E: apparently it needs to be said that I am not suggesting you switch to Linux on your phone today; just that development needs to accelerate. Please don't be one of the 34 people that replied to tell me Linux is not ready.
Android has always been a fairly open platform, especially if you were deliberate about getting it that way, but we've seen in recent months an extremely rapid devolution of the Android ecosystem:
- The closing of development of an increasing number of components in AOSP.
- Samsung, Xiaomi and OnePlus have removed the option of bootloader unlocking on all of their devices. I suspect Google is not far behind.
- Google implementing Play Integrity API and encouraging developers to implement it, which prevents apps from the Google Play Store from being downloaded without a system-wide OS-level account login. Notably the EU's own identity verification wallet requires this, in stark contrast to their own laws and policies, despite the protest of hundreds on Github.
- And finally, the mandatory implementation of developer verification across Android systems. Yes, if you're running a 3rd-party OS like GOS you won't be directly affected by this, but it will impact 99.9% of devices, and I foresee many open source developers just opting out of developing apps for Android entirely as a result. We've already seen SyncThing simply discontinue development for this reason, citing issues with Google Play Store. They've also repeatedly denied updates for NextCloud with no explanation, only restoring it after mass outcry. And we've already seen Google targeting any software intended to circumvent ads, labeling them in the system as "dangerous" and "untrusted". This will most certainly carry into their new "verification" system.
Google once competed with Apple for customers. But in a world where Google walks away from the biggest antitrust trial since 1998 with yet another slap on the wrist, competition is dead, and Google is taking notes from Apple about what they can legally get away with.
Android as we know it is dead. And/or will be dead very soon. We need an open replacement.
E2: thank you to everyone stopping by from Hacker News, Reddit, etc. to check out the threadiverse. I hope you'll stick around for a while. Check out https://phtn.app/ and the Voyager and Blorp apps for a nicer UI. Fuck Spez!
Unfortunately there's a lot(!) to do to make Linux enjoyable on a phone. I bought a Pinephone some years ago. And in addition to the slow hardware, the entire software/desktop experience isn't great. While everyone else has instant messengers, Linux doesn't have connected standby and emails and messages just don't arrive unless the screen is on. It wastes quite some power, and there are a bazillion small little quirks and annoyances and it's barely usable if compared to a regular smartphone. I think someone needs to invest quite some more time and money until this becomes a thing. I mean don't get me wrong, Linux and the low-level system is awesome. And it's brilliant on any server/laptop/desktop computer. It's just that there's so many things missing for a proper phone experience. And it's not just mildly inconvenient, but like people expect instant messages to be delivered...
It seems like you read the title as "everyone needs to switch to Linux mobile right now" but that's not what it says.
The point is, as you said, there's a lot of work that needs to be done, and that work is more important now than ever.
Sure. It's just that the timeframe is a bit disheartening. To me... so all of this is highly subjective. We had the Nokia N900 in like 2009. And I was expecting to live the full Linux experience within a few years and those things to become a bit more affordable. And today it's almost 16 years later and it doesn't feel like we've come substantially closer. More recently we had Librem and Pine64 put some effort and publicity into it, and that's also been 5 years. The mobile/touch desktops made some good progress. PostmarketOS is kind of nice. But there are entire layers missing like the app framework in Android which enables such app lifecycles, connected standby... Sandboxing and a fine-granular permission system for proprietary apps (or just modern mainstream usage) is kind of in its infancy. And I'm not even sure if everyone is going to use Flatpak for everything. And all of those missing things are huge undertakings.
So I'm not sure when to expect such an every-day phone... Maybe in 2030 or 2035? But that's kind of late if the headline is "more important now, than ever". Because all the while Google is moving more and more stuff from AOSP into their proprietary Play services and it's getting uncomfortable for me. We have a deadline with the Google messes with the allowed apps on a phone starting 2027. And my life includes more and more mandatory apps, or I have to forfeit taking part in society, culture, convenience or riding a train... This year, Google started giving the GrapheneOS devs a hard time... Now they're making it even more complicated.
So of course not everyone has to use it, and I'm first of all concerned with my own wellbeing. But I really don't see a solution in the near future which is going to address the important issues if today and the next few years. So I'm a bit unsure if a Linux phone will come around and help me before it's too late, or if I need to find other ways to deal with it.
True but I also never even considered it before because honestly open source Android works really great on most devices already. Now that that's all basically disappearing, hopefully more people will be more dedicated to creating an alternative.
Can you help me understand why Linux phones are the answer rather than a community maintained android fork?
