimes this can be because the people working on one aspect of the project are not the same as the people working on another, or the same number of people, and also maybe supporting later chips is less problematic than supporting external displays. The project as a whole doesn’t just work on one problem at a time in priority order.
I believe the blocker here is that USBC display support requires getting full Thunderbolt support working. Apple built their own Thunderbolt chips, which diverge from the Intel hardware that everyone else has used & that the Linux kernel drivers are written against enough to make support much harder than you'd expect.
IIRC Asahi has finally acquired Thunderbolt support but that support still doesn't have the pieces needed for external displays.
The entire section "Kept you waiting, huh?" is about this and some of the background is really interesting explaining why things are as they are. Here's just the last bit:
> Luckily for us, Sven was up to the task of dealing with all this nonsense. It just took slightly longer than expected! While the patchset to enable Thunderbolt was sent upstream first (another massive win for us!), we have also pulled it into our downstream kernel and are now shipping it to users!
> But there is a catch. Currently, only USB3, XDomain and Thunderbolt Networking are supported. You can plug in USB devices or connect two Thunderbolt computers together and make them ping each other, but that’s about it. PCIe and DisplayPort tunnelling — required for most Thunderbolt docks — are not supported. Yet. Watch this space.
> An enormous amount of gratitude is owed to Sven, who has stuck with this thankless chore for nearly 5 years in between all of his other incredible work.
As an old-time Linux user I wish Asahi all the best, and I get why they want to have the apple silicon working. But it just leaves a bad taste in my mouth to support a company that has historically been somewhat hostile towards Linux/FOSS compat and the power users while in the same breath trying to proclaim their platform as developer friendly, nevermind the "apple tax" that has to be paid on commodity hardware. The hacker spirit isn't aligned on this one.
Because sometimes people don’t go out of their way to buy Linux but want to repurpose old stuff they already own to experiment a bit and/or extend its life? For instance, I recently installed Ubuntu on my 10yo Intel MBP from 2015, and had to tweak lots of things to get it to a somewhat usable state that doesn’t flicker/tear/freeze. Had to use X11 instead of Wayland for anything Chromium-based. One of the most amusing ones (for some definition of amusing) is Bluetooth audio not working, turns out Ubuntu uses hot garbage codec that can’t fit into the available bandwidth due to AAC licensing (?), possibly among other problems, and switching to an AAC-enabled build made it reasonably stable.
I agree that you shouldn't rule out devices that don't have official support for Linux. And it's amazing that the community brings it to some many devices.
But if you already know at purchase that you'll install Linux on the device, you might as well prefer those with official support rather than just a community port, simply to show support for companies that agree with your priorities.
Also you should inform yourself beforehand. Don't just buy random hardware and then make the surprised pikachu face when it happens not to support Linux.
Uhh, this ethos of “can I make Linux work on this hardware” has been a huge driving force of Linux’s success. This sort of effort is deeply embedded in the culture and always has been.
Apple makes very high-quality hardware, and unlocking the full capability of that hardware via strong Linux support is absolutely a noble cause. Rather than giving in to corporate lockdown, the Asahi folks are gifting the world a way to take advantage of this new lineage of Apple Silicon based personal computers.
In 10 years when Apple has long abandoned support for the M1, M2, etc, that hardware will still be able to run strong thanks to the efforts of this team. I salute them.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon.
Is this referring to Omarchy? Anyone have context? TBH I'm surprised their work was ever MIT-licensed instead of GPL.
I saw that's there is an Omarchy for Mac "team" consisting mostly of one guy who is active on Twitter (and quickly amassed many followers) and seems to have no prior experience of working on Linux kernel. They do not mention Asahi anywhere which isn't surprising to me given the reputation of DHH and Omarchy. This hijacking of open source by bad faith actors is very concerning.
> Omarchy on...
> Apple M1/M2 chips#
> Asahi Alarm is a version of Arch for Apple M1/M2 computers built on top of Asahi Linux. You can get Omarchy running on top of that with some effort. See the user-driven guide.
Omarchy has gotten somewhat of a reputation for getting a lot of sponsorship money from commercial actors. If by monetize you mean "Get people to pay us" omarchy are definitely monetizing. Not much different to Asahi themselves of course, but is it right/moral to fork another active projects code and use it in order to gain your own patrons? I don't hold the answers. But no they aren't selling anything.
edit: worth noting this example, where DigitalOcean started cutting OSS support only to pledge millions of dollars to Omarchy. No wonder people consider the project a leech to the OSS community, but also consider correlation != causation: https://itsfoss.com/news/digitalocean-open-source-credits-en...
It is worth noting the same article's discussion of nodejs reads as "DO shut down a program, possibly planning a v2 of said program, but failed to communicate it to its dependents and didn't realize until they started planning to leave". Or they were just fighting the fires caused by a github issue named "Transition off of Digital Ocean" caused \o/
I've seen businesses that made the infrastructure to share a 100k machine. They needed it "on premise", but the setup did not look too hard and the results were outstanding compared to Claude Opus.
There's people running close to frontier model grade local AI on consumer and prosumer hardware, let's not act as if it was impossible to skip having to rent from OpenAI or Anthropic.
Omarchy is itself MIT licensed, with no restrictions that I'm aware of. If they later decide to make a paid version or something, the license wouldn't stop them, I think. But GPL would. All in all, I think it's a dumb move from Asahi because their GPL'd code will soon be eaten up LLMs that don't give a damn about licensing anyway.
I really don't get why people even start with permissive licenses if they have people they don't like to use their software or if they're just going to go "not like that" at the first real use.
It seems like many times it's open source but "just to look".
Because permissive licensing, and free software in general, rightfully comes with a certain cachet, and it turns out some people want the benefits of that reputation, but don’t actually believe in the freedom “to use, copy, modify, and distribute this software for any purpose.” Hence the treadmill from permissive licensing to copyleft, and then later from copyleft to “source available”…
“But software which OpenBSD uses and redistributes must be free to all (be they people or companies), for any purpose they wish to use it, including modification, use, peeing on, or even integration into baby mulching machines or atomic bombs to be dropped on Australia.”
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin.
Permissive licences require attribution. It's okay to complain when people publish your permissively-licenced work without attribution, and hence without honouring the licence.
changing the licence is much cheaper than pursuing legal action, and sends an immediate and very public signal of disapproval. given that legal remedies would be required either way, and are extremely time-consuming and expensive, it makes sense to send the signal first
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
It's one reason I've favored GPL of late, after many years using MIT. Not that many people these days would explicitly follow the terms of any license.
There are licenses that require attribution without being copyleft though. I'm not saying the Asahi developers are wrong to use copyleft, but it's weird to me to mention "without attribution" as a sticking point when there are fairly standard permissive licenses that require attribution like Apache and BSD.
MIT isn't copyleft and it does require attribution, it's basically the only requirement of it. I guess they just don't like that attribution means "put the license snippet deep in a secondary readme file hidden somewhere, along with 50 other licensees used in the project".
I think it's clear that they don't want people to use their code and close the door behind them. They also clearly want clearer attribution. I'm sorry but if those are things you desire, then deliberately choosing MIT because of a GPL hate-boner was plain stupid.
They are mad that vibecoders are building on top of their work, making real progress, then re-releasing it as "look it works" while containing a bunch of their work. Which is sort of fair but I don't get how that works in practice because in reality anyone can read the code which is open source and do whatever they want. Licensing won't stop them.
Your credit is your first commit. After that people copy/paste patches back and forth as they will. It happened before AI. I've run patches in prod that were basically someone's random RAID controller patch posted to the Gentoo forum. If it works, it works.
Right, this seems deeply confused. If you care about how your code is used, don’t use MIT. Especially, if you care about monetization, probably use GPL or AGPL.
The whole reason why people are trying to replace them is because they have stubbornly refused to use AI, which has directly impeded progress in getting M3/M4/M5/M6 working. It's fine of course for them to continue pursuing their hobby of doing everything manually, and honestly it is insanely impressive.
But most people just want stuff to work, so it's odd to complain about it when others do manage to get it to work using AI to build on top of their prior permissively-licensed work.
Wanting stuff to work is a valid reason to not use AI for more sophisticated development projects. Having a computer that kernel panics once every three days is useless to me, even if it works for you.
Do the massive amount of vulnerabilities not make you think twice?
At this point, I trust human code less than llm code. Design and architecture of the code still greatly benefit from a good human touch, but I'm not sure how long that will remain true.
So validate with an llm and humans. People committing code they haven't read - which will inevitability happen - is dangerous and hard to remotely verify.
In the end there is always a separate reviewer. Even if the producer is adversial and not even looking at the code then this process exists to catch it. This is not at all different from people attempting to get backdoors into Linux.
People using LLMs put up much larger changes, and much faster. Depending on the LLM used the quality is often much lower.
A lot of projects just stopped accepting any outside contributions rather than deal with the tidal wave of slop they were getting. Choosing to not accept LLM code is IMHO a better compromise.
LLMs will introduce new vulnerabilities. What Linux should have done is stop using non-standard C, or better, stop using C completely in favor of Rust, support code analysers.
And humans can introduce new vulnerabilities, too. Linux maintainers accept patches from random people over the internet. They already have to just the patch on its technical merit, not on who submitted it. For them I don't think it makes a difference (technically) if the patch was written with the help of an AI.
I believe that there may be valid reasons to refuse anything that was touched by an AI, but that's not about security. It's more about the non-technical problems coming from AI (centralising power, increasing inequalities, carbon footprint, ...).
Some people will always disagree with something. That's what's great about free software. There will probably be some Luddite anti AI Linux fork at some point.
That's not what they meant. It's AI friendly, not slop-friendly.
There's a reason no Omarchy patch of any kind has made it to the kernel list yet. I did personally look into some of their code and it is certified, hacked-togeter slop.
I also agree with that definition. But for example if I get an LLM to generate 95% of a program, but I understand and have read every line of code, is it still slop?
As an open source maintainer, when someone opens a PR, I don't know who they are. But I don't really care. I care about the PR: is it a "good" PR (there are multiple dimensions: is it understandable, does it add maintainable code, was it tested, etc) or not?
I think "AI-friendly" says "I don't care if it's written with an AI or not: I want good quality PRs".
I would think that "slop-friendly" is the kind of projects where the maintainers don't read the code. Maybe they try the feature (maybe not even?). I think the Omarchy guy was explaining somewhere that he does not read the code anymore, he just lets his AI write Rust. This is slop-friendly IMO.
If you are not AI-friendly, it means that you refuse to accept a good PR even if you cannot tell that an AI was involved. The mere fact that an AI was involved disqualifies the PR.
Are you one-shotting then reading/understanding? Then probably slop.
Are you iterating to help you understand, prompting to rename/restructure/document until you would have been able to easily understand without AI? Then probably not slop.
I'm pretty sure Asahi devs experiment with LLM tools, but not to Omarchy's extent. LLM ban is purely a filtration filter, so that they wouldn't be spending time overbearingly exhausting number of patches generated with AI, and not someone who cares about this than a shiny tool.
Everyone understands that LLM is a mean to an end, but it's clear that Omarchy got two confused and it's showing.
* Author posts patch with a manageable number of lines so maintainer doesn't gets burned out reviewing it.
* Tries to outline on the body what the patch is about and his rationale for the design decisions on the code.
* Promptly answers any maintaners questions about the line changes.
* Discloses which AI model he used to assist on the patch and how was it used.
AI slop patch:
* Submits patch with huge lines of code without explanation.
* The body is written in a self-patronizing tone with superfluous formatting that helps nothing but annoys who is looking for the actual substance of patch
* When questioned by maintainer about some line changes either goes radio silent or reply with the usual AI tone of marketin salesman.
Exactly. As an example, read that Gravity Linux guy's responses here. I will never ever trust him. I will stay away from his code as far as possible. Asahi team: I have my trust in you.
A kernel panic is the least of your worries with this hardware. Asahi had to introduce a custom daemon to avoid frying speakers. Apple doesn't need to care about making the hardware interfaces safe if they control the entire stack, so they don't, and expose dangerous capabilities to the OS.
There was a fun period of time where Boot Camp still existed, but Windows could fry the speakers because it didn't have any of macOS's thermal protection. I think this was after the 2016 redesigns, as I had a mid-2015 with no such issue.
In the case of speakers, though, what would a hardware failsafe even look like? It would probably be some sort of secondary chip; Apple's design is to just move that to the main processor, the same way they don't have separate SSD controllers.
Monitoring the CPU temperature feels very different to me. I'm pretty sure the CPU can't cook itself. This is probably documented somewhere so we don't need to speculate, but I can't research right now...
Good thing they only ever write correct code which consistently obeys every invariant in that entire stack ever. Excellent design, meet perfect engineering.
It's amusing that people still default to "human-written code is solid and AI code sucks" after the enormous quantity of bugs that AIs are finding in human-written code.
Nobody has ever said human code is perfect. But it is empirically the case that code written by LLMs is on average lower quality than code written by humans.
That's not really true, and it's only even potentially true if we talk about the absolute cream of the crop developers. Most human written code is quite bad actually
What is vibe-copying? The code was permissively licensed specifically so other people could do what they want with it. I'm so confused what people think permissive licensing is
There is no "stubbornly", the project is theirs and they do the f they want, their reasoning is sound. Permissive license does not mean that you cannot speak out about what people are doing with your software.
Do you think there would have been something working on mac if omarchy didn't start from the solid foundation of Asahi? Please don't make me laugh too hard.
And by the looks of it, they're very intent on gatekeeping. Deep down I don't like AI either, it has taken away a lot of what I love about coding. But it's here to stay. What if we can use it to wrestle hardware control away from Apple and other authoritarian vendors? Hell yes, we should.
The likely problem with the Asahi team is that they (by not accepting AI) feel that control is being wrestled away from them, and instead of adapting they choose to swallow the poison pill of GPL.
We might not like it, coding as it was pre-LLM is done for. No one is going to come groveling at the feet of those who stubbornly insist on not using it.
not sure that concept makes any sense at all. the entirety of LLMs' capability to perform software design and development tasks, such as it is, comes from the massive resource of copyleft projects, including those using GPL, that the AI companies have ingested
Well, it's a more restrictive license than MIT. A lot of popular projects nowadays are under the MIT license. AI labs still don't give a damn about how the code their LLMs ingest is licensed anyway, and GPL'd code won't stop them one bit. If their code had been used in proprietary ways, then yes, GPL would've been the way to go. AFAIK Omarchy is MIT-licensed, which still requires preserving the copyright notice of other MIT-licensed code it's based on. So I'm not sure what damage has been done to Asahi itself.
Linux itself has always been GPL licensed. It's the license switch that I have a problem with, not GPL itself. Switching from a permissive license to one less so, just so others would stop using it in their MIT-licensed projects.
> I'm surprised their work was ever MIT-licensed instead of GPL.
The intent was for their work to be usable on non-GPL distributions and OSs (including the BSDs).
The new licence is pretty restrictive — proprietary platforms will just regurgitate it with LLMs, but other open ecosystems will have to re-invent things from scratch.
Of course you can, its like putting candy for Halloween on the expectation that one kid is not going to get the whole bowl at once.
The way you can exploit an open license is to grab the code, invest massive amounts of resources in it and close the platform. Embrace, extend, and extinguish.
Spoiler: We have Vulkan 1.4 working, all of M4 and most of M5 working, are making significant progress towards full suspend to ram (no version of Asahi supports this) and this will all be shipped to users very soon!
