* before replying snarkily that Android is Linux, please take a long walk off a short pier, thanks
We have waydroid for that.
> You can't have a bank account on a Linux phone* because they won't let you, but you can on Android including on Graphene.
Unless of course it uses those stupid integrity apis to block anything that isn't stock.
https://privsec.dev/posts/android/banking-applications-compa...
What security problems does waydroid have that a VM wouldn't?
Waydroid uses an outdated fork of LineageOS running with namespaces and a compatibility layer on top of a much less private and secure base OS without similar kernel or userspace security protections. Running up-to-date AOSP in a virtual machine would at least be able to preserve the internal Android privacy and security model for apps to protect apps from each other and the OS from apps. It would also contain the overall OS within it too. Using the much less private and secure OS as the host OS with full access is quite backwards from a privacy and security perspective but would be a huge improvement.
AOSP is much more private and secure than traditional desktop Linux distributions. Moving to that software stack is inherently going to be moving much further away from competing with the privacy and security of iOS. The direction taken by GrapheneOS is to start from AOSP and greatly improve the privacy and security it provides to compete with and exceed the industry standard privacy and security provided by iPhones. That requires more than only software. Hardware and firmware security are very important too. Software security also increasingly depends on hardware-based security features such as hardware memory tagging, hardware control flow integrity protections, hardware-based virtualization and much more.
Keeping user data safe from access via encryption also depends on hardware security features for the vast majority of users not using a very strong passphrase. People take it for granted that they're going to have secure data via disk encryption with a random 6 digit PIN but that's not the case without a good secure element and OS integration with it. The approach used by desktop operating systems with TPMs is awful and makes security worse in a lot of ways rather than better. It's not at all the same thing, similarly to how what the desktop world calls secure boot is not a serious or complete implementation of it and doesn't provide nearly any useful security properties to end users unlike iOS or AOSP.
Interesting. Could you elaborate? I always thought that Secure Boot is reasonably secure (provided you can set it up in the first place).
Secure boot was widely used as a term prior to the UEFI usage of the term including many much more meaningful implementations. The term verified boot refers to the same concept but avoids it being confused with solely verifying a late stage bootloader and OS kernel from UEFI firmware.
Moving to a far less private and secure desktop software stack is going in the opposite direction from GrapheneOS. GrapheneOS doesn't exist to simply provide an alternative to mainstream operating systems but rather to offer much better privacy and security. We wouldn't be doing that if we were forking a desktop Linux environment and doing similar work for it. It would be nowhere close to the privacy and security of simply using an iPhone. That's a major part of why GrapheneOS is based on AOSP rather than it solely being about compatibility.
Desktop distributions are incredibly far behind on privacy/security and lack any clear path to achieving the same things. Every year, Android makes backwards incompatible privacy and security improvements as part of a new target SDK version. Android retains compatibility with legacy apps, but apps distributed through the Play Store (and other app stores to an extent) are required to move to the new target API level within around a year. This results in apps being forced to conform to a gradually improving privacy and security model. There's no such thing for desktop Linux apps but rather apps choose how much they want to participate in nascent sandboxing efforts.
GrapheneOS has near perfect compatibility apps from the Play Store via our sandboxed Google Play compatibility layer with the exception of banking and government apps. 90% of banking apps currently work on GrapheneOS because it greatly succeeds all of their security requirements and is only wrongly banned by a subset of those apps. These apps are gradually adding more anti-tampering and attestation checks for the hardware and OS, so maintaining compatibility has required us to gradually add more functionality working around it. We've also had to actively convince apps to stop banning non-Google-certified operating systems or to permit GrapheneOS and other secure options alongside doing it. A growing number of apps are choosing to stop banning using GrapheneOS due to pressure from our expanding userbase.
Waydroid uses namespaces and a compatibility layer to run an outdated fork of LineageOS. It's far from providing full functionality and compatibility with apps and nowhere close to the compatibility provided by GrapheneOS. More importantly from our perspective, Waydroid disables most of Android's standard privacy and security model through not supporting SELinux and other core parts of the security model. Android doesn't simply use SELinux as an additional layer of security with targeted policies but rather it's deeply integrated into the OS. It uses very strong whole system policies to implement the app sandbox, drastically reduced kernel attack surface and a lot more. Namespaces for the overall userspace environment are not a replacement for the app sandbox, kernel protections, verified boot and the rest of the security model.
1. Android has existing apps.
2. A subset of existing apps will refuse to run on various shades of non-stock/unofficial system.
I pointed out that 1. the existing apps largely do run on other Linux systems via waydroid, and that 2. apps refusing to run on non-stock systems are likely to refuse GOS as well.
If I cut down your page of text to the relevant points, I believe it amounts to:
1. GOS has better app compatibility than waydroid.
2. GOS is supported by some apps that would refuse other non-stock systems.
3. GOS is more secure/private. (This is unrelated and should at most have been a reply to the root comment.)
On 1. compatibility, yes you're probably ahead, though I have to question how much since waydroid is also 90% AOSP. Somewhat similarly on 2. yes GOS has gotten some buy-in, but if 90% is good enough, then again you shouldn't be holding up compatibility as your advantage over waydroid. Basically, either users want 100% app compat - in which case GOS is out - or they don't, in which case waydroid is likely to be an option. Point 3. is irrelevant to the conversation, and you probably don't want to start an argument about general features since GOS is missing vitally important features compared to other Android ROMs let alone "normal" Linux systems (how long has GOS been without full backup functionality now?).
> We've also had to actively convince apps to stop banning non-Google-certified operating systems or to permit GrapheneOS and other secure options alongside doing it.
What other secure options are those?
Android has a massive app ecosystem including the largest open source mobile app ecosystem. Those apps are nearly entirely available for GrapheneOS. Very few apps are unavailable for GrapheneOS. It's nearly entirely banking and government apps, but as we said 90% of those work on GrapheneOS.
AOSP and GrapheneOS have hardware-based virtualization usable to run desktop Linux software. It even provides opt-in GPU acceleration with gfxstream and will likely have native GPU virtualization support similar to NVIDIA's paid vGPU in the near future. AOSP and GrapheneOS have a very usable desktop mode which will be getting much better. Android is replacing ChromeOS and was already heavily improved for desktop use as part of being included within ChromeOS. We have that app ecosystem available too.
> the existing apps largely do run on other Linux systems via waydroid
A large portion of Android apps do not work via Waydroid. It has drastically lower compatibility than GrapheneOS and there are fundamental issues with how it approaches it.
The approach they take also severely harms privacy and security through throwing away most of the privacy and security model used by Android. That would not be the case for running AOSP in a hardware accelerated virtual machine, which is what we would recommend doing. It would still have greatly reduced compatibility due to apps using hardware APIs such as the hardware keystores and many apps actively trying to stop themselves being run in a virtual machine or other weird environment. It would work a lot better than using namespaces but would require modern hardware with virtualization acceleration.
> apps refusing to run on non-stock systems are likely to refuse GOS as well
A very tiny subset of overall Android apps entirely ban using a non-Google-certified OS. Most only have checks for the security model being intact and anti-tampering code interfering with running the apps in an atypical environment. We run into these issues with our privacy and security hardening features so we've done a lot of work to maintain compatibility.
> GOS is supported by some apps that would refuse other non-stock systems.