Android is already fully featured and has a solid ecosystem so it's usable now, not in 5-10 years with less of a need for adjustment for the people who want to switch.
Basically, why take several steps backwards and start from scratch?
Because a community maintained fork wont solve the problems in the OP?
I strongly disagree with this comment. I'll answer your numbered points from the original post one by one with my perspective:
Do you disagree with any of these? Would love to hear your thoughts
Wow. Ok.
All "community driven forks" are based on Google's AOSP. None of them have the resources to develop this stuff from the ground up.
No but someone would sell Linux devices if they were commercially viable, and no one would buy a Linux device with a locked bootloader.
You wouldn't need it on Linux mobile because...it's not Android.
I already addressed this in OP.
But then you need apps that work on Linux (optimised for mobile/touch). You can also easily create Apps for Android without play integrity API necessity.
Realistically an Android fork makes more sense.
Though in my ideal dream world a Rust based mobile wayland compositor (etc.) will be the future of open mobile OS. I hope there's enough (financial) interest to at some point reach that future.
Correct again! Running Linux apps on Linux, what a concept!
It doesn't, for all the reasons I listed in OP.
Don't get me wrong, I'm the first promoting an Android free mobile Linux, free of big company influences.
Though, what I meant is that there's very few mobile optimised apps on Linux, and I doubt that changes soon. The Android SDK is very matured (like Compose for UI). It's fairly easy to create a good native app experience in Android. Less so for non-Android Linux. (I've developed apps for either) Think about that alone, which further complicates adoption, which TBH is just necessary to get to an ecosystem that us usable for daily usage.
I hope that changes sooner than later, but the current alternatives are just not there yet.
So which is the best project to back?
I dunno! It will really require the participation of the entire community.
Gnome has been making great progress on the graphical front.
Notifications should be pretty simple, and probably should be provided by hardware manufacturers. But the support will need to be implemented into the apps that need them. That can potentially also fix the battery issue.
PostmarketOS I think is probably the most mature Linux mobile package currently but I'm no expert on the subject.
I'd argue that Ubuntu Touch and Sailfish are the most mature offerings. Both OSs are (or at least were at some point) developed as commercially viable alternatives to the duopoly. That gives them a headstart in terms of apps and overall pollish.
The postmarket shells are catching up, but you still get instructions like "drag and drop a file from your file manager to open it", which doesn't work on a phone. Phone UX still seems like an afterthought in many cases.
Postmarket OS is a desktop Linux system, but for phones. UT and Sailfish on the other hand are mobile OSs, that happen to use much of the same tech as desktop Linux. They are therefore much closer to the duopoly (for bettet or for worse).
Great, thanks! The next step would be to figure out if either is compatible with Syncthing and GPS-guided nav...
Syncthing:
Navigation:
Woohoo!! Now I can't decide between the two...
Have had both. Still have Sailfish because the phone is cheaper.
Also I thought Ubuntu Touch would be discontinued and I no longer use Ubuntu on my desktop,
but an Arch-based OS.
Best thing you can do is buy a phone that's most compatible to the OS.
So Fairphone 5 or Pixel 3a for Ubuntu Touch,
and Jolla 2 for the Sailfish.
Do not buy Pine64.
Pine64 is unusable.
Dang it, I gotta change, I guess, unless this info is old:
Isn't the VollaPhone Quintus the best option for Ubuntu Touch? (It's more expensive than the Fair Phone, but it ships with UT)
https://volla.online/en/index.php
If it does ship it with UT and Fairphone doesn't, then yes.
How cool would it be to out of nowhere see Valve come out with a SteamPhone based on Arch which does everything you ever hoped for and runs on high quality hardware including all the features that others took away (colour alert pixel, 3,5mm jack, replaceable battery), complete with dual boot or a containerised Android-mode for running apps that would never work like banking or eID. Would buy instantly.
comparison is the thief of joy and it's unrealistic to expect a product of grass roots community effort to compare to a product that benefits from deep pockets that can afford the finest talent and a considerably long head start.
I'm not sure. A phone is kind of a tool, same for a computer. Ideally we weigh our options (aka compare them) and pick which suits us best. And this grass roots effort isn't doing a half-assed job. I have almost everything available. I can do regular tasks, edit videos, do computer aided design, do heavy database stuff, run the bookkeeping for an entire mid-sized company, a server farm or almost anything I like with Free Software and Linux. I don't think a phone is fundamentally different and I kind of have the same high expectations for that niche. We're already doing great with lots of other things, both more niche and more ordinary stuff. And oftentimes it does not include money from tech giants.