Cool, I'm glad someone is taking a more aggressive and LLM friendly approach to doing Linux on Mac. Too bad there's no M3 since that's what I have :P
Not sure if it would be useful for you, but I did get an M3 GPU driver kinda working (it boots KDE, runs Quake and GLXGears, etc) but isn't quite usable yet. But feel free to yank the code if it's helpful: https://github.com/SuperDisk/asahi-plus-slop
Omarchy is not something I'm comfortable trusting long term, and comes with a lot of cruft.
I'd be very interested in seeing the parent post's project get M3/M3 Pro support; have it working in my system with a patchset on top of Asahi but suspend is leaving a lot to be desired. Unlike Omarchy, the parent post's diff would also be much more reviewable due to it's more limited scope.
Omarchy has no chance upstreaming any of their work, the code is abysmal.
Asahi at least guarantees the quality of their stuff because they upstream everything, sooner or later. Gravity apparently also intends to be upstream-first.
> Cool, I'm glad someone is taking a more aggressive and LLM friendly approach to doing Linux on Mac
Same. I've been waiting 6 years to run no-compromises Linux on my MacBook. At this point, I don't care if someone uses LLM decompiled firmware ported to Linux.
If I could run linux on apple hardware without compromise, my life would meaningfully be improved.
It's not well understood how important it is that humans feel in control of their environment. I feel like it's a fundamental to human happiness. Software that takes that away contributes to the general malaise and hopelessness that many are feeling today. It's just water to them at this point. They don't even see it. But it's the primary means of interaction.
Making that a true option to millions of people could do a lot of good for the world.
> I feel like it's a fundamental to human happiness. Software that takes that away contributes to the general malaise and hopelessness that many are feeling today.
I would like to see a full-featured Linux on Apple hardware too, but I think over 99% of people don't feel anywhere near this strongly. Do you feel this strongly about being able to replace the engine in your car with one you made yourself? I don't, and I think most people view computers the same way.
> Do you feel this strongly about being able to replace the engine in your car with one you made yourself?
I am not one of those people, as I do not do cars, but I would hazard a guess that most car people over a certain age do in fact feel very strongly about what modern cars have become. The ones who grew up without knowing what ownership feels like less so, which reflects the exact same dynamic with younger generations who are growing up without ever knowing what it feels like to own the hardware they use.
I think it would be disingenuous not to point out that this was also true of new cars 20 years ago - which is much more pertinent to the point of the people you're replying to.
Considering how easy it was to replace engines of most cars before the integration of electronics I'd say it used to be quite common. Just as most hardware vendors used to ship full specs for their components up until the 80s or even 90s.
So, yeah, yes, definitely. I wouldn't be working with computers on a daily basis if I had to use Windows.
I'm not gonna overly generalize this though. I don't think it'll be the same for most people but I tend to not care about what most people do 99% of the time anyways.
If it was as cheap and easy to install a new engine as it is for an OS, I imagine swapping engines would be even more popular than new OSs. (And I think far fewer would clamor to replace their OS if it cost $x,xxx for each OS swap and took many hours of manual labor - I.e. I think whether someone wants something can be heavily influenced by whether it is ~free vs cost $x,xxx, etc.)
I wholeheartedly agree. If I think about it, I must have thought "god I wish this ran Linux" at least once every single time I opened my MacBook - likely for the last several years.
I have run Asahi and while I love the project, it's not ready for that type of use.
Recently, I got a new MacBook and repurposed my old M1 MacBook Pro for local inference. The unbelievable number of compromises and hacks you need to do to run daemonized always-on software on MacOS was another reminder.
I know it goes completely against Apple's business model, but I'd love to be an Apple customer if I could buy their hardware and use it with software of my own choice. Holding an iPhone or a MacBook feels great, they're all easily among the best consumer electronics ever as far as the physical part goes. But Apple's software is completely different from how I want to use my devices so I just don't end up considering them when buying.
Would love for Asahi or similar projects to be fully usable but it also doesn't seem very likely. Apple is arguably not hostile to other OSes on MacBooks, they're designed to freely boot non-Apple operating systems, but by not publishing any hardware manuals Apple ensures that an Asahi-style project is always going to be well behind, likely years behind, in hardware support.
Oh, they’re definitely hostile, by the way, Darwin is still open source, but no one’s bothered in 26 years market inertia? Just like no one’s bothered to take a license out with SoftBank to design something from the ground up (like those three ex Apple engineers) using Linux, the easiest and fastest way to get something working from the ground up.
Wouldn’t the time be better spent taking a license out? With SoftBank to design? Arm chips that work from the ground up running Linux instead of this time-wasting exercise? Apple is one year away from M7 two years away from M8 processors Asahi are just chasing smoke…
Gravity Linux and Asahi Linux are two independent projects working towards the shared goal of a strong, open OS ecosystem on Apple Silicon devices. Gravity Linux was initially forked from Asahi Linux over differences in policy, namely regarding LLM use. We have tremendous respect for the Asahi Linux project and emphasize our shared goal of running Linux on Apple Silicon.
I just did a search for gravity linux from the search at the bottom of the HN page. HN is a fickle place as it seems you get more traction by hijacking Asahi posts that when you make your own posts.
Just remember that for however bad our code is (and I have no issues saying that most of our linux driver code is absolutely horrific, for now) it's better than Asahi code, because our code exists, and their code doesn't.
"Code is a liability" - is one of the fundamental tenets of programming.
So, "it's better because our code exists, and their code doesn't" isn't the brag you think it is. Yes, a quick thrown together utility is better than having to manually do something. But that's not the same for an OS, or god forbid, a filesystem.
Especially since the only reason your code exists in the first place is that they did the hard work and your project forked.
Yes, being able to buy a hacked-together microwave that can explode in your face and one that is tested and certified is amazing! Especially if you’re clueless about how they’re made and you just want a microwave.
Mainline Linux doesn't run on the M4. So in your metaphor, it's more like being able to choose between a microwave of uncertain quality and not having a microwave.
> however bad our code is (and I have no issues saying that most of our linux driver code is absolutely horrific
It's the operating system, mate. People don't just use operating systems to take screenshots and then post on social media.
> it's better than Asahi code, because our code exists, and their code doesn't
This is an especially horrific comment if you are really related to the project and this also adds to the perspective on why Asahi team might have wanted to have no nothing to do with this project and related people and their thought process. Anyway, good luck.
Wow man, "horrific code" can be worse than no code. Bad code can damage hardware, destroy data, allow exploits, etc. all manner of bad things.
Unless your meaning is that you already have work-in-progress code that you are working on to make much better whereas they have not even started, I don't see how your point makes sense.
But even then it debatable whether making bad code good is better than creating good code from the start.
Even good code is a liability. Admittedly "horrific" code is a 10x or 50x liability. Sometimes you accept that kind of technical debt because "it's really important" — e.g. the company will go bankrupt, or whatever.
(Historically, btw, this is why proprietary corporate code is often so much worse than even fairly incompetent open-source code. Because it's just not really that important to ship horrific slop when nobody's income depends on it.)
The idea that shipping "horrific" code now is somehow better than not shipping it until it is fit for human consumption works in some narrow business life-cycle scenarios, but not in any kind of user-facing scenario, and certainly absolutely fucking not any kind of user-facing OS implementation scenario... (T_T)
I prefer reliable software for the Operating System. Like, I can tolerate slop for a game or a one-time-thing of converting a doc or something. But for running the base software (your OS), I think you need better standards.
You want longterm support, not a set of pit of tree patches for 7.3 kernel that will stop working or misbehave in 7.4 or 7.10, with no one understanding why.
Besides not wanting hallucinations on my OS level code?
Or not feeling good for the intellectual theft involved in the training of AI to begin with?
Or not caring for the de-skilling of of human brains who are delegating coding to AI, and increasingly even the architectural and decision part?
Or the training material being AI degradation that awaits most code being written by AI?
Or the baragge of slop PRs that overwhelms my favorite FOSS projects?
Or deeping of big AI dependency, in times of climate and economic crisis that it worsens, but that it also risks not having the luxury of it forever?
Or the ease with which any stupid direction can be followed via AI, and the resulting always changing projects, already a problem with short attention span project leads even for big projects like Gnome?
Or the lack of deeper understanding to control all this in future generations of programmers reduced to "AI operators"?
>Are you saying "vibe-coded slop" because of the quality of the end result? Who cares if it gives you the features with good-enough performance?
There are a few significant reasons worth considering when getting down to the level of OS and hardware drivers. Long term maintainability is one. But another big one is that such software is capable of physically and permanently destroying the hardware it's operating on, with the textbook example for Macs being the speakers found in Macbooks. The only way to get the kind of performance they offer out of that physically light/thin package is to give them the power to destroy themselves if run at full force continuously but then run it from software so that they aren't run at full force continuously, but rather take advantage of the headroom that offers for very brief audio spikes whose overall integrated power still falls below safe limits. This all needs to be done from software since response of real world analog isn't instant. That enables much more range then if they were hardware capped at the maximum level they could run constantly, but it also means that if they're driven wrong, poof. The drivers have to never go to certain bad states, purely from software.
LLMs can't offer those kinds of guarantees, they don't really "know" how things work and then produce something deterministic off of that. It's even worse if, as-is the case with this sort of vibecoding effort, the LLM operator is in fact depending on the LLM for that part because they don't know how it really works either (vs someone who is just using it for boiler plate and knows exactly where the critical paths are).
Obviously everyone is welcome to roll the dice with their own hardware as they wish, and it may be ok. Or if using really old hardware one might just not care. And not all hardware is as sensitive as modern notebooks of course. Just be aware the analog world can be messy and you might not actually have the guardrails to screw up that you're used to in a higher level software environment.
>Otherwise one can also say Asahi is compiler-generated slop compared to hand written assembler code.
No, one cannot say that. Or at least not without looking really stupid.
If that issue with the speakers is documented, the LLMs of today would handle it correctly. That wasn’t true very recently. Also, assuming documentation exists is a big crutch that helps them more than us.
Hasn't been shipped yet, integration is being done now, we're shipping a new installer, from scratch graphics driver, and new kernel so there's a lot of moving parts we want to get right.
Any plans for supporting M3 Ultra? I understand it's a small percentage of the Apple Silicon Macs out there, but I'd love running Linux on my Mac Studio?
> It may surprise some to learn that WINE was originally made available under the MIT licence. The 90s and 2000s were a very different time in the Free Software ecosystem, and permissively licenced projects were viewed by tech companies as free real estate to pilfer and exploit. WINE was not immune to this, and many monetised proprietary forks of WINE were marketed and sold by various bad actors. The WINE development team grew increasingly concerned that these proprietary forks, while well within their rights to exist, would eventually cause so much fragmentation in the ecosystem that WINE itself would become unviable. In 2002, WINE was relicenced under the LGPL to prevent further proprietary forks from being created, and to encourage direct upstream contribution.
> When Asahi Linux was first started, it was with the intent that we develop a Free Software ecosystem for the Apple Silicon platform. To facilitate and encourage the particiation of our friends at the BSDs and other Free Software operating systems, we endeavoured to licence as much of our code as possible under the permissive MIT licence.
> Unfortunately, this is no longer viable. The malicious practices of the 90s and 2000s have reared their ugly heads once again, this time driven in large part by both the insatiable appetite of frontier AI labs for high-quality training data, and militant animosity toward the Free Software community.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
Hear, hear! Perfect clarity of thought. I love these people!
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
Attribution is the one and only condition of the MIT license, aside from the disclaimer of warranty, so I don't see how it "explicitly allows" publishing without attribution.
> Unfortunately, this is no longer viable. The malicious practices of the 90s and 2000s have reared their ugly heads once again, this time driven in large part by both the insatiable appetite of frontier AI labs for high-quality training data, and militant animosity toward the Free Software community.
It makes no sense to me why you'd adopt MIT in the first place for its permissiveness, and then be surprised that people you don't like are using your code. It's like being a free speech supporter and being surprised that people say things you find abhorrent. That's the point. If you only support speech you agree with, you don't believe in free speech, and if you only support code being developed by people you agree with, then you don't believe in free software.
>It makes no sense to me why you'd adopt MIT in the first place for its permissiveness, and then be surprised that people you don't like are using your code
It makes no sense to me that this makes no sense to you. The Asahi Linux project kicked off in December 2020. About 2 years before the first ChatGPT (Nov 2022). It had a lot of ideals including working with the BSDs at the time, and so they surveyed the landscape and considered the downsides at the time vs the upsides and decided a BSD approach was the right tradeoff. Now, 6 years later, the landscape has changed pretty radically, so they've decided that the downsides have gone up and any value they're getting from the BSDs is not high enough, so they're changing course. This is not hard to follow.
>It's like being a free speech supporter and being surprised that people say things you find abhorrent
No, it's nothing like this at all. Though utterly tortured, convoluted and wrong analogies to free speech are common from people who don't understand it.
>If you only support speech you agree with, you don't believe in free speech, and if you only support code being developed by people you agree with, then you don't believe in free software.
It is apparent that you are overloading "support". Free speech is about opposing the use of violence (direct or indirect) to decide debates as a near-universal default. That's it. Social and economic consequences unbacked by violence are not opposed to free speech, they ARE free speech themselves. If I think your opinion is wrong and stupid, I may "defend to death your right to say it" but then not invite you over to dinner, donate you any money/time/resources to get your message out further, etc etc. That is my own personal free speech in response to yours. And you can criticize that in turn and not invite me and so on and so forth. Free speech preserves this unending dynamic.
By the same token, believing in "free software" is easily overloaded, and claiming that using the GPL means someone doesn't believe in free software like you just did is one of the most absurd and also ancient, tired bad faith attacks that's been around for decades. It diminishes you to make it.
> Free speech is about opposing the use of violence (direct or indirect) to decide debates as a near-universal default. That's it.
I don't believe that's it. Free speech in addition to being opposed to violence to silence individuals is also against censorship. If you want to publish a book of opinions society finds repulsive, you should not incur violence for saying so, but also you should not be prohibited from publishing your views.
Now of course there's a difference between you not inviting this person to dinner and the government prohibiting this person from publishing their speech, and I understand that. Even before ChatGPT, there still existed bad actors and especially with MIT licensed software, it was inevitable that people were going to take this very valuable code and use it in closed source projects.
Asahi has every right to examine this new landscape and make changes to their licensing as a result. But speaking personally, I release my software under an MIT license specifically because I don't care who uses it or how. I'm not criticizing Asahi so much as I'm pointing out to avoid the MIT license or other similar ones if you have any opinion whatsoever on how your code is going to be used in the future by your friends and enemies alike.
> The Asahi Linux project kicked off in December 2020. About 2 years before the first ChatGPT (Nov 2022). It had a lot of ideals including working with the BSDs at the time, and so they surveyed the landscape and considered the downsides at the time vs the upsides and decided a BSD approach was the right tradeoff. Now, 6 years later, the landscape has changed pretty radically
Asahi themselves cited WINE forks from the early 2000's as a case study in "bad actors" abusing permissive licenses. So AI really has nothing to do with this, they were well aware of the precedent in this area.
To me it's just another example of this weird hypocritical stance some free software people have. They love putting the code out there and letting it spread far and wide, but once someone _else_ starts making money building off it, "hey no fair!"
I still don't understand the motivation for Asahi and other similar efforts. My impression is that Apple fully supports Linux virtualization on Apple Silicon. Why isn't that sufficient? Just asking.
From the perspective if the people working on it, I imagine it's immensely satisfying when they successfully reverse engineer something and get it working. It's not on the same scale or of the same complexity, but I feel that satisfaction when I finally understand something I've been trying to reverse engineer in a retro game.