GrapheneOS exceeds all of the security and attestation expectations of these apps. They have no reason to ban using it and we make sure to avoid that being the case. It's only a tiny number of apps banning using GrapheneOS in the first place and a growing number of those are beginning to permit it since they don't actually have a reason to ban it. They do have a reason to ban an OS not providing the security model and attestation functionality they want. We disagree with attestation being used this way, but they can do that while supporting GrapheneOS and a growing number of these apps are doing so.
> yes you're probably ahead
GrapheneOS provides drastically better compatibility.
> waydroid is also 90% AOSP
It's a fork of outdated LineageOS with a compatibility layer to make it run as part of a non-AOSP host OS. It has SELinux disabled which means the app sandbox and other protections aren't intact along with many other major differences. Apps can see these differences and many apps ban it because they can see the security model isn't intact or detect what's happening as a form of tampering.
In practice, GrapheneOS only gets banned by services enforcing the Play Integrity API device or strong integrity level without permitting GrapheneOS via hardware attestation. A growing number of apps are permitting it since they don't have any real reason to ban it. It's easy for them to stop endless negative reviews and customer support complaints from GrapheneOS users by simply implementing Android hardware attestation with Google's open source library for it and permitting GrapheneOS. They can also simply delete the code banning using non-Google-certified operating systems which is what we'd prefer over them hard-wiring permitting GrapheneOS and specific other alternatives.
> yes GOS has gotten some buy-in, but if 90% is good enough
That was only about banking apps. Over 99.99% of overall Android apps work on GrapheneOS. The vast majority of apps are not banking or government apps.
> compatibility as your advantage
It has an immense compatibility advantage along with far better privacy, security, usability and battery life.
> Basically, either users want 100% app compat - in which case GOS is out - or they don't, in which case waydroid is likely to be an option
Most users have 100% compatibility since most people don't use the around 10% of banks banning GrapheneOS. Our userbase is also growing large enough that apps cannot ban using it without it being a huge hassle for them. They have no actual reason to ban it, so that's why a growing number of apps are removing the ban on using it. They're usually not willing to stop using the Play Integrity API but once this is on their radar due to complaints they often become willing to implement using the Android hardware attestation API to permit GrapheneOS and other alternatives meeting their requirements. It's not a full solution since it won't allow people's self-signed builds or forks of GrapheneOS but it's progress and it enables the apps to easily permit more operating systems in the same way. They can just add more to their list. We would prefer the Play Integrity API being banned by regulators but this is good enough for now.
> since GOS is missing vitally important features compared to other Android ROMs let alone "normal" Linux systems
GrapheneOS has similar functionality to mainstream Android smartphones. We highly prioritize providing missing functionality not available through apps.
GrapheneOS is an operating system, not read-only memory firmware. There's a boot ROM which loads the main boot firmware from the SSD, verifies it and transfers control to it which then does the same with the OS. GrapheneOS is simply a regular OS installed on an SSD. Verified boot protects it from modification but it's not in read-only storage or anything like that. We don't misuse the term ROM to refer to it and prefer if others avoid it too.
> how long has GOS been without full backup functionality now?
GrapheneOS has much better backup support than the stock Pixel OS and most Google Mobile Services Android devices.
GrapheneOS has a built-in encrypted backup system. It backs up the same data as Google's device-to-device transfer for moving between Android devices. Every app targeting Android 12 or later gets backed up since allowBackup="false" was redefined to only disable cloud backups. Every Play Store app has been required to target Android 12 or later since under a year after the release of Android 12. The issue of apps opting out of backups has been in the past for years.
Backing up more of the system settings and other system data is planned and requires gradually expanding what gets converted into a portable format transferable between devices and OS versions.
> What other secure options are those?
It's up to these companies to define their security requirements. GrapheneOS is far more secure than anything permitted by the Play Integrity API device or strong integrity levels. It supports hardware-based attestation and we publish a signed JSON object providing our verified boot key fingerprints for use with it. They can easily use that and permit GrapheneOS. Other operating systems preserving the Android security model can do the same. Once they allow GrapheneOS, it's easy for them to allow more in the same way. It would be nicer if they stopped using the Play Integrity API and left OS security up to users but most of these apps are not open to doing that. A small portion of apps adopting the Play Integrity API didn't understand what it does and are willing to stop.
> Android has a massive app ecosystem including the largest open source mobile app ecosystem. Those apps are nearly entirely available for GrapheneOS. Very few apps are unavailable for GrapheneOS. It's nearly entirely banking and government apps, but as we said 90% of those work on GrapheneOS.
> AOSP and GrapheneOS have hardware-based virtualization usable to run desktop Linux software. It even provides opt-in GPU acceleration with gfxstream and will likely have native GPU virtualization support similar to NVIDIA's paid vGPU in the near future. AOSP and GrapheneOS have a very usable desktop mode which will be getting much better. Android is replacing ChromeOS and was already heavily improved for desktop use as part of being included within ChromeOS. We have that app ecosystem available too.
There was really no need to write 2 paragraphs to agree with me.
>> the existing apps largely do run on other Linux systems via waydroid
> A large portion of Android apps do not work via Waydroid. It has drastically lower compatibility than GrapheneOS and there are fundamental issues with how it approaches it.
What is a "large portion"? It's worked fine for me but I suppose if you have hard data that might be meaningful. Could you describe the "fundamental issues"? My impression is that they just run the whole Android userspace in a container, which seems like a reasonable approach.
> The approach they take also severely harms privacy and security through throwing away most of the privacy and security model used by Android. That would not be the case for running AOSP in a hardware accelerated virtual machine, which is what we would recommend doing. It would still have greatly reduced compatibility due to apps using hardware APIs such as the hardware keystores and many apps actively trying to stop themselves being run in a virtual machine or other weird environment. It would work a lot better than using namespaces but would require modern hardware with virtualization acceleration.
Again, I know security/privacy is your talking point, but this isn't actually the argument at hand.
>> apps refusing to run on non-stock systems are likely to refuse GOS as well
> A very tiny subset of overall Android apps entirely ban using a non-Google-certified OS. Most only have checks for the security model being intact and anti-tampering code interfering with running the apps in an atypical environment. We run into these issues with our privacy and security hardening features so we've done a lot of work to maintain compatibility.
So, yes.
>> GOS is supported by some apps that would refuse other non-stock systems.
> GrapheneOS exceeds all of the security and attestation expectations of these apps. They have no reason to ban using it and we make sure to avoid that being the case. It's only a tiny number of apps banning using GrapheneOS in the first place and a growing number of those are beginning to permit it since they don't actually have a reason to ban it. They do have a reason to ban an OS not providing the security model and attestation functionality they want. We disagree with attestation being used this way, but they can do that while supporting GrapheneOS and a growing number of these apps are doing so.
Which is a lot of words to agree that yes, many apps only allow stock and therefor block you just like any other OS.
>> yes you're probably ahead
> GrapheneOS provides drastically better compatibility.
I'd like evidence for "drastically", but again we mostly agree.
>> waydroid is also 90% AOSP
> It's a fork of outdated LineageOS with a compatibility layer to make it run as part of a non-AOSP host OS. It has SELinux disabled which means the app sandbox and other protections aren't intact along with many other major differences. Apps can see these differences and many apps ban it because they can see the security model isn't intact or detect what's happening as a form of tampering.
Right. So it's 90% AOSP. Old AOSP with worse security properties, but for compat purposes that's not important.