From a user's perspective, I can only speak for myself, but I'm grateful for the Asahi Linux project's work. I have less than zero interest in ever running macOS again. It's user-hostile, constraining, and not a good experience. I prefer Debian with Plasma 6. But Apple makes high quality hardware, and they are one of the few choices for a fast ARM laptop with excellent battery life. I avoided getting one for a long time until Asahi got to a point where it ran pretty well on the M1 and M2 with relatively few caveats. Then I bought an M1 Air second-hand, and have used it exclusively with Linux ever since. I'm very happy with it. My only complaint is that I have to waste a big chunk of disk space to keep a macOS install on a separate partition that I can never remove and occasionally need to boot into to get firmware updates since I can't get them through fwupdmgr like I can on most of my other systems.
Same as why so many people run Linux natively on their x86 machines. It's just part of the preferred setup because it's nicer in so many ways and there's no annoying proprietary host os in the way.
Some people might want to get all the apple software off the apple hardware. For example the dossier that macOS builds on users and their family behind the scenes; currently on the front page of this website.[0]
Unfortunately it appears that Asahi has promised to not allow doing that[1] but perhaps it will enable someone else to make it possible.
I'm glad that Asahi is switching from a cuck license to a reasonable copyleft license. Even though the circumstances that made this necessary are, of course, unfortunate.
To the people downvoting this comment: I get it, it's a vulgar choice of words. But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
Choosing GPL and similar licenses is the only way you can publish open source software without running this risk. As professional programmers, we must insist that people either pay us for our work or act as a equal collaborators.
And yes, there's a reckoning coming for AI companies who trained their models on GPL software.
> And yes, there's a reckoning coming for AI companies who trained their models on GPL software.
What would that look like in your opinion? Hard to fathom that the people with all the money and power could get into trouble over a licensing dispute.
> Hard to fathom that the people with all the money and power could get into trouble over a licensing dispute.
It's because of something called "Rule of Law"... Most free societies in the world have it.
If you find it difficult to fathom that rule of law applies to corporations, maybe it's time to start practicing fathoming it, or they will have won before the fight even started.
I’m aware of the rule of law, and enjoy living in a country where it (seamingly) still applies. I’m just questioning the viability (in practice, not in theory) of legal action in the US (where this would likely play out) given the economic hurdles to clear for the accusing party and the general chaos.
You're referencing "rule of law" but... what exactly do you mean? Bartz v. Anthropic resulted in a ruling that training from copyrighted books is fine if you buy them. Doe v. GitHub dismissed DMCA claim, saying Copilot generated code doesn't infringe on open source licences.
So far the precedent and the law implies the open source licence won't matter for training and the result. And these decisions will now be brought up in the US in any new cases raised.
> But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
GPL doesn't prevent that at all. That's not why Asahi made that change. It's about contributing back to the project, not the money they want to see.
It doesn't change the matter of fact that it's insulting language, and I downvoted it either way.
Developers have their own good reasons to stick with permissive licenses. Either because they are completely selfless, or they want to help users who want to kick-start their own business selling software. Just virtuous reasons.
> But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
That is not a fact at all. I choose to use permissive licenses because I am giving my work to the public out of charity. If someone uses it to turn a profit, and I never see a dime, I have not been exploited or abused in the least degree. It would certainly be virtuous for the people making a profit to give something back, but it isn't required. If I did require that, it wouldn't be much of a gift.
A reckoning for what? I've also read some GPL source, doesn't mean I'm not allowed to publish closed source software, as long as I'm not copying anything.
For the people wanting "Apple hardware quality with Linux", has anyone here been using the Framework Pro? I keep staring at it in my shopping cart but tell myself my regular Framework 13 is enough for the occasional use I give it.
This thread is full of folks saying, "Asahi can't keep up with Omarchy," when Omarchy Mac's own readme[0] states that it is, "Asahi Alarm + Omarchy." That's like bragging that someone sitting on Lebron's shoulders can dunk a higher rim than he can.
But at the end of the day, like in your example, you have "Lebron" vs "Someone sitting on top of LeBron's shoulders" facing off against each other. Who are you putting your money on, to get the higher dunk?
Ah yeah, a .dmg installer that asks for sudo permissions is much safer /s
Why do people get hung op on this so much? You can just read the script before running it. Of course you won’t though, just like you don’t read the source of the .dmg installer.
That's a very complex question that entirely depends on who you're trying to defend from, more than I can explain in a single comment.
The short of it is that you can make it relatively secure, but it will never reach the same levels of security as macOS. The tight integration apple has allows them to do far more for security, for better and for worse.
Worse being apple putting the entire might of the platform's security into making sure the Chess app can't be uninstalled or tampered with.
So many people that never wrote a line of code outside of school assignments in this thread.
People that did and have some opensource projects would never write some of the silly stuff I'm reading here. I don't expect permissive licenses to be popular for long.
honest question - how hard it is to use LLM and tools like REA to reverse engineer MacOS to figure out how its hardware works in terms of getting linux support on such hardware?
I thought it is already a solved problem? with large enough numbers of agents running in parallel reverse engineering, implementing kernel modules, testing in vm etc, the whole pipeline seems to be complete.
Yes, read that blog entry to understand how different that hardware is from any other Linux-supported ARM stuff to see how much deep understanding you need to actually bring that support upstream to Linux or Mesa. Some(most?) stuff requires fundamental changes and no LLM is going to design and write it in an uostreamable way.
Debugging is not an issue, Asahi seems to have it figured out, especially with the virtualization. It's the design that takes time.
Meanwhile Omarchy, for example, just brute-forces everything, with ugly patches that work today but cannot be upstreamed as such, so they'll break tomorrow with new kernel release. I really expect them to start struggling a lot going forward with maintaining the support they claim. Regressions may actually kill their effort.
You must not work with AI tools very much. The brute force approach as a first/second pass is very standard when doing reverse engineering with AI. You get the functionality figured out and get the basics working and then you just iterate on that and massage it into what you need it to be. Omarchy has practically unlimited tokens and every other week there is a new model being released that is more captable than the previous that will just massage things into place with each iteration.
And yet they have a total of zero upstreamed PRs, and haven't fixed enormous and glaring issues like PSCI control either.
It's almost as if the parent is correct, and Apple's ARM architecture isn't standardized enough for ordinary models to blindly reverse-engineer without documentation. If tokens were the only constraint on supporting Apple Silicon, we'd have reversed the entire SOC device tree years ago. Alas...
> And yet they have a total of zero upstreamed PRs, and haven't fixed enormous and glaring issues like PSCI control either.
Why should I care about if they upstream their patches or not? I have a working M3 macbook that runs Omarchy, why do I care that it's in the mainline Linux kernel? My GrapheneOS phone doesn't run mainstream Linux kernel, but it does the job that I need it to do.
upstreamed PR means nothing in 2026. I couldn't care less when some maintainer are just so political, still remember the rust in kernel drama a couple of years ago? why great work have to get approvals from those people?
They prohibit RE of the binaries. What Asahi doing is observing macOS as a black box with input and outputs, and translating observed behavior back to Linux.
>As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only.
This made me laugh. For context: some people used LLMs to import code to get Omarchy to work on macs. So, these people are switching licenses so that they could stop competitor distros from being compatible.
So, before, Asahi people were happy with Linux working on macs; now they want THEIR linux working on macs (they are not changing to GPL out of the goodness of their hearts or that they believe in everything being open, remember, they are only doing it so that Omarchy, or similar, can't copy their work without having to be GPL themselves).
Meanwhile Omarchy is MIT; which means Omarchy is allowing Asahi (and everyone else) to copy their work, with the actual intent of getting Linux everywhere.
True. Was Omarchy doing that? (I'm confused why changing the license matters, if both projects are already open source and releasing all the code to the public anyways)
This really leaves a bad taste in my mouth too. It's unfortunately inline with their absurd AI policy. Instead of working for the good of the users they literally are trying to halt progress. It's pretty clear the omarchy people will run circles around asahi very soon. Asahi people think this too and are trying to limit them as much as they can with license changes.
> with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon
which I hope everyone knows is completely dishonest. Nothing in Omarchy is about monetizing Linux; and monetizing Linux is and has been completely fine since its inception.
Asahi people just seem to be sour that someone else is joining the initiative but in a different project where AI use is encouraged to get to the same goal faster.
GPLv2 is a great license. Omarchy is literally taking their code without properly crediting and sloppily extending it. This is a good thing for everyone.
After seeing the blazing progress that Omarchy has been making with making Linux work on the latest macbook I'm planning on switching all 3 of the M1/M2 to Omarchy and planning on buying an M3 macbook air in the future. I appreciate the work that Asahi did to get the ball rolling, but they've just been left in the dust with their anti AI stance.
> but they've just been left in the dust with their anti AI stance.
How, exactly? AFAIK, Asahi officially supports just as many Macs as Omarchy does. The vast majority of Apple Silicon support in Omarchy comes from upstream Linux changes merged by Asahi, not any of the Omarchy devs.
The only feature I'm aware of Omarchy adding is the fingerprint reader driver, which is in such a poor state that it could never be merged upstream. Which, if that's the standard of quality Omarchy is holding themselves to, they're the ones getting left in the dust.
> they've just been left in the dust with their anti AI stance.
AI proponents always act like this is self-evidently a bad thing, but it isn't. When someone is going in the wrong direction, it is good for them to leave you behind. I believe that projects which embrace LLMs are going in a bad direction, and I'm glad to see more sensible maintainers get "left in the dust".
The patches to the Linux kernel have been about 50% made by AI in the past 6 months. The magic of open source is you can just fork it and go at your own pace and do whatever you want, and others can embrace whatever technology they want and then market adoption just speaks for whichever is superior.
How is making your project better — more feature complete, more secure, more performant and optimized — going in a bad direction? This is all platitudes and virtue signaling. Give me substantive reasons LLM “slop” is worse than human “slop”.
Three ex-Apple engineers did it by taking out a license with Arm Holdings/Softbank to form their own company and were successful enough to be bought out by Qualcomm 1.2 billion dollars, isn’t the Linux community big enough to support a processor from scratch, taking the same path? Taking that same path seems like a no-brainer if you have the skills.
Imagine deliberately buying a $2000+ computer that doesn't work with and is actively hostile towards your OS of choice, then embarking on a multi-year bitch-fest to shoehorn the OS onto it.
I struggle to wrap my mind around it. It's like buying an EV so you can convert it to work on gasoline.
- MacBooks are still among the nicest if not the nicest laptops hardware-wise, so some people like to run Linux on it.
- People have older Macs that they like to repurpose. E.g., we still have an M1 Mini that was used by our daughter before she got handed down a MacBook from my wife. It's now collecting dust, but I might install Linux to use it as a headless server.
> - MacBooks are still among the nicest if not the nicest laptops hardware-wise, so some people like to run Linux on it.
By the time said macbook is fully supported, it is in sorry state. Judging by what I found when looking for second hand laptops, macbooks don't age well.
> Judging by what I found when looking for second hand laptops, macbooks don't age well
Are you dumpster diving for your MacBooks, perchance? I buy all of mine second hand, they are all still mint by the 5 year mark, and I have several at the 10 year mark that all I've had to replace is the battery.
Huh? Apple has many problems. but one of them isn’t the resale value of their hardware. Or maybe I should say, if you buy Mac hardware, you don’t have a problem selling it on, not like a used PC the only thing worth selling on in a PC is the graphics card (Nvidia) and the memory.
I bought an M1 Ultra 64G off eBay in March with some 30 battery cycles, it looks new. This thing is a beast compared to most consumer hardware even today with its 800GB/s memory, runs small models well. When I get better HW for LLMs it's going to be a great Linux machine for a decade, well past the point Apple stops supporting M1 entirely.
Isn't that true for anything? Some people will sell 5yo hw in almost mint condition, whole others treat their hw like ass and it's garbage after two years already.
The problem with used Macs is that they're still really expensive, and if you want to only run them with Apple software, the support duration is limited (especially true for phones, even more for Android ones). You can pick up used Thinkpads or Dells in great condition for very cheap after a few years.
Not at all. I've used Linux on Thinkpads starting in roughly 2006 (switching away from MacOS), and they always worked flawlessly.
One major FU was with the 5.x kernel when suddenly it no longer supported my external monitor, this with Intel graphics (and one by one all distros upgraded to that kernel). Not sure what happened there and why, but I had to do using Linux for a while because of that.
For me the definition of actively hostile would require a locked down device that needs an exploit/jailbreak to install and a cat&mouse patch/exploit cycle - but it seems that Apple intentionally designed this to be allowed - what is giving you the “actively hostile” impression?
I used Windows and Linux all my life, but got my first MacBook Air M this year. There is literally zero competition to it. It's an incredibly powerful and cheap yet fanless device. Also, it has uniform memory, which is awesome.
I wish I could install windows or linux on it because MacOS kind of sucks, but the device itself is amazing and reason enough to switch away from PCs.
> but the device itself is amazing and reason enough
By continuing to reward Apple, they lose any incentive to improve macOS and it will continue to get objectively worse because people keep giving them money. Why would they fix things if you keep buying it?
Strange argument. Their hardware is without competition, that's the reason I use it and also warrants rewarding them with my money. Why should I reward Linux for (mostly) only working on hardware that sucks in comparison?
I used MacOS back in the Panther and Tiger days (10.4), then switched away, and started using Macs at work starting with maybe Monterey.
I was very surprised that almost nothing had changed in that long time! They got virtual desktops (finally), and widgets changed from an expose-like effect to only becoming visible at the right side of the screen, and tbh that's about it. All other "features" seem to be changes to Apple's built-in apps.
I think the OS still sucks about as much as in 2004, it's very much a matter of taste.
I got an M2 macbook air with 16GB ram and 1TB SSD for $700 and install Asahi Arch Linux and am planning on migrating it to Omarchy now that it supports M chips. It's my favorite laptop out of the Dell XPS, Precisions, and thinkpads that I own. The fact I can get 7-9 hours of battery life, have everything feels snappy while I'm doing Dev work and have a very crisp display all at the same time puts the Intel/AMD chips to shame.
Supposedly Intel is actually giving Apple a run for its money this year after 6 years of being left in the dust, so maybe I'll get something Intel next year.
To go with your analogy, if I can buy an electric car and replace the engine in it with something that does what I want and gives me 95% of the usable features of the locked down engine that it ships with, why wouldn't I do that? Especially when the rest of the cars on the market are getting 100 miles to a charge when my Apple M2 is getting 300?
Not sure what will happen when people will directly contribute to the Linux kernel with LLM patches though for these devices, I guess they will have to live with it.
Asahi is patches on top of the linux kernel. If you independently get your drivers in to the kernel which does allow LLMs, the drivers will end up on Asahi too.
I saw they managed to apply their no-llm policy on top of their upstream mesa driver but I'm not sure how long that will hold if the forks are getting too popular
Sure, but a human could read the code and look for ideas/parts and write the actual code better than an AI can. Omarchy is 99% hype, but the fact that its working on osx is telling me that it is indeed a goal you can accomplish. I rather not have my OS be a slopfest, so thats why asahi is way better than omarchy ever can be.
I didn't know about their LLM stance, but I feel like Asahi are shooting themselves in the foot. Giving up the advantage to toxic teams like Omarchy is strategically wrong. They could use llms for reverse engineering and drafting and as you said still rewrite the final versyof the code.
Maybe to them it’s not a race, and more a journey?
Or to put it differently, why the rush? It’s not like they’re trying to corner the “Linux on Apple Silicon” market, last I checked?
And to add to that "journey" story, they probably also understand that writing the code is only one part of it, and they probably want to remain in a position to maintain it
It's not about the journey if you don't want (certain) other people to use it as they have so far, but rather about status, and they're afraid of losing that status.
Asahi is working on Macs too. Omarchy just took their existing drivers and wrapped them in fascism. They're also supposedly vibecoding support for newer Macs (which Asahi is working on too, but without slop, and Apple making changes to the platform every year doesn't help).
I’ve just build the kernel from Omarchy on my fedora asahi remix and got thunderbolt and Touch ID working. You do need to patch m1n1 with their changes as well. But I’ve been able to finally ditch an extra usb hub and hdmi cable.
In the PR’s that were merged in that kernel I also see they are actively working towards m4 and m5 as well. Must be hard for Asahi to keep up this way.
It's hard because they write things with intention to upstream it. Omarchy patches simply cannot be, because they go absolutely against the established way of doing things. No one consciously designs anything in their code, it's all LLM output.
I personally vetted that TouchID code you mentioned and it's an absolute hellhole, with everything done in-kernel and no userspace separation what so. That code is also a security nightmare.
Its basically useless gui thing. Nothing i cant have in another distro. What im intested in is the osx port. Apple hardware is unbeatable, but the OS has been going downhill since jobs passes. This is why true linux on mac is my dream setup. Asahi is the closest thing so far.
apple hw is unbeatable indeed. but the os was always for normies, a walled garden.
regarding 'the useless gui': it was what actually was missing in in the linux world.
i used kde before, which is fine, but it is too bloated and kwin_wayland is unusually heavy on the machine: it maters, especially for mobile devices like laptops.
of course if you use macos and apple hw, you don't care about it , since the stance. i don't run (yet) omarchy on my servers. but there are a ton of other, especially developer / ai friendly, defaults that would make one tinker for days with other distros. as i said i had my own "dot files", but a community built distro with the right defaults is much better energy management than everyone on their own.
I have used Linux for 21 years, on and off on the desktop, and all the time on the server. In the last 10 years I've had basically not time to catch up with what's new in terms of desktop environments and other tools (e.g. neovim, terminal emulators), much less have the time to tinker with any of them.
Omarchy (first it was Omakub) changed that by serving it all up to me in less than five minutes. There are, perhaps, other similar projects, but this one got to me first. For me, it's an improvement. Someone else's defaults aren't a constraint as long as I find them useful. I only wish M4 Pro was supported, because macOS (even though I'm still running 15.x) has gone downhill.
It's, of course, less useful if you have 20 years' worth of dotfiles and custom workflows that you aren't keen to replace and I fully respect that.
There's nothing to catch up on. It's a kernel. It's systemd. And then pick a desktop manager. Every distro installs in 5 minutes. Because on Linux you literally just copy/paste. It's just a sloppy fork of Arch.
You are tech dreaming any group working on this? Should be looking to do their own arm processor. by taking out a license with SoftBank and designing something from the ground up.
> Fedora Linux 45 was released on 3 November, and we shipped Fedora Asahi Remix on the same day. This is the first time we have been able to get a new version of Fedora Asahi Remix out on the same day as upstream!
This is a wonderful thing to read. I've been a proud financial backer of Asahi since the first announcement of their Patreon account (now OpenCollective). I've only installed and booted it once on an M1 MacBook Air to kick the tires. I liked what I saw. It worked well enough to validate the work being done. I then reinstalled MacOS so that my partner had dedicated modern laptop. I also doubled my donation.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
Oh man, I just doubled my donation again. I'm now at 4x my original donation based on reading this because hell yeah. Congrats to the team doing this stellar work. Now that my partner has a MacBook Neo, I'm gonna wipe that MacBook Air and test out the latest release. I'm so happy folks are doing this.
I support Asahi due to ideological conviction and self-interest (in case I ever want to have an alternative on Apple hardware) rather than because I want to replace MacOS. But who knows. If Apple's rent-seeking behavior with the App Store grows and further disrupts my opinion of them, maybe I'll end up on Asahi full time.
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
Unfortunately, the license allows this. It’s also a social issue - why shouldn’t a faster mover be able to slop a feature faster than the project owners? Who cares if it’s disruptive to them?
I’ve seen this happen on other projects and beyond rejecting the changes, I’m not sure what to do. It’s damaging morale when someone has worked weeks or months on a feature, worked with countless individuals on planning and implementation, and then someone releases a vibe coded slop fork with the last 5-10% and then takes credit as if they’re some genius doing something useful or helpful.
Ultimately the issue is manners, but when there is opportunity and generosity, the generous are going to be taken advantage of.
LLM haters are the worst type of toxic; they will not stop until everyone else stops using it, too.
Punitively changing a license with an intent to further restrict what someone does with the code is childish. If the concern is "someone will feed it into an LLM to obfuscate it" — what stops them from doing that with GPL code?
GPL code is, by the way, ripped off on a daily basis and routinely ignored because it's bordering on impossible to enforce.
In some cases LLMs are a legal issue. In other cases they point to low effort contributions. Those changes require the same amount of effort to review, but are no effort to produce. When there is no mechanism to control a resource sink, everyone suffers.
Tragedy of the commons where the resource being depleted is our time and attention. The typical solutions don’t work on that resource!
Maybe they're hoping that GPL scares off the larger commercial players. Works in my case.
GPL will possibly scare off serious, risk averse companies with lawyers. It's not going to scare of Chad the AI Bro who wants to move fast and break things, crush it in the market, ride the VC rocket, and has no qualms about monetizing the work of others.
Oh right, because serious, risk averse companies are going to be using a linux distribution that does not even currently run on an M6 Mac.
> the larger commercial players
Who are these large commercial players worried about Asahi Fucking Linux of all things?
Oracle? HP?
Is EMC building a SAN out of recycled MacBooks?
Not to discredit any of the work they've done but this is at its core a hobby project, and no large commercial player is touching this with a 10m pole.
People who oppose the use of LLMs have the right to do what they want with their software just like anyone else does. It’s not “childish” or “toxic” for them to express their opinion by changing the license on their own software or writing about it on their own blog.
Furthermore, if the GPL is really as unenforceable as you say, then there’s no reason why their license change is “punitive”. Either the GPL works, in which case the Asahi Linux maintainers have the right to restrict use of their software, or it doesn’t and there’s no reason why anyone should be upset by a gesture of free speech.
The Asahi Linux maintainers have the opinion that LLMs are harming their ecosystem, and if they want to prevent people from using LLMs on their software, then they should be able to do so without being unfairly attacked.
> if they want to prevent people from using LLMs on their software, then they should be able to do so without being unfairly attacked
They are of course free to wish whatever they like, but also rightfully deserve — and have earned — steadfast mockery.
It's like the space program banning computers and going back to chalk and black ladies doing long division because programmers might make mistakes.
You got a really thoughtful comment, and you reply with the most bizarre case of false equivalence I've read in a couple of days...
It's not that I am against LLM because I'm using them, but I get sick of reading that mentality of AI-must-learn-from-everything-and-if-you're-against-you're-against-progress.
It's an incredible yet immature technology, and (too much IMO) people want to use it for everything without asking themselves any question. There was and still is a whole world of technology and knowledge before AI/LLM that left us teachings, laws, technologies that we must live with. Some of these things are here for good and maybe others for bad reasons.
In any case you should take your time to understand why something is there before blindly disagreeing with it. That's what moves discussion forward.
The only thing that is holding me back is easily encrypting the drive.
I just ran this the other day on a new setup: https://github.com/osx-tools/asahi-encrypt Worked well.
Thank you!
Hello since people are moving to Linux on their macs what about all their data on mac like uk iCloud Photos notes calendar reminders etc.
it's not like icloud's website stops working
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
I appreciate all of the hard work that's gone into Asahi linux, but this statement is basically nonsensical. Making the license copyleft does not prevent people from monetizing it, using AI to develop on it, improve it or wreck it. All it does is make sure that anyone who does will share their source code modifications.
It doesn't sound like they want their code to be open source. It sounds like they want it to be closed source or at least restrictively licensed enough that they can tell users of the source how they're allowed to interact with it and modify it.
The entire point of open source / free software is that it lets you do what you want with it. The original author doesn't need to accept your AI patches or write AI code themselves, but downstream users are not restricted.
I am just baffled by how confused even very smart people seem to be about how open source and permissive licenses work and what they are for.
(Now, should they use trademark law to ensure others can't use the name Asahi Linux confusingly? Absolutely. But that's got nothing to do with copyright.)
You are confusing why they choose MIT (to further adoption by other existing distributions and allow other developers to reuse pieces) and how it is being used (to build cheap distributions on it slapping a different name on front), they understand well the implications as anyone else.
> I appreciate all of the hard work that's gone into Asahi linux
Actually no, you don't seem to. Regardless of the license, remind yourself that they didn't lose ownership of what they did. They can still be disgusted at how people are building lazy derivatives just because they wanted to share what they did with a different kind of developers. I know I would.
Frankly, I'm not sure if permissive licenses have a future. No one wants dev marketeer to save tokens reusing their work.
They’re switching to GPL and you think that’s anti-open source? MIT has basically zero restrictions and assumes a good-faith collaboration environment. GPL is more realistic about the motives of much of the tech world, and in practice is much more encouraging of collaborative, open, and free software development. Kudos to them on making the switch.
The only blocking point for me before I would use Asahi Linux is the support of the USB-C displays. I was considering and reconsidering to install Asahi so many times, but I just cannot give up the ability to plug an external display into my MacBook Air.
I am just surprised that other features like support of the newer CPUs still get a higher priority than this one. I am not complaining or anything, I am not even related to the project in any way, so not in the right place to ask for anything, but I really am surprised that this seems to not to bother most of the users?
This works, tried last night in my M1, you need to install the fairydust kernel and reboot. https://copr.fedorainfracloud.org/coprs/lacamar/fairydust-ke...
My Dell external monitor (+builtin usb/ethernet hub) connected via usb-c thunderbolt started working right away.
My problem is I can't use my 2nd external monitor. :)
Also no luks fde support which is a big deal, basically I can't take it anywhere.
Huh. Thank you for the info!
I am always looking at this "Device Support" page [1] and it's still red there. I even have a calendar reminder to check it out every once in a while, but if it's not being updated I might drop my calendar reminder then :)
[1] https://asahilinux.org/fedora/#device-support
I developed the habit that letting AI assistant look up information like this for me. It might be hidden in the docs I usually over-look.
> no luks fde support
You can get an "unecrypted boot, encrypted root" setup though.
https://davidalger.com/posts/fedora-asahi-remix-on-apple-sil...
Did you consider that? I am personally using that and feeling pretty good about it, but maybe I am missing something?
Thanks for sharing, I'll try that.
Oh wait, there is also this tool which automates some of that. I think I used that tool instead of the blog article itself. It's the same idea though.
https://github.com/osx-tools/asahi-encrypt
Sometimes this can be because the people working on one aspect of the project are not the same as the people working on another, or the same number of people, and also maybe supporting later chips is less problematic than supporting external displays. The project as a whole doesn’t just work on one problem at a time in priority order.
I believe the blocker here is that USBC display support requires getting full Thunderbolt support working. Apple built their own Thunderbolt chips, which diverge from the Intel hardware that everyone else has used & that the Linux kernel drivers are written against enough to make support much harder than you'd expect.
IIRC Asahi has finally acquired Thunderbolt support but that support still doesn't have the pieces needed for external displays.
The entire section "Kept you waiting, huh?" is about this and some of the background is really interesting explaining why things are as they are. Here's just the last bit:
> Luckily for us, Sven was up to the task of dealing with all this nonsense. It just took slightly longer than expected! While the patchset to enable Thunderbolt was sent upstream first (another massive win for us!), we have also pulled it into our downstream kernel and are now shipping it to users!
> But there is a catch. Currently, only USB3, XDomain and Thunderbolt Networking are supported. You can plug in USB devices or connect two Thunderbolt computers together and make them ping each other, but that’s about it. PCIe and DisplayPort tunnelling — required for most Thunderbolt docks — are not supported. Yet. Watch this space.
> An enormous amount of gratitude is owed to Sven, who has stuck with this thankless chore for nearly 5 years in between all of his other incredible work.
Thank you, Sven!
As an old-time Linux user I wish Asahi all the best, and I get why they want to have the apple silicon working. But it just leaves a bad taste in my mouth to support a company that has historically been somewhat hostile towards Linux/FOSS compat and the power users while in the same breath trying to proclaim their platform as developer friendly, nevermind the "apple tax" that has to be paid on commodity hardware. The hacker spirit isn't aligned on this one.
Unfortunately this has been going on for 20 years and people refuse to learn.
"My [GPU/whatever] doesn't work right"
"Why don't you buy hardware that officially supports Linux?"
"I couldn't possibly"
Because sometimes people don’t go out of their way to buy Linux but want to repurpose old stuff they already own to experiment a bit and/or extend its life? For instance, I recently installed Ubuntu on my 10yo Intel MBP from 2015, and had to tweak lots of things to get it to a somewhat usable state that doesn’t flicker/tear/freeze. Had to use X11 instead of Wayland for anything Chromium-based. One of the most amusing ones (for some definition of amusing) is Bluetooth audio not working, turns out Ubuntu uses hot garbage codec that can’t fit into the available bandwidth due to AAC licensing (?), possibly among other problems, and switching to an AAC-enabled build made it reasonably stable.
Wild to limit yourself to what the manufacturers “officially support”. May as well just run macOS with that attitude.
I agree that you shouldn't rule out devices that don't have official support for Linux. And it's amazing that the community brings it to some many devices.
But if you already know at purchase that you'll install Linux on the device, you might as well prefer those with official support rather than just a community port, simply to show support for companies that agree with your priorities.
Also you should inform yourself beforehand. Don't just buy random hardware and then make the surprised pikachu face when it happens not to support Linux.
Uhh, this ethos of “can I make Linux work on this hardware” has been a huge driving force of Linux’s success. This sort of effort is deeply embedded in the culture and always has been.
Apple makes very high-quality hardware, and unlocking the full capability of that hardware via strong Linux support is absolutely a noble cause. Rather than giving in to corporate lockdown, the Asahi folks are gifting the world a way to take advantage of this new lineage of Apple Silicon based personal computers.
In 10 years when Apple has long abandoned support for the M1, M2, etc, that hardware will still be able to run strong thanks to the efforts of this team. I salute them.
If you think Apple has historically been somewhat hostile to FOSS, you'd hate to hear about the company whose logo is on every other laptop
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon.
Is this referring to Omarchy? Anyone have context? TBH I'm surprised their work was ever MIT-licensed instead of GPL.
Its the monetizing part that makes me think its not Omarchy - they have no monetizing things.
They're referring to Omarchy, it's very clear on their IRC
I saw that's there is an Omarchy for Mac "team" consisting mostly of one guy who is active on Twitter (and quickly amassed many followers) and seems to have no prior experience of working on Linux kernel. They do not mention Asahi anywhere which isn't surprising to me given the reputation of DHH and Omarchy. This hijacking of open source by bad faith actors is very concerning.
Yeah, on their GitHub for example, in their m1n1 fork they tend to refer to "upstream" as... the main branch of their fork.
Very interesting choice of words, but again, all their work including all issues and docs are LLM-generated.
Nonsense.