> In practice, GrapheneOS only gets banned by services enforcing the Play Integrity API device or strong integrity level without permitting GrapheneOS via hardware attestation. A growing number of apps are permitting it since they don't have any real reason to ban it. It's easy for them to stop endless negative reviews and customer support complaints from GrapheneOS users by simply implementing Android hardware attestation with Google's open source library for it and permitting GrapheneOS. They can also simply delete the code banning using non-Google-certified operating systems which is what we'd prefer over them hard-wiring permitting GrapheneOS and specific other alternatives.
You and I have different impressions of how apps approach attestatoin.
>> yes GOS has gotten some buy-in, but if 90% is good enough
> That was only about banking apps. Over 99.99% of overall Android apps work on GrapheneOS. The vast majority of apps are not banking or government apps.
And most apps work on waydroid. Of course the apps that go out of their way to block non-stock are the pain point.
>> compatibility as your advantage
> It has an immense compatibility advantage along with far better privacy, security, usability and battery life.
(You've carved out 3 words in a way that doesn't make sense alone; this is part of either the previous or next bit)
>> Basically, either users want 100% app compat - in which case GOS is out - or they don't, in which case waydroid is likely to be an option
> Most users have 100% compatibility since most people don't use the around 10% of banks banning GrapheneOS. Our userbase is also growing large enough that apps cannot ban using it without it being a huge hassle for them. They have no actual reason to ban it, so that's why a growing number of apps are removing the ban on using it. They're usually not willing to stop using the Play Integrity API but once this is on their radar due to complaints they often become willing to implement using the Android hardware attestation API to permit GrapheneOS and other alternatives meeting their requirements. It's not a full solution since it won't allow people's self-signed builds or forks of GrapheneOS but it's progress and it enables the apps to easily permit more operating systems in the same way. They can just add more to their list. We would prefer the Play Integrity API being banned by regulators but this is good enough for now.
90% of the time, it works 100% of the time. Anyways, you aren't 100% compatible, so my point stands.
>> since GOS is missing vitally important features compared to other Android ROMs let alone "normal" Linux systems
> GrapheneOS has similar functionality to mainstream Android smartphones. We highly prioritize providing missing functionality not available through apps.
I'm not comparing to stock, and you refuse to give apps root access so the shortcomings that they can fix are limited. (Yes, I know your threat model demands that the stupid helpless users can never be allowed to control their device. That's actually defensible if security+privacy is your #1 goal, but it does undermine the features that your OS can provide.)
> GrapheneOS is an operating system, not read-only memory firmware. There's a boot ROM which loads the main boot firmware from the SSD, verifies it and transfers control to it which then does the same with the OS. GrapheneOS is simply a regular OS installed on an SSD. Verified boot protects it from modification but it's not in read-only storage or anything like that. We don't misuse the term ROM to refer to it and prefer if others avoid it too.
I love pedantry as much as the next guy, but ROM has not meant read-only memory in like 20 years.
>> how long has GOS been without full backup functionality now?
> GrapheneOS has much better backup support than the stock Pixel OS and most Google Mobile Services Android devices.
> GrapheneOS has a built-in encrypted backup system. It backs up the same data as Google's device-to-device transfer for moving between Android devices. Every app targeting Android 12 or later gets backed up since allowBackup="false" was redefined to only disable cloud backups. Every Play Store app has been required to target Android 12 or later since under a year after the release of Android 12. The issue of apps opting out of backups has been in the past for years.
> Backing up more of the system settings and other system data is planned and requires gradually expanding what gets converted into a portable format transferable between devices and OS versions.
So, you're trailing, say, Debian, and have done for years, and have at most a plan for someday being able to compete.
>> What other secure options are those?
> It's up to these companies to define their security requirements. GrapheneOS is far more secure than anything permitted by the Play Integrity API device or strong integrity levels. It supports hardware-based attestation and we publish a signed JSON object providing our verified boot key fingerprints for use with it. They can easily use that and permit GrapheneOS. Other operating systems preserving the Android security model can do the same. Once they allow GrapheneOS, it's easy for them to allow more in the same way. It would be nicer if they stopped using the Play Integrity API and left OS security up to users but most of these apps are not open to doing that. A small portion of apps adopting the Play Integrity API didn't understand what it does and are willing to stop.
So when you wrote "GrapheneOS and other secure options", you just meant GOS because nothing else matches your definition of security. Again, I don't exactly mind that, but it's disingenuous to gesture at other options that don't exist.
20% of apps from the Play Store not working as opposed to less than 1/10000.
> Could you describe the "fundamental issues"?
Many Android apps use functionality such as the hardware keystore APIs and other hardware APIs.
> My impression is that they just run the whole Android userspace in a container, which seems like a reasonable approach.
Using namespaces in the way they do is not a reasonable approach since they're running it with SELinux disabled which means most of the privacy and security model is disabled. It also isn't set up to run that way and requires very problematic hacks. It also ends up not being possible to run more than one instance of it due to those hacks.
> Again, I know security/privacy is your talking point, but this isn't actually the argument at hand.
It is not a 'talking point' and is very relevant in all cases. You may not care about it but most people reading this thread do.
> So, yes. > > Which is a lot of words to agree that yes, many apps only allow stock and therefor block you just like any other OS.
GrapheneOS has better app compatibility than plenty of Google certified devices. In practice, the only issues are with apps going out of the way to ban other operating systems and GrapheneOS does better than anything else. Apps are increasingly permitting it since it exceeds all their requirements and it's just a little bit of extra work for them to not solely depend on Google's API.
> I'd like evidence for "drastically", but again we mostly agree.
It's easy to check and other people have done so.
> Right. So it's 90% AOSP. Old AOSP with worse security properties, but for compat purposes that's not important.
No, you're wrong. It's very important for compatibility. Apps can and do detect large parts of the security model being disabled. A far larger portion of banking and government apps won't run on it. Having 90% compatibility with those is certainly better with 30% or less.
> You and I have different impressions of how apps approach attestatoin.
You don't have the experience and data we do. We have a large number of users and reports from them. We also have conversations ongoing with numerous banking apps about it and have seen multiple apps permit GrapheneOS. Revolut is currently talking to us, although the app works on GrapheneOS since we worked around how it was being banned in January 2025. We're also talking to Square (Cash App, etc.) and others. It is a problem which can be solved and these companies are largely not against solving it or even actively want to solve it, it just isn't something they've gotten to yet. The GrapheneOS userbase growing has begun to make this problem solve itself.
> And most apps work on waydroid. Of course the apps that go out of their way to block non-stock are the pain point.
No, a large portion of Android apps do not work with it because they use APIs unavailable or broken with it. Many banking and similar apps check that the security model is intact which it clearly isn't so they ban it but they would happily run on a non-stock OS which appears to have it intact.
> You've carved out 3 words in a way that doesn't make sense alone
What we wrote makes sense.
> 90% of the time, it works 100% of the time. Anyways, you aren't 100% compatible, so my point stands.
90% compatibility with banking apps is far better than less than 30%. You're trying to make it seem black and white when it isn't. Google certified Android devices don't truly have 100% app compatibility in the first place. Apps have plenty of required features, minimum Android version requirements, requirements for specific models (some apps only work on Pixels, Samsung flagships, etc.) and a lot more. Some low-end devices have broken support for Camera2 and other APIs despite the CTS. GPU drivers are buggy and games are buggy so there are compatibility issues. Perfect compatibility never exists. Nearly every non-banking app working fine and 90% of the special case of banking apps working fine is very good compatibility.