> Omarchy on... > Apple M1/M2 chips# > Asahi Alarm is a version of Arch for Apple M1/M2 computers built on top of Asahi Linux. You can get Omarchy running on top of that with some effort. See the user-driven guide.
Source: https://omarchy.org/manual/omarchy-on/
> [...] You finished what Asahi started, and you’re already pushing on to M4, M5, and M6. Thank you!
Source: https://omarchy.org/news/2026/11/marcelo-alcantara-and-birk-...
It’s a little more complicated than this (but not better IMO)
They do have large entities donating to them
Omarchy has gotten somewhat of a reputation for getting a lot of sponsorship money from commercial actors. If by monetize you mean "Get people to pay us" omarchy are definitely monetizing. Not much different to Asahi themselves of course, but is it right/moral to fork another active projects code and use it in order to gain your own patrons? I don't hold the answers. But no they aren't selling anything.
edit: worth noting this example, where DigitalOcean started cutting OSS support only to pledge millions of dollars to Omarchy. No wonder people consider the project a leech to the OSS community, but also consider correlation != causation: https://itsfoss.com/news/digitalocean-open-source-credits-en...
DigitalOcean obviously cut their OSS support because they gave it as credits for a year and it required constant human review and interaction.
At the same time it was mostly to small OSS projects that dont have formal non-profits so no tax exemption benefits to DigitalOcean.
It is worth noting the same article's discussion of nodejs reads as "DO shut down a program, possibly planning a v2 of said program, but failed to communicate it to its dependents and didn't realize until they started planning to leave". Or they were just fighting the fires caused by a github issue named "Transition off of Digital Ocean" caused \o/
Yes. But how is GPL going to prevent that?
Is Omarchy trying to monetize Linux?
I mean, I get the poisoning the well of open source thing that some companies do. I didn't think Omarchy was trying to do that.
With how good LLMs are at reverse engineering, I've been looking forward to being able to ditch MacOS and still have decent hardware soonish.
Taking back the ownership of our devices to run our software is the positive future LLMs are giving us. We need only to take action.
The same LLMs you have to rent from the very richest people on earth? I would be careful to call that freedom
You know you can run LLMs on your own computer, right?
If you can afford a computer that can run LLMs, sure.
"afford" in this case has to be compared to the cost of renting.
Of course. Someone who pays 150$ a year for Claude Pro would need years to get to a 4k machine.
150$/month, for year...
I've seen businesses that made the infrastructure to share a 100k machine. They needed it "on premise", but the setup did not look too hard and the results were outstanding compared to Claude Opus.
There's people running close to frontier model grade local AI on consumer and prosumer hardware, let's not act as if it was impossible to skip having to rent from OpenAI or Anthropic.
Have you seen hardware prices? They are so high that it's not even affordable for upper middle class in most of the world.
Omarchy is itself MIT licensed, with no restrictions that I'm aware of. If they later decide to make a paid version or something, the license wouldn't stop them, I think. But GPL would. All in all, I think it's a dumb move from Asahi because their GPL'd code will soon be eaten up LLMs that don't give a damn about licensing anyway.
I really don't get why people even start with permissive licenses if they have people they don't like to use their software or if they're just going to go "not like that" at the first real use.
It seems like many times it's open source but "just to look".
Because permissive licensing, and free software in general, rightfully comes with a certain cachet, and it turns out some people want the benefits of that reputation, but don’t actually believe in the freedom “to use, copy, modify, and distribute this software for any purpose.” Hence the treadmill from permissive licensing to copyleft, and then later from copyleft to “source available”…
“But software which OpenBSD uses and redistributes must be free to all (be they people or companies), for any purpose they wish to use it, including modification, use, peeing on, or even integration into baby mulching machines or atomic bombs to be dropped on Australia.”
—Theo de Raadt (2001)
That'a weird how you placed copyleft next to source available side of spectrum.
That's opposite because copyleft is ultimately about user freedom not developers freedom.
GPL wont help Asahi team for a bit.
> That's opposite because copyleft is ultimately about user freedom not developers freedom.
Yeah, the whole idea of GPL is to restrict developers from doing certain things, because it would be harmful to users
Yeah, the whole point of GPL is that people can fork it and improve it as long as they share the code back.
If Asahi is rejecting LLM produced code, then they will get the modified source but will be unable to use it because of their own rules.
They explain that they expected that MIT licensed code would encourage contributions from BSD OSes
I agree it was a poor choice of license and that GPL helps build a better community, I'm glad they mad ethe license choice
They say:
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin.
Permissive licences require attribution. It's okay to complain when people publish your permissively-licenced work without attribution, and hence without honouring the licence.
There are legal remedies to this that can be taken before changing the license entirely.
These remedies would still be needed to actively police misuse of the new license.
changing the licence is much cheaper than pursuing legal action, and sends an immediate and very public signal of disapproval. given that legal remedies would be required either way, and are extremely time-consuming and expensive, it makes sense to send the signal first
I don't get this complain. MIT does not forbid commercial use. So what's the fuss about?
From TFA:
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
It's one reason I've favored GPL of late, after many years using MIT. Not that many people these days would explicitly follow the terms of any license.
There are licenses that require attribution without being copyleft though. I'm not saying the Asahi developers are wrong to use copyleft, but it's weird to me to mention "without attribution" as a sticking point when there are fairly standard permissive licenses that require attribution like Apache and BSD.
MIT isn't copyleft and it does require attribution, it's basically the only requirement of it. I guess they just don't like that attribution means "put the license snippet deep in a secondary readme file hidden somewhere, along with 50 other licensees used in the project".
I think it's clear that they don't want people to use their code and close the door behind them. They also clearly want clearer attribution. I'm sorry but if those are things you desire, then deliberately choosing MIT because of a GPL hate-boner was plain stupid.
They are mad that vibecoders are building on top of their work, making real progress, then re-releasing it as "look it works" while containing a bunch of their work. Which is sort of fair but I don't get how that works in practice because in reality anyone can read the code which is open source and do whatever they want. Licensing won't stop them.
Sort of fair? Or actually fair because attribution for work is… fair?
Your credit is your first commit. After that people copy/paste patches back and forth as they will. It happened before AI. I've run patches in prod that were basically someone's random RAID controller patch posted to the Gentoo forum. If it works, it works.
Right, this seems deeply confused. If you care about how your code is used, don’t use MIT. Especially, if you care about monetization, probably use GPL or AGPL.
The whole reason why people are trying to replace them is because they have stubbornly refused to use AI, which has directly impeded progress in getting M3/M4/M5/M6 working. It's fine of course for them to continue pursuing their hobby of doing everything manually, and honestly it is insanely impressive.
But most people just want stuff to work, so it's odd to complain about it when others do manage to get it to work using AI to build on top of their prior permissively-licensed work.
Wanting stuff to work is a valid reason to not use AI for more sophisticated development projects. Having a computer that kernel panics once every three days is useless to me, even if it works for you.
Linux kernel itself is AI development friendly.
Some of us think that is a mistake.
Hmm very vague. "Some of us".
Do the massive amount of vulnerabilities not make you think twice?
At this point, I trust human code less than llm code. Design and architecture of the code still greatly benefit from a good human touch, but I'm not sure how long that will remain true.
https://www.phoronix.com/news/OpenAI-948-Linux-LPE-Bugs
So validate with an llm and humans. People committing code they haven't read - which will inevitability happen - is dangerous and hard to remotely verify.
In the end there is always a separate reviewer. Even if the producer is adversial and not even looking at the code then this process exists to catch it. This is not at all different from people attempting to get backdoors into Linux.
People using LLMs put up much larger changes, and much faster. Depending on the LLM used the quality is often much lower.
A lot of projects just stopped accepting any outside contributions rather than deal with the tidal wave of slop they were getting. Choosing to not accept LLM code is IMHO a better compromise.
LLMs will introduce new vulnerabilities. What Linux should have done is stop using non-standard C, or better, stop using C completely in favor of Rust, support code analysers.
And humans can introduce new vulnerabilities, too. Linux maintainers accept patches from random people over the internet. They already have to just the patch on its technical merit, not on who submitted it. For them I don't think it makes a difference (technically) if the patch was written with the help of an AI.
I believe that there may be valid reasons to refuse anything that was touched by an AI, but that's not about security. It's more about the non-technical problems coming from AI (centralising power, increasing inequalities, carbon footprint, ...).
Some people will always disagree with something. That's what's great about free software. There will probably be some Luddite anti AI Linux fork at some point.
That's not what they meant. It's AI friendly, not slop-friendly.
There's a reason no Omarchy patch of any kind has made it to the kernel list yet. I did personally look into some of their code and it is certified, hacked-togeter slop.
How do you define the difference between ai friendly and ai slop?
experience
My definition, FWIW.
AI-friendly: you can use the LLMs however you wish, but you must thoroughly understand every part of submitted code.
Slop-friendly: understanding anything is optional.
I also agree with that definition. But for example if I get an LLM to generate 95% of a program, but I understand and have read every line of code, is it still slop?
Some would say yes unfortunately.
Not the OP, but I wanted to share my 2 cents:
As an open source maintainer, when someone opens a PR, I don't know who they are. But I don't really care. I care about the PR: is it a "good" PR (there are multiple dimensions: is it understandable, does it add maintainable code, was it tested, etc) or not?
I think "AI-friendly" says "I don't care if it's written with an AI or not: I want good quality PRs".
I would think that "slop-friendly" is the kind of projects where the maintainers don't read the code. Maybe they try the feature (maybe not even?). I think the Omarchy guy was explaining somewhere that he does not read the code anymore, he just lets his AI write Rust. This is slop-friendly IMO.
If you are not AI-friendly, it means that you refuse to accept a good PR even if you cannot tell that an AI was involved. The mere fact that an AI was involved disqualifies the PR.
Are you one-shotting then reading/understanding? Then probably slop.
Are you iterating to help you understand, prompting to rename/restructure/document until you would have been able to easily understand without AI? Then probably not slop.
I'm pretty sure Asahi devs experiment with LLM tools, but not to Omarchy's extent. LLM ban is purely a filtration filter, so that they wouldn't be spending time overbearingly exhausting number of patches generated with AI, and not someone who cares about this than a shiny tool.
Everyone understands that LLM is a mean to an end, but it's clear that Omarchy got two confused and it's showing.
No, they’re pretty against AI in general.
Ai friendly patch:
* Author posts patch with a manageable number of lines so maintainer doesn't gets burned out reviewing it.
* Tries to outline on the body what the patch is about and his rationale for the design decisions on the code.
* Promptly answers any maintaners questions about the line changes.
* Discloses which AI model he used to assist on the patch and how was it used.
AI slop patch:
* Submits patch with huge lines of code without explanation.
* The body is written in a self-patronizing tone with superfluous formatting that helps nothing but annoys who is looking for the actual substance of patch
* When questioned by maintainer about some line changes either goes radio silent or reply with the usual AI tone of marketin salesman.
* rarely if ever discloses AI usage.
Exactly. As an example, read that Gravity Linux guy's responses here. I will never ever trust him. I will stay away from his code as far as possible. Asahi team: I have my trust in you.
A kernel panic is the least of your worries with this hardware. Asahi had to introduce a custom daemon to avoid frying speakers. Apple doesn't need to care about making the hardware interfaces safe if they control the entire stack, so they don't, and expose dangerous capabilities to the OS.
There was a fun period of time where Boot Camp still existed, but Windows could fry the speakers because it didn't have any of macOS's thermal protection. I think this was after the 2016 redesigns, as I had a mid-2015 with no such issue.
Does this generalize to basically anything outside of speakers though? I think a pretty easy solution is to not use AI for that specifically.
thermal management is the first thing that comes to mind. and battery management
You would hope both those systems have built in hw failsafes
A blown fuse is a failsafe but that’s not helpful for users.
If they removed the hardware failsafes from speakers to save a cent per device, I wouldn't be so sure.
In the case of speakers, though, what would a hardware failsafe even look like? It would probably be some sort of secondary chip; Apple's design is to just move that to the main processor, the same way they don't have separate SSD controllers.
Monitoring the CPU temperature feels very different to me. I'm pretty sure the CPU can't cook itself. This is probably documented somewhere so we don't need to speculate, but I can't research right now...
Good thing they only ever write correct code which consistently obeys every invariant in that entire stack ever. Excellent design, meet perfect engineering.
Has anyone blown out some component of their Apple Silicon Macs over the years, whether via traditional or AI-run RE?
Haven't encountered any reports about it yet.
It's amusing that people still default to "human-written code is solid and AI code sucks" after the enormous quantity of bugs that AIs are finding in human-written code.
Nobody has ever said human code is perfect. But it is empirically the case that code written by LLMs is on average lower quality than code written by humans.
That's not really true, and it's only even potentially true if we talk about the absolute cream of the crop developers. Most human written code is quite bad actually
> absolute cream of the crop developers
You’re reading their blog
This has not been true for 6 months. Update your priors.
Having a computer where half the features don't work isn't particularly great either...
Human slop is far worse than then AI slop.
So why those people using AI do not write the code from scratch without breaking the license and without vibe-copying existing code?
What is vibe-copying? The code was permissively licensed specifically so other people could do what they want with it. I'm so confused what people think permissive licensing is
No talent. no skills is why?
The code was MIT, I don't get why people write code use the most permissive licences and then get grumpy that folks use it as offered.
It's literally why GPL was invented.
There is no "stubbornly", the project is theirs and they do the f they want, their reasoning is sound. Permissive license does not mean that you cannot speak out about what people are doing with your software.
Do you think there would have been something working on mac if omarchy didn't start from the solid foundation of Asahi? Please don't make me laugh too hard.
> they have stubbornly refused to use AI
And by the looks of it, they're very intent on gatekeeping. Deep down I don't like AI either, it has taken away a lot of what I love about coding. But it's here to stay. What if we can use it to wrestle hardware control away from Apple and other authoritarian vendors? Hell yes, we should.
The likely problem with the Asahi team is that they (by not accepting AI) feel that control is being wrestled away from them, and instead of adapting they choose to swallow the poison pill of GPL.
We might not like it, coding as it was pre-LLM is done for. No one is going to come groveling at the feet of those who stubbornly insist on not using it.
> the poison pill of GPL
????
> the poison pill of GPL
not sure that concept makes any sense at all. the entirety of LLMs' capability to perform software design and development tasks, such as it is, comes from the massive resource of copyleft projects, including those using GPL, that the AI companies have ingested
Well, it's a more restrictive license than MIT. A lot of popular projects nowadays are under the MIT license. AI labs still don't give a damn about how the code their LLMs ingest is licensed anyway, and GPL'd code won't stop them one bit. If their code had been used in proprietary ways, then yes, GPL would've been the way to go. AFAIK Omarchy is MIT-licensed, which still requires preserving the copyright notice of other MIT-licensed code it's based on. So I'm not sure what damage has been done to Asahi itself.
You realize that Linux itself is GPL-licensed? How is it a poison pill precisely?
Linux itself has always been GPL licensed. It's the license switch that I have a problem with, not GPL itself. Switching from a permissive license to one less so, just so others would stop using it in their MIT-licensed projects.
> I'm surprised their work was ever MIT-licensed instead of GPL.
The intent was for their work to be usable on non-GPL distributions and OSs (including the BSDs).
The new licence is pretty restrictive — proprietary platforms will just regurgitate it with LLMs, but other open ecosystems will have to re-invent things from scratch.
> exploit our permissive licensing to hijack our work
You cannot exploit permissive licensing when its entire purpose is to encourage others to use it.
Of course you can, its like putting candy for Halloween on the expectation that one kid is not going to get the whole bowl at once.