> I'm not comparing to stock
Most people are using OEM Android forks or iOS.
> and you refuse to give apps root access so the shortcomings that they can fix are limited
A userdebug build of GrapheneOS has root access. It's not suitable for production builds used by the vast majority of users and isn't in those. It's also incredibly niche and not very useful. What you're actually referring to is using a rootkit framework providing app accessible root access where a huge portion of the OS has root access. You're not talking about the user having root but rather a framework for apps built with disregard and even disdain for security being granted root via a dialog which a user cannot truly ever revoke or undo the consequences of granting beyond a reinstall of the whole OS.
> I know your threat model demands that the stupid helpless users can never be allowed to control their device
The app accessible root frameworks you're talking about drastically reduce the privacy and security of the OS for all users including ones never using it to grant root access to an app. Granting root access to apps built with such disregard for following security best practices is also not somehow unproblematic because you think they're trustworthy since they're open source.
> but it does undermine the features that your OS can provide
No, it means we build features with proper implementations following security best practices instead of doing it with half-baked hacks via a massive portion of the OS having full root access and granting that to apps.
> I love pedantry as much as the next guy, but ROM has not meant read-only memory in like 20 years.
It means read-only memory in embedded, information security and in the context of GrapheneOS which is an overlap of both. GrapheneOS works on security including boot security where it's very relevant. When someone says ROM recovery in the context of GrapheneOS, that means the boot ROM recovery mode loaded when firmware on the SSD can't be loaded. It does not mean the OS recovery mode. It's not correct terminology to refer to GrapheneOS. It propagates unnecessary misconceptions and confusion.
> So, you're trailing, say, Debian, and have done for years, and have at most a plan for someday being able to compete.
No, it doesn't provide a similar backup system portable across devices and major OS versions.
> So when you wrote "GrapheneOS and other secure options", you just meant GOS because nothing else matches your definition of security. Again, I don't exactly mind that, but it's disingenuous to gesture at other options that don't exist.
No, our response said the apps should set a security standard and enforce it fairly. They should not enforce Google certification but rather define the standard security expectations and permit anything meeting them. GrapheneOS is much more secure than any Google certified OS, so it would meet actual security standards they set. Whether other alternate operating systems meet those standards depends on what the app developers set as the requirements. Many Google certified operating systems won't meet reasonable security standards and would need to be banned if the enforcement is fair.
Unfortunately, the EU is currently developing their age-verification-app, and it mandates hardware attestation[1], and it seems that their reference implement those requirements using Play Integrity[2].
[1]: https://news.ycombinator.com/item?id=49148128 [2]: https://github.com/eu-digital-identity-wallet/av-doc-technic...
Regulators should force Google to stop further closing up Android and to permit alternate operating systems to pass the Play Integrity API device and strong integrity levels. It should not require certification by Google and complying with their arbitrary requirements based around their business model. There should be another path to obtaining certification without Google's involvement where Google has to respect it and permit those devices and operating systems to pass. Most important is stopping them from closing things down more against the original terms they provided Android and gained market share with it. Next most important is a reasonable path to alternatives passing the Play Integrity API based on security standards which do not block updates while waiting for certification.
We don't think it's realistic to stop apps adopting attestation but people should try regardless. A single country with a large market banning would make a huge difference.
https://grapheneos.org/articles/attestation-compatibility-gu... is our guide for app developers on permitting GrapheneOS via hardware attestation. Other operating systems can publish signed keys in a similar way to enable easily supporting them once this is implemented. An organization certifying operating systems could collect these and sign an overall list. It would also be possible to support a broader hardware ecosystem with alternate roots of trust by providing a signed list of those too. This is not the future we want but it is the future we believe we can obtain through a lot of pressure. We would greatly prefer apps not making an allowlist of operating systems. People should be able to make their own GrapheneOS build and use the same apps.
It's not a very good guide, frankly. It's basically a wall of text without clear instructions on what to do, besides "look at the examples Google gave". And even looking at Google's examples, there are no examples or explanations that tell me how to download the list of allowed signatures, and what should I do with the Json schema.
You've given me hope for the future!
Soon, there won't even be an alternative flow. There are a lot of places where there already isn't.
However at that point you'll likely pay for the privilege.
The "security module" they require you to install on your computer. In the past, when browsers had plugins, this was a browser plugin; nowadays, it's an always-on service (running as root) which exposes a local HTTP server which the bank site connects to to validate your computer. For an example from a major bank in this country (the same "security module" is used by several banks in this country), https://seg.bb.com.br/home.html is the diagnostic page for that "security module" (the FAQ page there has links to the installers).
Its app is really nice and native, though. That, along with a decent interest rate, is why I use it.
That would be inconvenient if this was 1985.
Or in other words: of course you can have mobile bankid without a smartphone, just use a tablet computer ;)
Consider switching your bank to one not forcing you into American megacorps.
Well, I can buy a train ticket without it, so I could still leave.
I wonder why there would be such a difference in policy between countries, not only in government but across the private sector? This doesn't make any sense to me. If anything, I'd expect Sweden to have a more sensible, left wing attitude than Oz.
For eg. There's no browser based alternative to make UPI payments that i know of.
It's not just for clients with no smart phones. It's also clients who activate developer mode on their smartphone etc... clients who use VPN. Clients who run AdBlocker. Clients who use Firefox.
There needs to be a severe push back against this.
Things have gotten so crazy that the homeless here are walking around with QR codes printed out when begging for money. I'm guessing they're a network of people who share the same smartphone/payment method..
The relevant regulations are Commission Delegated Regulation (EU) 2018/389 and the earlier Directive (EU) 2015/2366.
While these laws are deliberately vague when it comes to specific technologies, they do require at least two independent factors from different categories, such as knowledge (password) and possession (phone).
That, in and of itself, wouldn't be a problem. The way most banks implement it, however, is by giving you two choices:
1. You use their mobile app (which likely requires device attestation and Google Play Services, so won't work on a plain LineageOS install)
2. You use their CardTAN device, which is extremely inconvenient to always carry around.
Sure, we nerds might argue they should just let us use our Yubikeys or regular old TOTP, but pretty much no bank implements that. (Why? Your guess is as good as mine.)
Personally, I had to buy a second stock Samsung phone just for banking apps. And yes, there are still alternatives (only very very few though), but no, none of them are convenient, for various unrelated reasons.
I wouldn't be surprised if this continues to spread to the US too, under the sneaky disguise of "security".
In reality, most of time it's just a matter of choosing the right bank wherever you are. I'm writing this on a GNU/Linux phone that does my banking just fine. I'm from the EU.
crying in EU :(
This is a social issue - not a technical one.
Source: my bank which recently 'upgraded' a browser version to a glorified SPA which even renders as a vertical oriented app on a landscape 4K monitor.
A pure linux non-android phone would be great however you wouldn't have access to properly working apps and would not be able to participate in modern society.
Anyone getting BMW digital keys to work on Graphine OS?
Somehow I do.
It seems not all of them, and that things will only get worse if recent news comes true.
That said, I think wasi containers support for a mobile OS would be a game changer.