The way you can exploit an open license is to grab the code, invest massive amounts of resources in it and close the platform. Embrace, extend, and extinguish.
Shameless plug for my own project:
https://gravitylinux.org
Spoiler: We have Vulkan 1.4 working, all of M4 and most of M5 working, are making significant progress towards full suspend to ram (no version of Asahi supports this) and this will all be shipped to users very soon!
What about ProMotion/120hz and HDR? Last I checked Asahi didn’t really make full use of the Macbook Display. But very interesting!
Works! The code is public at https://github.com/GravityLinux/linux
We also have vrr!
Cool, I'm glad someone is taking a more aggressive and LLM friendly approach to doing Linux on Mac. Too bad there's no M3 since that's what I have :P
Not sure if it would be useful for you, but I did get an M3 GPU driver kinda working (it boots KDE, runs Quake and GLXGears, etc) but isn't quite usable yet. But feel free to yank the code if it's helpful: https://github.com/SuperDisk/asahi-plus-slop
Omarchy works on M3 with GPU acceleration.
Omarchy is not something I'm comfortable trusting long term, and comes with a lot of cruft.
I'd be very interested in seeing the parent post's project get M3/M3 Pro support; have it working in my system with a patchset on top of Asahi but suspend is leaving a lot to be desired. Unlike Omarchy, the parent post's diff would also be much more reviewable due to it's more limited scope.
Omarchy has no chance upstreaming any of their work, the code is abysmal.
Asahi at least guarantees the quality of their stuff because they upstream everything, sooner or later. Gravity apparently also intends to be upstream-first.
> Cool, I'm glad someone is taking a more aggressive and LLM friendly approach to doing Linux on Mac
Same. I've been waiting 6 years to run no-compromises Linux on my MacBook. At this point, I don't care if someone uses LLM decompiled firmware ported to Linux.
Sounds extremely exciting!
If I could run linux on apple hardware without compromise, my life would meaningfully be improved.
It's not well understood how important it is that humans feel in control of their environment. I feel like it's a fundamental to human happiness. Software that takes that away contributes to the general malaise and hopelessness that many are feeling today. It's just water to them at this point. They don't even see it. But it's the primary means of interaction.
Making that a true option to millions of people could do a lot of good for the world.
It's important work.
> I feel like it's a fundamental to human happiness. Software that takes that away contributes to the general malaise and hopelessness that many are feeling today.
I would like to see a full-featured Linux on Apple hardware too, but I think over 99% of people don't feel anywhere near this strongly. Do you feel this strongly about being able to replace the engine in your car with one you made yourself? I don't, and I think most people view computers the same way.
> Do you feel this strongly about being able to replace the engine in your car with one you made yourself?
I am not one of those people, as I do not do cars, but I would hazard a guess that most car people over a certain age do in fact feel very strongly about what modern cars have become. The ones who grew up without knowing what ownership feels like less so, which reflects the exact same dynamic with younger generations who are growing up without ever knowing what it feels like to own the hardware they use.
As a person over a certain age I mostly like the fact that cars are so much more reliable than they were when I started driving.
I like the creature comforts on a long journey. I would have given my right arm for satnav when I was driving for work.
I’m much the same when it comes to technology. A phone that doesn’t break is much more valuable to me than a repairable device that fails.
I think it would be disingenuous not to point out that this was also true of new cars 20 years ago - which is much more pertinent to the point of the people you're replying to.
I am looking forward to swapping the engine in my MX5, if I ever manage to kill it.
Considering how easy it was to replace engines of most cars before the integration of electronics I'd say it used to be quite common. Just as most hardware vendors used to ship full specs for their components up until the 80s or even 90s.
So, yeah, yes, definitely. I wouldn't be working with computers on a daily basis if I had to use Windows.
I'm not gonna overly generalize this though. I don't think it'll be the same for most people but I tend to not care about what most people do 99% of the time anyways.
If it was as cheap and easy to install a new engine as it is for an OS, I imagine swapping engines would be even more popular than new OSs. (And I think far fewer would clamor to replace their OS if it cost $x,xxx for each OS swap and took many hours of manual labor - I.e. I think whether someone wants something can be heavily influenced by whether it is ~free vs cost $x,xxx, etc.)
If it was easy, someone would’ve done something. with Darwin or many the other abandon/obsolete OS’s from the past…
I wholeheartedly agree. If I think about it, I must have thought "god I wish this ran Linux" at least once every single time I opened my MacBook - likely for the last several years.
I have run Asahi and while I love the project, it's not ready for that type of use.
Recently, I got a new MacBook and repurposed my old M1 MacBook Pro for local inference. The unbelievable number of compromises and hacks you need to do to run daemonized always-on software on MacOS was another reminder.
I know it goes completely against Apple's business model, but I'd love to be an Apple customer if I could buy their hardware and use it with software of my own choice. Holding an iPhone or a MacBook feels great, they're all easily among the best consumer electronics ever as far as the physical part goes. But Apple's software is completely different from how I want to use my devices so I just don't end up considering them when buying.
Would love for Asahi or similar projects to be fully usable but it also doesn't seem very likely. Apple is arguably not hostile to other OSes on MacBooks, they're designed to freely boot non-Apple operating systems, but by not publishing any hardware manuals Apple ensures that an Asahi-style project is always going to be well behind, likely years behind, in hardware support.
Absolutely. In some ways it makes me wonder if the ability to boot unsigned OSes was just to avoid anti trust lawsuits.
Despite the ethical grey area of LLMs regarding stolen IP, I'm at the point where I'd cheer on a complete set of reverse engineered firmware appear.
Damn the consequences
Oh, they’re definitely hostile, by the way, Darwin is still open source, but no one’s bothered in 26 years market inertia? Just like no one’s bothered to take a license out with SoftBank to design something from the ground up (like those three ex Apple engineers) using Linux, the easiest and fastest way to get something working from the ground up.
Wouldn’t the time be better spent taking a license out? With SoftBank to design? Arm chips that work from the ground up running Linux instead of this time-wasting exercise? Apple is one year away from M7 two years away from M8 processors Asahi are just chasing smoke…
From the FAQ:
Gravity Linux vs Asahi Linux
Gravity Linux and Asahi Linux are two independent projects working towards the shared goal of a strong, open OS ecosystem on Apple Silicon devices. Gravity Linux was initially forked from Asahi Linux over differences in policy, namely regarding LLM use. We have tremendous respect for the Asahi Linux project and emphasize our shared goal of running Linux on Apple Silicon.
Does all of M4 include full GPU support on those chips? And what about the Apple Neural Engine stuff?
A shame that the community is split like this but happy to see progress regardless.
Sometimes it can be good to have two teams work independently towards a common goal. I'm sure they'll learn a lot from each other.
It really seems like the learning is flowing in one direction here.
Vulkan 1.4 on the GPU, D3D12, full AAA gaming support
I just did a search for gravity linux from the search at the bottom of the HN page. HN is a fickle place as it seems you get more traction by hijacking Asahi posts that when you make your own posts.
Are you upstreaming your changes?
> all of M4 and most of M5 working
This is confusing - your website shows M4 as Wip (i.e. pre-alpha) and only M4 mini as alpha. M5 is not listed at all.
Wip is described as "Bring-up in progress. Not usable yet." So is it "not usable" or "all of M4 (...) working"?
It's vibe-coded slop. Probably the reason why the Asahi team wanted to exclude these contributions in the first place.
Just remember that for however bad our code is (and I have no issues saying that most of our linux driver code is absolutely horrific, for now) it's better than Asahi code, because our code exists, and their code doesn't.
You're talking about code with the potential to damage hardware. So, no.
Exactly, and that attitude omg. Overall I'm glad I saw his messages - gravity linux is now on my ban list. Not going to touch it in a million years.
> Hasn't been shipped yet
Does it though?
"Code is a liability" - is one of the fundamental tenets of programming.
So, "it's better because our code exists, and their code doesn't" isn't the brag you think it is. Yes, a quick thrown together utility is better than having to manually do something. But that's not the same for an OS, or god forbid, a filesystem.
Especially since the only reason your code exists in the first place is that they did the hard work and your project forked.
I absolutely don't want to run "horrific code" as my OS. Why would anyone?
Especially in kernel mode. That interfaces with actual hardware.
If a kernel coded using AI assistance is unacceptable, then I have bad news about Linus' official policy on AI use in the Linux kernel.
No one criticized AI assisted work, but the authors admission that the code is shit.
It's certainly bad news, but there's not much any of us can do about it, can we?
Enshittification comes for everybody.
Isn't it great how free software means having a choice? I love it.
Yes, being able to buy a hacked-together microwave that can explode in your face and one that is tested and certified is amazing! Especially if you’re clueless about how they’re made and you just want a microwave.
> and one that is tested and certified
I must have missed that; could you share a link to the Linux kernel that's tested and certified as functional on M4 hardware?
Having your code accepted to Linux mainline is the certification in this paraphrase.
Mainline Linux doesn't run on the M4. So in your metaphor, it's more like being able to choose between a microwave of uncertain quality and not having a microwave.
Exactly my point. Some things you simply should not implement experimentally and tell the people it’s ok, just like Omarchy does to their.
> however bad our code is (and I have no issues saying that most of our linux driver code is absolutely horrific
It's the operating system, mate. People don't just use operating systems to take screenshots and then post on social media.
> it's better than Asahi code, because our code exists, and their code doesn't
This is an especially horrific comment if you are really related to the project and this also adds to the perspective on why Asahi team might have wanted to have no nothing to do with this project and related people and their thought process. Anyway, good luck.
Wow man, "horrific code" can be worse than no code. Bad code can damage hardware, destroy data, allow exploits, etc. all manner of bad things.
Unless your meaning is that you already have work-in-progress code that you are working on to make much better whereas they have not even started, I don't see how your point makes sense.
But even then it debatable whether making bad code good is better than creating good code from the start.
The website makes it obvious that it's in early alpha stages and is transparent about it:
> This release is intended for developers.
RE always ran a risk of everything you mentioned.
If this project had been started before the AI craze by some RE amateur you probably wouldn't be this dismissive.
IOW if it weren't slop then people wouldn't dismiss it as slop. You don't say!
People would still be nervous about the legal ramifications of accepting code from an Apple employee though.
Uhm, I don’t think this is the flex you think it is.
That's just... not how it works.
Even good code is a liability. Admittedly "horrific" code is a 10x or 50x liability. Sometimes you accept that kind of technical debt because "it's really important" — e.g. the company will go bankrupt, or whatever.
(Historically, btw, this is why proprietary corporate code is often so much worse than even fairly incompetent open-source code. Because it's just not really that important to ship horrific slop when nobody's income depends on it.)
The idea that shipping "horrific" code now is somehow better than not shipping it until it is fit for human consumption works in some narrow business life-cycle scenarios, but not in any kind of user-facing scenario, and certainly absolutely fucking not any kind of user-facing OS implementation scenario... (T_T)
You should be more respectful considering that the only part that works 100% of your fork come from Asahi.
Why didn't you start from scratch building something of higher quality? Lack of skill or only of tokens?
Are you saying "vibe-coded slop" because of the quality of the end result? Who cares if it gives you the features with good-enough performance?
Otherwise one can also say Asahi is compiler-generated slop compared to hand written assembler code.
I prefer reliable software for the Operating System. Like, I can tolerate slop for a game or a one-time-thing of converting a doc or something. But for running the base software (your OS), I think you need better standards.
>Who cares if it gives you the features with good-enough performance?
Some of us still care.
Could you elaborate on why?
You want longterm support, not a set of pit of tree patches for 7.3 kernel that will stop working or misbehave in 7.4 or 7.10, with no one understanding why.
Besides not wanting hallucinations on my OS level code?
Or not feeling good for the intellectual theft involved in the training of AI to begin with?
Or not caring for the de-skilling of of human brains who are delegating coding to AI, and increasingly even the architectural and decision part?
Or the training material being AI degradation that awaits most code being written by AI?
Or the baragge of slop PRs that overwhelms my favorite FOSS projects?
Or deeping of big AI dependency, in times of climate and economic crisis that it worsens, but that it also risks not having the luxury of it forever?
Or the ease with which any stupid direction can be followed via AI, and the resulting always changing projects, already a problem with short attention span project leads even for big projects like Gnome?
Or the lack of deeper understanding to control all this in future generations of programmers reduced to "AI operators"?
Only your first sentence has anything to do with why you would care in relation to Asahi Linux & Gravity Linux.
The rest is all just why you don't love AI generally.
(Not saying you're wrong; just that you expanded the topic substantially.)
>Only your first sentence has anything to do with why you would care in relation to Asahi Linux & Gravity Linux.
Well, none of those stop to matter or get a free pass just because it's Asahi Linux & Gravity Linux.
Especially since I wasn't just answering for Asahi Linux & Gravity Linux. I was answering the question I quoted in general:
"Who cares [if AI is used on a project] if it gives you the features with good-enough performance?"
>Are you saying "vibe-coded slop" because of the quality of the end result? Who cares if it gives you the features with good-enough performance?
There are a few significant reasons worth considering when getting down to the level of OS and hardware drivers. Long term maintainability is one. But another big one is that such software is capable of physically and permanently destroying the hardware it's operating on, with the textbook example for Macs being the speakers found in Macbooks. The only way to get the kind of performance they offer out of that physically light/thin package is to give them the power to destroy themselves if run at full force continuously but then run it from software so that they aren't run at full force continuously, but rather take advantage of the headroom that offers for very brief audio spikes whose overall integrated power still falls below safe limits. This all needs to be done from software since response of real world analog isn't instant. That enables much more range then if they were hardware capped at the maximum level they could run constantly, but it also means that if they're driven wrong, poof. The drivers have to never go to certain bad states, purely from software.
LLMs can't offer those kinds of guarantees, they don't really "know" how things work and then produce something deterministic off of that. It's even worse if, as-is the case with this sort of vibecoding effort, the LLM operator is in fact depending on the LLM for that part because they don't know how it really works either (vs someone who is just using it for boiler plate and knows exactly where the critical paths are).
Obviously everyone is welcome to roll the dice with their own hardware as they wish, and it may be ok. Or if using really old hardware one might just not care. And not all hardware is as sensitive as modern notebooks of course. Just be aware the analog world can be messy and you might not actually have the guardrails to screw up that you're used to in a higher level software environment.
>Otherwise one can also say Asahi is compiler-generated slop compared to hand written assembler code.
No, one cannot say that. Or at least not without looking really stupid.
If that issue with the speakers is documented, the LLMs of today would handle it correctly. That wasn’t true very recently. Also, assuming documentation exists is a big crutch that helps them more than us.
Hasn't been shipped yet, integration is being done now, we're shipping a new installer, from scratch graphics driver, and new kernel so there's a lot of moving parts we want to get right.
So it's vaporware.
All the code is already public
https://github.com/GravityLinux/linux https://github.com/niklassheth/mesa-wip
but as a shipped product you are currently right. That'll change when we ship.
As coincidence would have it, I checked your website just yesterday and M4 Max was listed as unsupported. Has that changed in the past 24 hours?
Don't bother. Just check his replies. I would not trust him with my hardware, software and data.
Any plans for supporting M3 Ultra? I understand it's a small percentage of the Apple Silicon Macs out there, but I'd love running Linux on my Mac Studio?
how's thunderbolt and external display support?
You do realize you’re basically called out directly in their post, right?
> Okay okay, Battlefront is a very simple Direct3D 9.0 game from 22 years ago.
Damn I feel old now. Lol.