(^ Here's a startup idea if you're looking out for one. I for one throw out all my devices from the balcony to get this)
I haven't figured out how to make them talk with each other though. I imagine that's the place where I'd notice this lack of support.
Almost none support strong dedicated HW authenticators or second factors. Not even as an option to those who care.
Anyway it's always possible to just reverse their web api and use it directly. 2FA that consists of copying some code from SMS is no barrier, especially not on the Linux phone that you fully control.
I should make a chip called Super Security Sauce Silicon, and make cards with them, and market them to phone vendors so their customers won't need the card, and market them to banks so their customers will be secure. It performs Dual_EC_DRBG with my keys.
If any of the Motorola devices have GOS as a pre-installed option, now the companies don't have the excuse that the device is modified.
I'm guessing the companies will continue to be difficult, but it'll be amusing to watch, at least.
Why can't the prescription be sent to email or collected on a website? I don't see why an "app," especially one signed by Google is necessary for something like this. It wouldn't surprise me if the app requested location or sensor access in order to provide a "personalized experience."
They surely must have gotten feedback from me and others because the next update it worked again. But I and probably others where already a lost customer.
Diminishingly few apps do not work, and it’s down to them.
https://privsec.dev/posts/android/banking-applications-compa...
Long term in the future, surely the open source community forks AOSP itself and maintains it. The worst option would be giving up and trying to focus on Linux (desktop) imo.
You're misinformed about security preview patches. It has nothing to do with Google being hostile. Additionally, our security preview releases have regularly included patches for components not specific to Android such as wpa_supplicant prior to those being publicly disclosed by the upstream projects.
Yes, for some definition of "heavyly" that you chose. Yet no single megacorp can decide single-handedly on the future of Linux or do this: https://news.ycombinator.com/item?id=49364745
> AOSP has a massive amount of non-Google companies and individuals working on operating systems based on it.
Can you define "massive" in, e.g., percentage of lines of code? This looks unsubstantiated to me.
The parent has a good point and shouldn't be flagged.
based on is the important part here. Yes, AOSP would survive without Google. But as long as Google is in control of AOSP, it controls the direction in which AOSP and all downstream projects are headed. Because of breaking changes in AOSP, there is a cost in maintaining downstream changes. So downstream projects cannot veer off too far, short of a hard fork.
> You're misinformed about security preview patches. It has nothing to do with Google being hostile.
Does it not? From the 2025092500 GrapheneOS release notes: "We strongly disagree with broadly distributing patches to OEMs 3-4 months before the official publication date. It further delays getting patches to users and sophisticated attackers will have no issue getting the patches from one of many people at Android OEMs with early access."
Sounds pretty hostile to me.
Leaving phones behind on old incompatible OS versions 2 times in 3 years and switching app frameworks 3 times in 4 years does not a good app developer experience make.
Piled on top of that, Google became actively hostile to 3rd party developers building support for YouTube (and Gmail and Gmaps, but those had workarounds / alternatives).
The advantage being, we can manage packages using a regular Linux distro
Drivers doesn't just mean the kernel. It's the user space binary blobs and services that need to talk to the kernel to enable the hardware.
Other than that there's waydroid, alien dalvik etc.. that run another Android instance in a container.
The thing is a lot of Android applications use safety net/other methods to make sure they only run on. "Approved"/stock hardware.
Google thought about this, don't worry. They learned their lesson after CyanogenMod tried to compete by offering an alternative. Non-Google Android are now dead except in China.
> The advantage being, we can manage packages using a regular Linux distro
It's much easier, more plausible, and more logical to sandbox a regular Linux package-managed distro inside of Android, which numerous solutions exist for.
People will tell you that chromium is "more up to date" as if Google wasn't the one setting the standards, making it impossible for anyone else to be similarly complete. Seems like we have a very similar problem here...
In principle I agree about a Linux phone, but the gaps are much greater. I am also sympathetic to the GOS team's arguments that sandboxing on Android is better, and important on a device that allows control of essentially my whole life (2FA apps etc.)
We provided a much more detailed reply at https://news.ycombinator.com/item?id=49364220.
There is still a philosophical reason for supporting a version of Linux for mobile devices that is not dependent on Google, but is community-driven, for users with different priorities.
That is not really relevant to the GOS project though, and the fact that GOS is probably the easiest way to get a Play Services-free mobile device kind of drags it into discussions that it doesn't necessarily belong in.
Again, thanks for your work, and I appreciate your team's focus on your goal of the best security possible.
in China - there's no google apps available on their 'android' versions.
their platforms are already performant and fluid - so people should build on that.
Linux crowd can not even agree on compositor, and if systemd or sudo is a good idea.
I think "linux smartphone" would end up looking very similar -- some UI on top of a kernel with a very non-LSB userspace; therefore the concept is really not all that compelling
But take that far enough and you get Stallman'd into a corner where someone points out that's great, but no phone for you.
This is actually a feature.
Especially in this political climate it’s insane to recommend people switch to a phone OS that can be pwned via hardware or software means in virtually zero time versus something like GOS + a Pixel or an iPhone.
The hardware and software security of Android/iOS/iPadOS/GOS devices are leagues ahead of Linux servers in all aspects and it’s not even close (but they also serve different purposes so to some extent I’m not sure how 1:1 this comparison is). However, this discussion focuses on desktop style Linux distros which is what these phones are running which have inadequate security. Desktop Linux is inferior security-wise to a server environment where each process/app is running as it’s own user etc. Desktop-style Linux phones also have virtually zero hardware security versus the alternatives.
It doesn’t matter if a user knows what they are doing. For example installing so-called “trusted” apps means they’re only trustworthy at that moment in time and not forward-looking. All it takes is one supply chain or runtime security exploit. Software security is also extremely poor with a lack of comprehensive sandboxing and so on, I would just read the grapheneos accounts replies on here since they’re writing entire novels on this.
I mean even GrapheneOS shows how relevant this is with Samuel Tunick, on the front page today. If he’d had a desktop Linux phone that was seized from him while in AFU, sure it would be “private” but there would be virtually nothing stopping officials from running through his entire phone since so many of these desktop Linux distros rely on poor security elements if any on top of FDE with LUKS so your entire device is basically exposed and unencrypted after the first unlock.
This whole comment was borderline stream of consciousness so I apologize if anything is unclear here.
claude code makes stuff like that super accessible to do in your spare time. another example is installing debian on a synology 918+, there's no way i would've had the grit to do that without ai. it's open season for any gadget that's got a debug uart port.
Yes, you can, although it may require tinkering and compromises: https://itsfoss.com/whatsapp-linux-desktop/ and https://news.ycombinator.com/item?id=49363246
Also if we use atomic distros with flatpaks and whatnot that mimic Android security the end user basically ends up having to essentially use Termux (but busybox or something similar) on their Linux phone as well
GrapheneOS makes our own major improvements to the permission model, but we cannot enforce apps adopting it and need to design it all to be backwards compatible. We do that for Contact Scopes, Storage Scopes, our Sensors toggle and other features but not everything can be done that way. For our exploit protections, we have toggles to work around apps with memory corruption bugs or bad practices such as dynamic code loading. Android can enforce apps improving in a way we can't do. We're starting from a platform with a mandatory app sandbox and relatively modern exploit protections though. Android also does support using hardware memory tagging or HWASan to test apps even though many developers aren't using those to clean up their memory corruption bugs.
You don't think, say, isolating apps has anything to do with it?
As far as I understand, the Linux phone enthusiast's POV is that we should start seeing the phone as another general purpose computer that can be set as per one's needs and can be suitably hardened as and when the need arises.
I wonder what environment the GrapheneOS devs use for development and other general computing needs, and if such an environment could be emulated on a phone (that is assuming, of course, that their development environment is way more hardened than current desktop Linux).
Traditional desktop Linux distributions have atrocious privacy and security. The security record is very poor. The security record does speak for itself in that it has been a disaster. iOS and AOSP have massively improved upon the legacy Unix security model. A traditional desktop OS cannot properly protect users from applications, remote attacks or physical attacks such as extracting data from an After First Unlock state device. It's not only the software that's very lacking but also the hardware and firmware for the Windows and desktop Linux ecosystem. macOS has a made a lot of progress for the hardware, firmware, hardware-based security within the OS and a gradual move towards a mandatory app sandbox and other protections which have not happened in the Windows or desktop Linux world.
AOSP has an increasingly usable desktop mode and supports running traditional desktop Linux within hardware accelerated virtual machines. It's not fully ready as a desktop replacement yet but it's getting there. Android will be shipped on many laptops in the future as a replacement for ChromeOS. The desktop mode is going to be the main way it functions on those so it's going to get much better. GrapheneOS has all of this functionality. Many people are trying out the latest Android 17 desktop mode on GrapheneOS and were already using the earlier experimental mode. Major improvements to that are coming. Many people are quite happy with this even if you don't want it.
No, it's not a bold claim. It's a strong consensus among people working on privacy and security engineering or research.
> Care to elaborate?
We did in our response to you at https://news.ycombinator.com/item?id=49366321 and posted more elsewhere. Others have posted far more details about it in articles and papers.
> I think you are somewhat conflating the needs of desktop/laptop/mobile users.
No, we aren't. Desktops need far better privacy and security than they have today. That does not require giving up functionality. Development can be done in contained hardware accelerated virtual machines specific to projects. It doesn't require running an OS where anything goes and software doesn't run in a sandbox on the host. QubesOS could have a far more secure OS as the one people interact with despite running far less secure operating systems in many of the guests. It would not prevent people getting work done. It can also be done with far higher usability than is currently provided.
> Hardware security tied to unlocking modes makes little sense on desktops.
Verified boot and hardware support for disk encryption is very valuable to a desktop and laptop.
> I can lose my phone, or possibly my laptop, but my servers and workstation are at home, guarded by a dog and a shotgun.
Servers and workstations being in a trusted location doesn't mean the security of those against physical attacks doesn't matter. It also doesn't address remote attacks.
> Disk encryption already covers a realistic threat model such as a break-in.
You do not have comparable disk encryption. You have it in a form that only truly works when the device hasn't had the encryption passphrase entered since boot. It's wide open to data extraction via a bunch of debugging interfaces. There's incredibly poor security against physical attacks at both a hardware and OS level.
> I don't expect protection against apps, and don't want to pay the performance tax that comes with "enhanced" memory allocators and memory-safe languages/runtimes.
The performance cost of exploit protections can be determined in a granular way. GrapheneOS has a toggle to use the standard allocator for any apps. Memory safe languages do not imply inherent overhead and can be faster for some uses. Rust is very high performance and even Java can be higher performance for certain uses where there are lots of short lived allocations.
> The irony is that Android is practically designed to run commercial malware. The sandbox you have in mind does very little in practice.
Android's sandbox provides enormous benefits and the permission model gets significantly better ever year. Android has heavily focused on improving a lot of what you're talking about itself upstream and GrapheneOS improves it much more.
Android can force backwards incompatible changes on the whole app ecosystem to improve privacy and security which it does with each new major target SDK version for the major yearly releases. They limit how much they do this in each major release and aren't as focused on privacy and security as us so we wish they did more, but no other mainstream platform has made such substantial improvements to the privacy and security model for apps. Android has improved drastically more than iOS since it had a much worse starting point and had to catch up. It has caught up in most areas and surpassed iOS in some but still needs improvements in others. GrapheneOS is heavily focused on those areas where it's still behind iOS and has done a lot to improve it.
> Users consent to sharing files, contacts, locations, photos, when asked to because they want to use some app, and all of that is sent straight to Facebook, Google, or others.
GrapheneOS has Contact Scopes and Storage Scopes to choose which contacts and files/media are shared with the app seeing it as the permissions it requested being granted. This is an extension of Android's standard approach built on the infrastructure it provides for this. It's already heavily moving in this direction and has increasingly forced apps to use a case-by-case consent approach. Location, Microphone and Camera access in Android can be granted on a one-time basis. It heavily pushes apps to using the system file picker, media picker, contact picker, etc. and store policy is increasingly used as a deterrent against requesting bulk access. GrapheneOS does much better and it's able to do much better because of the infrastructure Android provides.
> Android apps can also communicate in some obscure and unexpected ways to discover each other and share information behind the user's back
Communication better apps is possible within profiles where both apps want to engage in it. Users do have the ability to prevent apps seeing each other and communicating via profiles (Private Spaces, work profiles and secondary users). GrapheneOS will be providing finer-grained nested profiles beyond a single Private Space per user or a single work profile along with it being more convenient.
Apps getting access to user data in the first place is what's most relevant rather than whether they can communicate regardless. They could also send it to a service rather than another app. Android is focused on preventing apps getting access to sensitive data including data from other apps or the user's profile data (files, media, contacts, etc.) which GrapheneOS greatly enhances. We do have things like a Network toggle but it's not intended as a substitute for the main approach of avoiding apps getting access to sensitive data. Android is currently phasing in a standard local network access permission for apps will provide one of the main benefits of our Network toggle.
Hardware-based security features including hardware memory tagging, a high quality secure element providing high quality interfaces for improving security, proper verified boot and more are definitely important too, but they cannot simply be bolted onto an OS. The OS needs to be built around being able to take advantage of these features.
Moving to a far less private and secure desktop software stack is going in the opposite direction from GrapheneOS. GrapheneOS doesn't exist to simply provide an alternative to mainstream operating systems but rather to offer much better privacy and security. We wouldn't be doing that if we were forking a desktop Linux environment and doing similar work for it. It would be nowhere close to the privacy and security of simply using an iPhone. That's a major part of why GrapheneOS is based on AOSP rather than it solely being about compatibility.
Desktop distributions are incredibly far behind on privacy/security and lack any clear path to achieving the same things. Every year, Android makes backwards incompatible privacy and security improvements as part of a new target SDK version. Android retains compatibility with legacy apps, but apps distributed through the Play Store (and other app stores to an extent) are required to move to the new target API level within around a year. This results in apps being forced to conform to a gradually improving privacy and security model. There's no such thing for desktop Linux apps but rather apps choose how much they want to participate in nascent sandboxing efforts.
GrapheneOS has near perfect compatibility apps from the Play Store via our sandboxed Google Play compatibility layer with the exception of banking and government apps. 90% of banking apps currently work on GrapheneOS because it greatly succeeds all of their security requirements and is only wrongly banned by a subset of those apps. These apps are gradually adding more anti-tampering and attestation checks for the hardware and OS, so maintaining compatibility has required us to gradually add more functionality working around it. We've also had to actively convince apps to stop banning non-Google-certified operating systems or to permit GrapheneOS and other secure options alongside doing it. A growing number of apps are choosing to stop banning using GrapheneOS due to pressure from our expanding userbase.