This is very cool. I have a project that built on Asahi[1], and the better the hardware support, the wider the audience.
[1] https://github.com/hnarayanan/hello-guix
> It may surprise some to learn that WINE was originally made available under the MIT licence. The 90s and 2000s were a very different time in the Free Software ecosystem, and permissively licenced projects were viewed by tech companies as free real estate to pilfer and exploit. WINE was not immune to this, and many monetised proprietary forks of WINE were marketed and sold by various bad actors. The WINE development team grew increasingly concerned that these proprietary forks, while well within their rights to exist, would eventually cause so much fragmentation in the ecosystem that WINE itself would become unviable. In 2002, WINE was relicenced under the LGPL to prevent further proprietary forks from being created, and to encourage direct upstream contribution.
> When Asahi Linux was first started, it was with the intent that we develop a Free Software ecosystem for the Apple Silicon platform. To facilitate and encourage the particiation of our friends at the BSDs and other Free Software operating systems, we endeavoured to licence as much of our code as possible under the permissive MIT licence.
> Unfortunately, this is no longer viable. The malicious practices of the 90s and 2000s have reared their ugly heads once again, this time driven in large part by both the insatiable appetite of frontier AI labs for high-quality training data, and militant animosity toward the Free Software community.
> Some time ago, we were made aware of multiple attempts to exploit our permissive licensing to hijack our work, with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon. In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
> As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only. Those who do not learn from history are doomed to repeat it.
Hear, hear! Perfect clarity of thought. I love these people!
> In some extreme cases, in-development feature code was being lifted straight out of our personal forks and published without attribution or lazily rewritten with LLM assistance to obfuscate its origin. The MIT licence explicitly allows this, and has no carveouts to protect against bad faith actors.
Attribution is the one and only condition of the MIT license, aside from the disclaimer of warranty, so I don't see how it "explicitly allows" publishing without attribution.
> Unfortunately, this is no longer viable. The malicious practices of the 90s and 2000s have reared their ugly heads once again, this time driven in large part by both the insatiable appetite of frontier AI labs for high-quality training data, and militant animosity toward the Free Software community.
It makes no sense to me why you'd adopt MIT in the first place for its permissiveness, and then be surprised that people you don't like are using your code. It's like being a free speech supporter and being surprised that people say things you find abhorrent. That's the point. If you only support speech you agree with, you don't believe in free speech, and if you only support code being developed by people you agree with, then you don't believe in free software.
>It makes no sense to me why you'd adopt MIT in the first place for its permissiveness, and then be surprised that people you don't like are using your code
It makes no sense to me that this makes no sense to you. The Asahi Linux project kicked off in December 2020. About 2 years before the first ChatGPT (Nov 2022). It had a lot of ideals including working with the BSDs at the time, and so they surveyed the landscape and considered the downsides at the time vs the upsides and decided a BSD approach was the right tradeoff. Now, 6 years later, the landscape has changed pretty radically, so they've decided that the downsides have gone up and any value they're getting from the BSDs is not high enough, so they're changing course. This is not hard to follow.
>It's like being a free speech supporter and being surprised that people say things you find abhorrent
No, it's nothing like this at all. Though utterly tortured, convoluted and wrong analogies to free speech are common from people who don't understand it.
>If you only support speech you agree with, you don't believe in free speech, and if you only support code being developed by people you agree with, then you don't believe in free software.
It is apparent that you are overloading "support". Free speech is about opposing the use of violence (direct or indirect) to decide debates as a near-universal default. That's it. Social and economic consequences unbacked by violence are not opposed to free speech, they ARE free speech themselves. If I think your opinion is wrong and stupid, I may "defend to death your right to say it" but then not invite you over to dinner, donate you any money/time/resources to get your message out further, etc etc. That is my own personal free speech in response to yours. And you can criticize that in turn and not invite me and so on and so forth. Free speech preserves this unending dynamic.
By the same token, believing in "free software" is easily overloaded, and claiming that using the GPL means someone doesn't believe in free software like you just did is one of the most absurd and also ancient, tired bad faith attacks that's been around for decades. It diminishes you to make it.
> Free speech is about opposing the use of violence (direct or indirect) to decide debates as a near-universal default. That's it.
I don't believe that's it. Free speech in addition to being opposed to violence to silence individuals is also against censorship. If you want to publish a book of opinions society finds repulsive, you should not incur violence for saying so, but also you should not be prohibited from publishing your views.
Now of course there's a difference between you not inviting this person to dinner and the government prohibiting this person from publishing their speech, and I understand that. Even before ChatGPT, there still existed bad actors and especially with MIT licensed software, it was inevitable that people were going to take this very valuable code and use it in closed source projects.
Asahi has every right to examine this new landscape and make changes to their licensing as a result. But speaking personally, I release my software under an MIT license specifically because I don't care who uses it or how. I'm not criticizing Asahi so much as I'm pointing out to avoid the MIT license or other similar ones if you have any opinion whatsoever on how your code is going to be used in the future by your friends and enemies alike.
> The Asahi Linux project kicked off in December 2020. About 2 years before the first ChatGPT (Nov 2022). It had a lot of ideals including working with the BSDs at the time, and so they surveyed the landscape and considered the downsides at the time vs the upsides and decided a BSD approach was the right tradeoff. Now, 6 years later, the landscape has changed pretty radically
Asahi themselves cited WINE forks from the early 2000's as a case study in "bad actors" abusing permissive licenses. So AI really has nothing to do with this, they were well aware of the precedent in this area.
To me it's just another example of this weird hypocritical stance some free software people have. They love putting the code out there and letting it spread far and wide, but once someone _else_ starts making money building off it, "hey no fair!"
Permissive licences require attribution. They clearly state that there have been cases where their code was published without attribution.
The only thing you ask with a permissive licence is attribution. Why would it not make sense to you that they want it honoured?
Omitted attribution is of course a license violation, but their issues are not limited to just that case, and it's the other issues I'm addressing.
If they didn't even give attribution, why would we think they'd care about copyleft at all?
I still don't understand the motivation for Asahi and other similar efforts. My impression is that Apple fully supports Linux virtualization on Apple Silicon. Why isn't that sufficient? Just asking.
Because we want computers that work without macos
Why go outsite when we have VR? Yes, you answered this: it isn't sufficient.
Because you have limited system resources with virtualization.
From the perspective if the people working on it, I imagine it's immensely satisfying when they successfully reverse engineer something and get it working. It's not on the same scale or of the same complexity, but I feel that satisfaction when I finally understand something I've been trying to reverse engineer in a retro game.
From a user's perspective, I can only speak for myself, but I'm grateful for the Asahi Linux project's work. I have less than zero interest in ever running macOS again. It's user-hostile, constraining, and not a good experience. I prefer Debian with Plasma 6. But Apple makes high quality hardware, and they are one of the few choices for a fast ARM laptop with excellent battery life. I avoided getting one for a long time until Asahi got to a point where it ran pretty well on the M1 and M2 with relatively few caveats. Then I bought an M1 Air second-hand, and have used it exclusively with Linux ever since. I'm very happy with it. My only complaint is that I have to waste a big chunk of disk space to keep a macOS install on a separate partition that I can never remove and occasionally need to boot into to get firmware updates since I can't get them through fwupdmgr like I can on most of my other systems.
Motivation: OS/software freedom on Apple's hardware.
Same as why so many people run Linux natively on their x86 machines. It's just part of the preferred setup because it's nicer in so many ways and there's no annoying proprietary host os in the way.
Some people might want to get all the apple software off the apple hardware. For example the dossier that macOS builds on users and their family behind the scenes; currently on the front page of this website.[0]
Unfortunately it appears that Asahi has promised to not allow doing that[1] but perhaps it will enable someone else to make it possible.
[0] https://news.ycombinator.com/item?id=50403097
[1] https://asahilinux.org/docs/platform/security/#apples-unspok...
As just one example, GPU performance is much better on native Linux than in a VM.
Why do you want to climb it? Because it is there. The apple arm platform is there, and we're going to climb it.
Nice. One thing I like about Asahi is that they're relatively anti-LLM.
I'm glad that Asahi is switching from a cuck license to a reasonable copyleft license. Even though the circumstances that made this necessary are, of course, unfortunate.
To the people downvoting this comment: I get it, it's a vulgar choice of words. But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
Choosing GPL and similar licenses is the only way you can publish open source software without running this risk. As professional programmers, we must insist that people either pay us for our work or act as a equal collaborators.
And yes, there's a reckoning coming for AI companies who trained their models on GPL software.
> And yes, there's a reckoning coming for AI companies who trained their models on GPL software.
What would that look like in your opinion? Hard to fathom that the people with all the money and power could get into trouble over a licensing dispute.
> Hard to fathom that the people with all the money and power could get into trouble over a licensing dispute.
It's because of something called "Rule of Law"... Most free societies in the world have it.
If you find it difficult to fathom that rule of law applies to corporations, maybe it's time to start practicing fathoming it, or they will have won before the fight even started.
I’m aware of the rule of law, and enjoy living in a country where it (seamingly) still applies. I’m just questioning the viability (in practice, not in theory) of legal action in the US (where this would likely play out) given the economic hurdles to clear for the accusing party and the general chaos.
I mean, I hope you’re right…
You're referencing "rule of law" but... what exactly do you mean? Bartz v. Anthropic resulted in a ruling that training from copyrighted books is fine if you buy them. Doe v. GitHub dismissed DMCA claim, saying Copilot generated code doesn't infringe on open source licences.
So far the precedent and the law implies the open source licence won't matter for training and the result. And these decisions will now be brought up in the US in any new cases raised.
> But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
GPL doesn't prevent that at all. That's not why Asahi made that change. It's about contributing back to the project, not the money they want to see.
The inverse judo throw of GPL also turning free software advocates into copyright maximalists is quite unfortunate.
It doesn't change the matter of fact that it's insulting language, and I downvoted it either way.
Developers have their own good reasons to stick with permissive licenses. Either because they are completely selfless, or they want to help users who want to kick-start their own business selling software. Just virtuous reasons.
> But the fact of the matter is, if somebody else is getting paid based on your work, and you aren't being compensated your fair share, that's extractive and abusive.
That is not a fact at all. I choose to use permissive licenses because I am giving my work to the public out of charity. If someone uses it to turn a profit, and I never see a dime, I have not been exploited or abused in the least degree. It would certainly be virtuous for the people making a profit to give something back, but it isn't required. If I did require that, it wouldn't be much of a gift.
A reckoning for what? I've also read some GPL source, doesn't mean I'm not allowed to publish closed source software, as long as I'm not copying anything.
For the people wanting "Apple hardware quality with Linux", has anyone here been using the Framework Pro? I keep staring at it in my shopping cart but tell myself my regular Framework 13 is enough for the occasional use I give it.
Do you need better battery life? Otherwise the gains are minor.
This thread is full of folks saying, "Asahi can't keep up with Omarchy," when Omarchy Mac's own readme[0] states that it is, "Asahi Alarm + Omarchy." That's like bragging that someone sitting on Lebron's shoulders can dunk a higher rim than he can.
[0]https://github.com/omacom/omarchy-mac
But at the end of the day, like in your example, you have "Lebron" vs "Someone sitting on top of LeBron's shoulders" facing off against each other. Who are you putting your money on, to get the higher dunk?
The person who can still dunk if the the other one sits down
They're supposedly vibecoding support for hardware Asahi doesn't yet.
https://omarchy.org/news/2026/09/introducing-omarchy-m/
https://github.com/omacom/omarchy/graphs/contributors?select...
Intereting top two contributors.
Y'all should just let asahi cook. I trust them more to make actual working stable OS than meat proxies.
Yeah I don't trust anyone's judgement whose installation process is:
> Install from MacOS
> `curl https://alx.sh | sh`
https://asahilinux.org/
Of course not. Everyone knows you've got to do
for it to work. Duh.
Ah yeah, a .dmg installer that asks for sudo permissions is much safer /s
Why do people get hung op on this so much? You can just read the script before running it. Of course you won’t though, just like you don’t read the source of the .dmg installer.
It really depends. I'd say here it's warranted because idk how one would `brew install asahi-linux`.
If it's someone's fuckass vibecoded app though, just why.
I do it. I just found it hard to use it without the touch id. Hope that changes.
By the way how safe that OS and data in it are? Or even that partition is safeguarded by the safeties offered by Apple on a silicon mac?
That's a very complex question that entirely depends on who you're trying to defend from, more than I can explain in a single comment.
The short of it is that you can make it relatively secure, but it will never reach the same levels of security as macOS. The tight integration apple has allows them to do far more for security, for better and for worse.
Worse being apple putting the entire might of the platform's security into making sure the Chess app can't be uninstalled or tampered with.
I get you :)
And also that Safari can't be updated without a macOS update.
The other day my Mac running 26.7.1 updated to Safari 27.2, while still only offering 27.0.1 as an upgrade (no 26.7.2 or 27.2).
So many people that never wrote a line of code outside of school assignments in this thread.
People that did and have some opensource projects would never write some of the silly stuff I'm reading here. I don't expect permissive licenses to be popular for long.
honest question - how hard it is to use LLM and tools like REA to reverse engineer MacOS to figure out how its hardware works in terms of getting linux support on such hardware?
I thought it is already a solved problem? with large enough numbers of agents running in parallel reverse engineering, implementing kernel modules, testing in vm etc, the whole pipeline seems to be complete.
Did I miss something here?
Yes, read that blog entry to understand how different that hardware is from any other Linux-supported ARM stuff to see how much deep understanding you need to actually bring that support upstream to Linux or Mesa. Some(most?) stuff requires fundamental changes and no LLM is going to design and write it in an uostreamable way.
Debugging is not an issue, Asahi seems to have it figured out, especially with the virtualization. It's the design that takes time.
Meanwhile Omarchy, for example, just brute-forces everything, with ugly patches that work today but cannot be upstreamed as such, so they'll break tomorrow with new kernel release. I really expect them to start struggling a lot going forward with maintaining the support they claim. Regressions may actually kill their effort.
You must not work with AI tools very much. The brute force approach as a first/second pass is very standard when doing reverse engineering with AI. You get the functionality figured out and get the basics working and then you just iterate on that and massage it into what you need it to be. Omarchy has practically unlimited tokens and every other week there is a new model being released that is more captable than the previous that will just massage things into place with each iteration.
> Omarchy has practically unlimited tokens
And yet they have a total of zero upstreamed PRs, and haven't fixed enormous and glaring issues like PSCI control either.
It's almost as if the parent is correct, and Apple's ARM architecture isn't standardized enough for ordinary models to blindly reverse-engineer without documentation. If tokens were the only constraint on supporting Apple Silicon, we'd have reversed the entire SOC device tree years ago. Alas...
> And yet they have a total of zero upstreamed PRs, and haven't fixed enormous and glaring issues like PSCI control either.
Why should I care about if they upstream their patches or not? I have a working M3 macbook that runs Omarchy, why do I care that it's in the mainline Linux kernel? My GrapheneOS phone doesn't run mainstream Linux kernel, but it does the job that I need it to do.
Just read other people explaining this in this thread.
upstreamed PR means nothing in 2026. I couldn't care less when some maintainer are just so political, still remember the rust in kernel drama a couple of years ago? why great work have to get approvals from those people?
Sure, let's see how the hardware you enjoy in November behaves in March after they rebase the kernel to 7.5.
I work on reverse engineering tools and previously trained models, and no, this is not correct.