Android is a large Linux operating system family. It's the mainstream form of Linux on personal computers. Android users are Linux users. For privacy and security, using a monolithic kernel written in C is definitely not a good thing. Doing much better than we are today partly requires moving away from so heavily depending on the Linux kernel for security. Android does a lot of Linux kernel hardening with attack surface reduction and exploit protections which are improved by GrapheneOS, but it's not enough. The massive torrent of severe vulnerabilities being discovered in the Linux kernel is going to get worse before it gets better and will remain a problem. Adopting hardware-based virtualization for isolation of apps and OS components including drivers is an important part of our roadmap.
You could be dividing up your projects into highly sandboxed environments on Android too. It has support for running multiple hardware accelerated virtual machines running desktop Linux and it wouldn't be that hard to support creating those with NixOS and other distributions instead of only the standard the Debian images provided by Android. That will happen as it gets more mature.
See the explanation at https://news.ycombinator.com/item?id=49364220 for why desktop Linux is nowhere close to AOSP for privacy and security. If you want more details, there's a lot of deeper coverage.
That is fine but obviously we are talking about threat model in comparison to each other.
>Google Play Services, which are needed
They are "needed" if you want to use other closed source apps that rely on play service feature. If you don't want/use closed source apps there is no reason to use Google play services at all, that's why they are optional.
>running with full access to all my data
GrapheneOS sandboxes play services specifically to run with user permission instead of system and can be used in a completely different isolated profile for your other private user data.
NixOS by itself does nothing to increase security outside of the supply chain. There is for example no default application sandboxing, Mandatory access control or hardened memory allocation (software or hardware). Just to name the most basic security feature.
It's almost like Android has put millions of expert dev hours into making it the most used OS in the world. Like GNU+linux on laptops only works the way it does because of android-upstreamed battery saver kernel features.
But a mobile is also people's most used devices with all of their data, bank accounts etc there - it has to be safe. And GNU+linux has not even a single thought about security, while android just has it worked out (every app runs as its own user, so it's even built on standard UNIX security).
A mobile OS also has to race to suspend and for that it needs cooperation from "apps" -- desktop apps just run, they don't care about anything besides SIGKILL. That's not a workable model on a mobile and android solves it.
And I say all that as someone who runs linux everywhere I can and I absolutely love it. It's imo the best kernel out there -- but the userspace is not where it should be and if anything, the correct question would be what can we take from Android and add to GNU+Linux. (And nix is fantastic, but it's a packaging solution, I don't really see how it comes into question here. I can run nix on my android phone just fine by the way)
As to your other userspace concerns... these are all solveable. Perhaps with some elbow grease, but devices like the Steam Deck prove that mobile linux isnt as much of a problem re: userspace as you claim.
Ed.: I'd also like to add that the fuss around security is _mostly_ Google propaganda. Android is not meaningfully more secure, _without application level changes_ than Linux.
> but devices like the Steam Deck prove that mobile linux isnt as much of a problem re: userspace as you claim.
So.... by relying on yet another company with multiples of $10B in revenue who spent a lot of time and money to build (and maintain) an entire custom frontend/compatibility layer to provide a stable UX for managing apps without intimidating non-tech savvy users? Where security isn't even a distant priority as it's a restricted store in a proprietary walled garden exclusively for games (and not your banking app or crypto wallet or browser tabs)?SteamOS, very similarly to Android, entirely depends on a corporate benefactor leveraging a parallel user space that happily bypasses the "normal" community-supported parts of a typical Linux desktop distro wherever they present an obstacle. It proves that a company with deep pockets could indeed create a new mobile Linux distro (by writing lots of checks and cutting out large chunks of desktop Linux in favor of custom implementations entirely under their control).
You mean, a browser?
It's almost like the free market only works when we have well-defined and regulated markets. This has been known by Adam Smith and quite logical, yet people expect Google and Apple giga-corporations with monopolies to somehow abide by the laws of selling grains on the market.
No. This is the end state of any semblance of laissez-faire economics. Full stop. Massive accumulation at the top, power to those with capital, rags for the poorest.
(which doesn't mean there's not a lot farther from here).
Not in the mobile world no, it's not a free market by any means
GrapheneOS’ security model makes that of desktop Linux look like a joke.
This is an objective analysis based on x86 security, GrapheneOS hardening (including isolation and hardened mem allocator), Pixel hardware security.
It has much better adoption of modern exploit protections and far better testing with sanitizers for both the kernel and userspace. It enables us to do even better because most of the memory corruption bugs caught by MTE are resolved. We do still need to resolve more, but it would be far more impractical for us to do it on the desktop.
Android uses SELinux for both whole system MAC and MLS policies with deep OS integration. It uses it for a massive amount of kernel attack surface reduction with allowlists for socket protocols, devices, ioctl commands and other functionality. It's far different from the traditional targeted approach used by desktops or even rare use of whole system SELinux for desktops/servers. It's nearly a completely different thing in practice. The OS has it deeply integrated in userspace for enforcement beyond in the kernel and it's developed around it. It's the main basis for the app sandbox and a lot of other isolation in the OS. OS processes are specifically split up and have IPC set up in a way that they can be contained well with it.
The mandatory app sandbox with yearly backwards incompatible privacy and security improvements as part of new target SDK versions is the most important difference. It's the basis for GrapheneOS being able to do much better. Having the infrastructure it already has available means we can add our features such as Contact Scopes, Storage Scopes and our Sensors toggle on top. We plan to add a lot more, but there are also the yearly improvements we get in the baseline such as how Camera, Microphone and Location have supported one-time grants for years, can only be used while apps are in use once granted (with Location have an extra layer of background opt-in) and precise vs. coarse location.
Because there is no such thing as "mainstream Linux" when it comes to anything related to user-facing consumer software. Not on desktops, not on tablets, not on phones.
I mean you invoked "mainstream Linux" and "Nix" in the same paragraph. That alone should clue you into why this, absolutely, does not work.
See here is the problem:
https://wiki.pine64.org/wiki/PinePhone_Software_Releases
"Linux enthusiests" would rather muck around with rewriting the same software over and over and over again because they dislike using GTK or whatever, and put monumental amount of efforts making new package managers, then, say, getting the ability to take simple photographs using a phone camera.
I mean... In that page there is no less then 25 different "Linux Phone OSes" listed.
None of them actually work.
They are all going to be slow, they are all going to burn through battery life. There is no meaningful security to speak of.
If I handed a unlocked "Linux phone" to somebody and said "take a photograph of me"... The chances of that actually working is slim to none.
Meanwhile we have Android OS that is proven to work. It is open source. It is used by, literally, billions of people. The security model is as good as it gets. It has better application support then Windows.
Taking something that works and then making it more secure and more open and more privacy focused is infinitely more productive and meaningful then trying over from scratch because you want a phone based around Nix packages or whatever.
Even if Google decided to close source Android from now on and be actively hostile to any open source kernel modules... Forking the Android that exists today and trying to make it work is exponentially more likely to yield positive results then, say, starting on a Debian-based "Linux phone OS".
And Android can still use nix-pkgs if you really wanted to.
Yes, there is: https://linux.org (aka mainline Linux). You can have various DEs on top of it, but the main functionality depends on the drivers in the mainline Linux, not so much on the DEs. Most of the operating systems for Pinephone listed in your link have more or less the same problems and features, because they all use the same kernel. They all depend on the upstreamed support of the device.
The existence of such a framework would make the various tradeoffs with going web-only sting less and make that the advantageous route, not just the cost-cutting route that it’s seen as now (and why those bad browser wrappers continue to proliferate).
If by "mainstream" Linux you mean something like postmarketOS, I'd suggest you look up reviews or give it a try yourself. A few months ago, people were reporting a hard time placing a call, taking a photo, etc.
Moving to a far less private and secure desktop software stack is going in the opposite direction from GrapheneOS. GrapheneOS doesn't exist to simply provide an alternative to mainstream operating systems but rather to offer much better privacy and security. We wouldn't be doing that if we were forking a desktop Linux environment and doing similar work for it. It would be nowhere close to the privacy and security of simply using an iPhone. That's a major part of why GrapheneOS is based on AOSP rather than it solely being about compatibility.
Desktop distributions are incredibly far behind on privacy/security and lack any clear path to achieving the same things. Every year, Android makes backwards incompatible privacy and security improvements as part of a new target SDK version. Android retains compatibility with legacy apps, but apps distributed through the Play Store (and other app stores to an extent) are required to move to the new target API level within around a year. This results in apps being forced to conform to a gradually improving privacy and security model. There's no such thing for desktop Linux apps but rather apps choose how much they want to participate in nascent sandboxing efforts.
GrapheneOS has near perfect compatibility apps from the Play Store via our sandboxed Google Play compatibility layer with the exception of banking and government apps. 90% of banking apps currently work on GrapheneOS because it greatly succeeds all of their security requirements and is only wrongly banned by a subset of those apps. These apps are gradually adding more anti-tampering and attestation checks for the hardware and OS, so maintaining compatibility has required us to gradually add more functionality working around it. We've also had to actively convince apps to stop banning non-Google-certified operating systems or to permit GrapheneOS and other secure options alongside doing it. A growing number of apps are choosing to stop banning using GrapheneOS due to pressure from our expanding userbase.
Android is a large Linux operating system family. It's the mainstream form of Linux on personal computers. Android users are Linux users. For privacy and security, using a monolithic kernel written in C is definitely not a good thing. Doing much better than we are today partly requires moving away from so heavily depending on the Linux kernel for security. Android does a lot of Linux kernel hardening with attack surface reduction and exploit protections which are improved by GrapheneOS, but it's not enough. The massive torrent of severe vulnerabilities being discovered in the Linux kernel is going to get worse before it gets better and will remain a problem. Adopting hardware-based virtualization for isolation of apps and OS components including drivers is an important part of our roadmap.
GrapheneOS is a non-profit open source project. We provide documentation and other information on the project.
> Android is not a Linux distro nor is it directly derived from Linux
Android is a family of Linux distributions. Linux is a kernel without a specific userspace and there are many different choices of libc, shell, init system, command-line utilities and far more. There are even more choices for the high level user interface. Android's userspace is largely specifically written to run on top of the Linux kernel with extensive use of SELinux and other Linux-specific APIs. A lot of the software you're considering to be Linux also runs on FreeBSD and elsewhere. FreeBSD is not Linux and the traditional desktop software stack largely shared with FreeBSD isn't Linux either.
Android Open Source Project uses the upstream Linux LTS branches. The upcoming devices with GrapheneOS support from Motorola use latest Linux 6.18 LTS branch. We plan to port to newer Linux kernel branches throughout the lifetime of the devices. Google already does that for Pixels and it needs to start happening for other devices too.
The upstream Linux LTS branches largely exist for Android with Google contributing a lot of the resources for it. Ubuntu and RHEL make their own LTS branches rather than using the ones from the Linux project. The lifetime for the upstream branches was increased from 2 to 6 years alongside Pixels increasing the support time to 5 years with the Pixel 6. It was decreased from 6 to 2 years because Pixels have been smoothly moving to newer kernel branches in production. It demonstrated the lack of need for the longer LTS branches not only for Android but more broadly. Pixels moved to 7 years of support with the Pixel 8 which wouldn't have been possible sticking to 1 kernel branch without a further lifetime increase.
> Mobile phones are not "personal computers".
Mobile phones are absolutely personal computers and should be treated as such. Many people use a mobile phone as their main personal computer. In large parts of the world, mobile phones are people's only personal computer.
GrapheneOS can even be hooked up to a monitor, mouse, keyboard, game controller and other accessories. It has a very functional desktop mode. It can also run desktop Linux applications via the hardware accelerated virtualization support. It even has opt-in GPU acceleration via gfxstream. That's all standard functionality from the Android Open Source Project.
> Trying to act like it being "mainstream" or "common" relative to desktop distros matters is just laughable.
There are billions of Android users. Each of those users is a Linux user. Linux is the dominant OS family on personal computers due to Android.
> And I take issue with the notion that a monolithic kernel cannot be secure.
A monolithic kernel written in C with an extreme focus on performance at the cost of having immense complexity is not reasonably secure. Piling on better exploit protections is useful but has a limited impact and those protections don't work as well in the Linux kernel as they do in the much more security-oriented userspace. Everything in the kernel being in the same address space with no isolation for regular code is not a good setup for security and severely hinders the usefulness of many exploit protections.
Further reducing attack surface even more than Android does is possible but is difficult while maintaining app compatibility. Hardware-based virtualization is the most realistic approach to dealing with the insecurity of the Linux kernel to properly protect it from userspace.
Protecting drivers from exploitation largely involves moving most of the code to userspace which is the norm in the mobile world and should expand much further. Bluetooth, Wi-Fi and other code can be largely moved out of the kernel. Moving TCP/IP, filesystems and many other drivers out of the kernel is entirely possible. However, it starts to run up against the Linux kernel having a culture and approach strongly against doing it. The same thing has happened with the attempt to introduce Rust in the Linux kernel with immense resistance and obstruction to simply providing bindings to write non-core drivers in it. There's no current attempt to use Rust in the core kernel and it would be an immensely slow process even if that was happening.
Hundreds of severe vulnerabilities being found per month with only a subset of those getting fixed is not a sustainable approach to making a reasonably secure OS.
> then why arent you guys working on the next kernel for personal and high-performance computing
We're doing initial work on expanding use of hardware-based virtualization to further reduce reliance on the Linux kernel for security. We're also building up our resources to be able to take on much bigger projects. Aside from that, we don't need to be doing things entirely on our own.
A lot of what we want aligns with where AOSP is already headed and many others want the same things. AOSP added hardware-based virtualization with the userspace code written in Rust and a more secure approach to handling it in the kernel (pKVM) without our involvement. AOSP also began to make very limited use of this for OS sandboxing itself. We want to do much more with it and use it for sandboxing apps, which are not things Android is currently working on as far as we know but likely will be in the future.
We're using a project account making it clear these responses are from the GrapheneOS project. That's the opposite of a sockpuppet account. We've moved to using this project account instead of individual accounts because of personal attacks directed towards our team. A community member created this account and transferred it to us to encourage us to keep posting on Hacker News.
Also if you only appeal to those that care about having exactly that, the sales will never be high enough to keep a sustainable business, unless it is at the scale of a phone vendor at least like Jolla.