They prohibit RE of the binaries. What Asahi doing is observing macOS as a black box with input and outputs, and translating observed behavior back to Linux.
who are "they" here?
clean room RE is legal across the world. with AI, it is increasingly easier to do it right.
>As such, we made the difficult decision to begin the process of transitioning all of our repositories to the GNU General Public License version 2 or later. Going forward all Asahi Linux code will be licenced under strong copyleft licences only.
This made me laugh. For context: some people used LLMs to import code to get Omarchy to work on macs. So, these people are switching licenses so that they could stop competitor distros from being compatible.
So, before, Asahi people were happy with Linux working on macs; now they want THEIR linux working on macs (they are not changing to GPL out of the goodness of their hearts or that they believe in everything being open, remember, they are only doing it so that Omarchy, or similar, can't copy their work without having to be GPL themselves).
Meanwhile Omarchy is MIT; which means Omarchy is allowing Asahi (and everyone else) to copy their work, with the actual intent of getting Linux everywhere.
GPL protects against someone taking your code, adding proprietary extensions, and closing off access to the source code.
True, but knowing why they went GPL, it is hilarious. They had MIT and they were mad about another MIT-licensed project using their work.
I think the concerns are more long term than that.
True. Was Omarchy doing that? (I'm confused why changing the license matters, if both projects are already open source and releasing all the code to the public anyways)
They don't need to wait for the last step of a rug pull to occur, the trajectory is worrisome enough to take action before the shit hits the fan.
This really leaves a bad taste in my mouth too. It's unfortunately inline with their absurd AI policy. Instead of working for the good of the users they literally are trying to halt progress. It's pretty clear the omarchy people will run circles around asahi very soon. Asahi people think this too and are trying to limit them as much as they can with license changes.
Also, they say this:
> with a direct view to replacing the project and aggressively monetising Linux on Apple Silicon
which I hope everyone knows is completely dishonest. Nothing in Omarchy is about monetizing Linux; and monetizing Linux is and has been completely fine since its inception.
Asahi people just seem to be sour that someone else is joining the initiative but in a different project where AI use is encouraged to get to the same goal faster.
GPLv2 is a great license. Omarchy is literally taking their code without properly crediting and sloppily extending it. This is a good thing for everyone.
Big fan of Omarchy. Asahi could learn a few things from them :)
After seeing the blazing progress that Omarchy has been making with making Linux work on the latest macbook I'm planning on switching all 3 of the M1/M2 to Omarchy and planning on buying an M3 macbook air in the future. I appreciate the work that Asahi did to get the ball rolling, but they've just been left in the dust with their anti AI stance.
> but they've just been left in the dust with their anti AI stance.
How, exactly? AFAIK, Asahi officially supports just as many Macs as Omarchy does. The vast majority of Apple Silicon support in Omarchy comes from upstream Linux changes merged by Asahi, not any of the Omarchy devs.
The only feature I'm aware of Omarchy adding is the fingerprint reader driver, which is in such a poor state that it could never be merged upstream. Which, if that's the standard of quality Omarchy is holding themselves to, they're the ones getting left in the dust.
People outside Asahi are making M4/M5/M6 work. Pretty soon Omarchy will support M1-M6 while Asahi will only support M1-M3.
> they've just been left in the dust with their anti AI stance.
AI proponents always act like this is self-evidently a bad thing, but it isn't. When someone is going in the wrong direction, it is good for them to leave you behind. I believe that projects which embrace LLMs are going in a bad direction, and I'm glad to see more sensible maintainers get "left in the dust".
The patches to the Linux kernel have been about 50% made by AI in the past 6 months. The magic of open source is you can just fork it and go at your own pace and do whatever you want, and others can embrace whatever technology they want and then market adoption just speaks for whichever is superior.
How is making your project better — more feature complete, more secure, more performant and optimized — going in a bad direction? This is all platitudes and virtue signaling. Give me substantive reasons LLM “slop” is worse than human “slop”.
To get the ball rolling? All the stuff you listed, the hard part, is already available in asahi.
Are you a marketing bot? This is so insane and incorrect
Why doesn't the Asahi team just design their own silicon? With AI it is easier than ever. And no this is not a joke.
How to get it fabbed? That's even easier as there are lots of companies with $$$ who'd like to commoditize Apple.
That would be missing the point entirely though?
Asahi Linux is a project to provide an option to people who have for one reason or another got themselves an Apple Silicon Mac.
Building their own silicon would serve an entirely different goal.
Yet building their own silicon/with your in-house Linux distro would be the quickest and easiest way. Solving the need…
Why doesn't Apple just contribute to Asahi?
Isn’t Darwin open sourced? 26 years ago, no one’s picked up the gauntlet, what more help do you need?
How would designing a new chip commoditize Apple?
Three ex-Apple engineers did it by taking out a license with Arm Holdings/Softbank to form their own company and were successful enough to be bought out by Qualcomm 1.2 billion dollars, isn’t the Linux community big enough to support a processor from scratch, taking the same path? Taking that same path seems like a no-brainer if you have the skills.
Imagine deliberately buying a $2000+ computer that doesn't work with and is actively hostile towards your OS of choice, then embarking on a multi-year bitch-fest to shoehorn the OS onto it.
I struggle to wrap my mind around it. It's like buying an EV so you can convert it to work on gasoline.
I don't.
- MacBooks are still among the nicest if not the nicest laptops hardware-wise, so some people like to run Linux on it.
- People have older Macs that they like to repurpose. E.g., we still have an M1 Mini that was used by our daughter before she got handed down a MacBook from my wife. It's now collecting dust, but I might install Linux to use it as a headless server.
Also, this is Hacker News.
> - MacBooks are still among the nicest if not the nicest laptops hardware-wise, so some people like to run Linux on it.
By the time said macbook is fully supported, it is in sorry state. Judging by what I found when looking for second hand laptops, macbooks don't age well.
> Judging by what I found when looking for second hand laptops, macbooks don't age well
Are you dumpster diving for your MacBooks, perchance? I buy all of mine second hand, they are all still mint by the 5 year mark, and I have several at the 10 year mark that all I've had to replace is the battery.
Huh? Apple has many problems. but one of them isn’t the resale value of their hardware. Or maybe I should say, if you buy Mac hardware, you don’t have a problem selling it on, not like a used PC the only thing worth selling on in a PC is the graphics card (Nvidia) and the memory.
I bought an M1 Ultra 64G off eBay in March with some 30 battery cycles, it looks new. This thing is a beast compared to most consumer hardware even today with its 800GB/s memory, runs small models well. When I get better HW for LLMs it's going to be a great Linux machine for a decade, well past the point Apple stops supporting M1 entirely.
Isn't that true for anything? Some people will sell 5yo hw in almost mint condition, whole others treat their hw like ass and it's garbage after two years already.
The problem with used Macs is that they're still really expensive, and if you want to only run them with Apple software, the support duration is limited (especially true for phones, even more for Android ones). You can pick up used Thinkpads or Dells in great condition for very cheap after a few years.
>Imagine deliberately buying a $2000+ computer that doesn't work with and is actively hostile towards your OS of choice
That's how Linux has been for PCs also for most of its lifetime. Only in the recent 5-10 years things have kind of improved, slightly.
Macbook hardware is superb. I understand why people are interested in replacing the (in some parts) lackluster operating system Apple forces into it.
Not at all. I've used Linux on Thinkpads starting in roughly 2006 (switching away from MacOS), and they always worked flawlessly.
One major FU was with the 5.x kernel when suddenly it no longer supported my external monitor, this with Intel graphics (and one by one all distros upgraded to that kernel). Not sure what happened there and why, but I had to do using Linux for a while because of that.
imagine being so obtuse you go around judging how other people choose to have fun and spend their own money and "pretend you don't get it".
Dino fuel generator hooked up to Tesla EV: https://youtube.com/shorts/46bRaGP4uqM
For me the definition of actively hostile would require a locked down device that needs an exploit/jailbreak to install and a cat&mouse patch/exploit cycle - but it seems that Apple intentionally designed this to be allowed - what is giving you the “actively hostile” impression?
Apple hardware is so much better than the alternatives that people will dump thousands of hours into getting it to work with their preferred software
I used Windows and Linux all my life, but got my first MacBook Air M this year. There is literally zero competition to it. It's an incredibly powerful and cheap yet fanless device. Also, it has uniform memory, which is awesome.
I wish I could install windows or linux on it because MacOS kind of sucks, but the device itself is amazing and reason enough to switch away from PCs.
> because MacOS kind of sucks
It didn't used to, but now does.
> but the device itself is amazing and reason enough
By continuing to reward Apple, they lose any incentive to improve macOS and it will continue to get objectively worse because people keep giving them money. Why would they fix things if you keep buying it?
> By continuing to reward Apple
Strange argument. Their hardware is without competition, that's the reason I use it and also warrants rewarding them with my money. Why should I reward Linux for (mostly) only working on hardware that sucks in comparison?
I used MacOS back in the Panther and Tiger days (10.4), then switched away, and started using Macs at work starting with maybe Monterey.
I was very surprised that almost nothing had changed in that long time! They got virtual desktops (finally), and widgets changed from an expose-like effect to only becoming visible at the right side of the screen, and tbh that's about it. All other "features" seem to be changes to Apple's built-in apps.
I think the OS still sucks about as much as in 2004, it's very much a matter of taste.
How do you think Linux gets to run on proprietary undocumented hardware generally?
I got an M2 macbook air with 16GB ram and 1TB SSD for $700 and install Asahi Arch Linux and am planning on migrating it to Omarchy now that it supports M chips. It's my favorite laptop out of the Dell XPS, Precisions, and thinkpads that I own. The fact I can get 7-9 hours of battery life, have everything feels snappy while I'm doing Dev work and have a very crisp display all at the same time puts the Intel/AMD chips to shame. Supposedly Intel is actually giving Apple a run for its money this year after 6 years of being left in the dust, so maybe I'll get something Intel next year.
To go with your analogy, if I can buy an electric car and replace the engine in it with something that does what I want and gives me 95% of the usable features of the locked down engine that it ships with, why wouldn't I do that? Especially when the rest of the cars on the market are getting 100 miles to a charge when my Apple M2 is getting 300?
Apple is surprisingly non-hostile toward these projects. They even fix bugs that only affect Linux: https://developer.apple.com/documentation/macos-release-note...
> I struggle to wrap my mind around it. It's like buying an EV so you can convert it to work on gasoline.
What about buying a really nice gasoline car and convert it to an EV?
Could the projects bring in stuff from omarchy? Omarchy is a toy, but they did slop the osx thing, so maybe some ideas can be ported over?
Asahi is against LLMs so they won't.
Not sure what will happen when people will directly contribute to the Linux kernel with LLM patches though for these devices, I guess they will have to live with it.
Asahi is patches on top of the linux kernel. If you independently get your drivers in to the kernel which does allow LLMs, the drivers will end up on Asahi too.
I believe that’s what parent was saying.
Yes indeed that's what I meant.
I saw they managed to apply their no-llm policy on top of their upstream mesa driver but I'm not sure how long that will hold if the forks are getting too popular
Sure, but a human could read the code and look for ideas/parts and write the actual code better than an AI can. Omarchy is 99% hype, but the fact that its working on osx is telling me that it is indeed a goal you can accomplish. I rather not have my OS be a slopfest, so thats why asahi is way better than omarchy ever can be.
I didn't know about their LLM stance, but I feel like Asahi are shooting themselves in the foot. Giving up the advantage to toxic teams like Omarchy is strategically wrong. They could use llms for reverse engineering and drafting and as you said still rewrite the final versyof the code.
At the end of the day the decision is on them. If they want to be slower in development than clank-using competitors, so be it.
Would really prefer they didn’t lose to DHH of all people tho
Maybe to them it’s not a race, and more a journey? Or to put it differently, why the rush? It’s not like they’re trying to corner the “Linux on Apple Silicon” market, last I checked?
And to add to that "journey" story, they probably also understand that writing the code is only one part of it, and they probably want to remain in a position to maintain it
It's not about the journey if you don't want (certain) other people to use it as they have so far, but rather about status, and they're afraid of losing that status.
Asahi is working on Macs too. Omarchy just took their existing drivers and wrapped them in fascism. They're also supposedly vibecoding support for newer Macs (which Asahi is working on too, but without slop, and Apple making changes to the platform every year doesn't help).
In the end, absolutely it still takes human software skills at the end of the trail. Pressing a button and walking away isn’t going to work.
I’ve just build the kernel from Omarchy on my fedora asahi remix and got thunderbolt and Touch ID working. You do need to patch m1n1 with their changes as well. But I’ve been able to finally ditch an extra usb hub and hdmi cable.
In the PR’s that were merged in that kernel I also see they are actively working towards m4 and m5 as well. Must be hard for Asahi to keep up this way.
It's hard because they write things with intention to upstream it. Omarchy patches simply cannot be, because they go absolutely against the established way of doing things. No one consciously designs anything in their code, it's all LLM output.
I personally vetted that TouchID code you mentioned and it's an absolute hellhole, with everything done in-kernel and no userspace separation what so. That code is also a security nightmare.
sorry to put a dent in your fantasy world, but omarchy is awesome for day-to-day work, great upgrade on top of arch which is also awesome.
Its basically useless gui thing. Nothing i cant have in another distro. What im intested in is the osx port. Apple hardware is unbeatable, but the OS has been going downhill since jobs passes. This is why true linux on mac is my dream setup. Asahi is the closest thing so far.
apple hw is unbeatable indeed. but the os was always for normies, a walled garden.
regarding 'the useless gui': it was what actually was missing in in the linux world.
i used kde before, which is fine, but it is too bloated and kwin_wayland is unusually heavy on the machine: it maters, especially for mobile devices like laptops.
of course if you use macos and apple hw, you don't care about it , since the stance. i don't run (yet) omarchy on my servers. but there are a ton of other, especially developer / ai friendly, defaults that would make one tinker for days with other distros. as i said i had my own "dot files", but a community built distro with the right defaults is much better energy management than everyone on their own.
> i used kde before,
If you're comparing Omarchy to KDE, I don't think you understand what either actually is. You can use Hyprland/Quickshell in other systems.
> but a community built distro
Omarchy is also not a community built distro.
you can, of course. but here it is very nicely packaged.
The omarchy glazing by noobs is exhausting for those of us who have used Linux for 20+ years.
I have used Linux for 21 years, on and off on the desktop, and all the time on the server. In the last 10 years I've had basically not time to catch up with what's new in terms of desktop environments and other tools (e.g. neovim, terminal emulators), much less have the time to tinker with any of them.
Omarchy (first it was Omakub) changed that by serving it all up to me in less than five minutes. There are, perhaps, other similar projects, but this one got to me first. For me, it's an improvement. Someone else's defaults aren't a constraint as long as I find them useful. I only wish M4 Pro was supported, because macOS (even though I'm still running 15.x) has gone downhill.
It's, of course, less useful if you have 20 years' worth of dotfiles and custom workflows that you aren't keen to replace and I fully respect that.
There's nothing to catch up on. It's a kernel. It's systemd. And then pick a desktop manager. Every distro installs in 5 minutes. Because on Linux you literally just copy/paste. It's just a sloppy fork of Arch.
You are tech dreaming any group working on this? Should be looking to do their own arm processor. by taking out a license with SoftBank and designing something from the ground up.
How do you justify calling it a toy when it does certain quite fundamental things a lot better?
Because TWM ricing isn't new. Nor is shipping pre-riced distros, or using an Arch base.
Do you think TWM ricing is a fundamental feature of a distro?
It's just about the only headlining feature of Omarchy.