Pocketblue – Fedora Atomic for mobile devices(github.com) |
Pocketblue – Fedora Atomic for mobile devices(github.com) |
I have yet to attempt daily-driving it, but just trying it and easily switching mobile shells (e.g., from Plasma Mobile to Phosh) so easily[0] without have weird side-effects from the previous environment has been quite exciting!
[0]: https://pocketblue.github.io/devices/oneplus-sdm845/#images-...
Very similar to how Universal Blue, Bazzite, Bluefin etc. build at https://github.com/ublue-os/bazzite (see their Containerfile), but for mobile.
Has a similar mission to https://postmarketos.org, but with a different build system AFAICT
> Dockerfile
1. https://docs.fedoraproject.org/en-US/bootc/getting-started/
Boot images should be Dm-verity protected EROFS images. We should not be building new things on OCI. It's really mind-blowing to me that this is a new direction people who are supposed to be top of class OS builders are moving to as a direction.
They took the CoreOS dream and threw everything in the trash
> Boot images should be Dm-verity protected EROFS images
Maybe I'm misunderstanding you - I gather that you think the boot images are distributed as OCI images? That's not the case, bootc is more about building the image, updating it and the overall structure. Booting an image built with bootc does not involve any container infrastructure (unless you start services that depend on containers, I guess - but that's deep in userspace). There's technically nothing preventing this from using verified read-only images.
> Dm-verity protected EROFS images
I've tried the silverblue desktop version of this and while I'm not convinced that a mix of OS/Brew/Flatpak/Containers is making things more approachable it's interesting to see these concepts progress and the tools improve.
My eyes are currently on the Lenovo IdeaPad Duet 3 as it seems to tick most boxes on PMOS' compatibility matrix.
Mobile Linux is a super interesting but difficult area so always good to see another effort in the space, hopefully Pocketblue and postmarketOS can benefit from each other.
Yea. I hope the same thing. I just wish that a defacto affordable device can finally emerge and the whole process can just be made easier and people might even base their hardware decisions on top of it.
A lot of the times I find Mobile Linux un-approachable because a lot of essential features only work in phones which are extremely expensive with less specs compared to its peers and in normal phones, half the features might not work (which is understandable given they are community efforts but still)
I just hope that the community/such-projects can decide on one single affordable phone and then make it feature complete and maybe even recommend people to use it as well imo.
I assume you are refering to the Librem 5? These days the Google Pixel 3a is coming close, as multiple people report reliable phone calls. Camera is still lacking, but once libcamera add auto-focus and the device gets a driver for the focus actuator (already in the works), it's going to be just as competent with better battery-life for less than USD/EUR 100. On the more pricey side, the Fairphone 5 has a good mind share and contributor count, making it quite likely that remaining issues are being solved soon.
- Apps: It's Linux (like desktop or server), but "image-based" so you install apps in containers like iOS or Android do (and therefore OS updates basically-never break). https://flathub.org is generally the main app store for Linux containerized phone apps.
- Screenshots: It'll look the same as other Linux-on-phones, so like https://en.wikipedia.org/wiki/PostmarketOS for instance. It's just built differently.
[0]: https://framagit.org/linuxphoneapps/linuxphoneapps.frama.io/...
[1]: https://framagit.org/linuxphoneapps/linuxphoneapps.frama.io/...
That said, as mentioned in another thread here, work is being done to add PinePhone support.
Written on a Pinephone.
Anyway, Fedora is sooo good. :D
Xiaomi pad 5/6
Oneplus 6/6T
... That's it.
Does anyone know if there's plans for more? Is this project in very early stages, or is it going to be another Graphene OS with an extremely limited device support?
These projects (Linux on mobile) are even more limited due to very poor support from the manufacturer for anything more than OEM and device specific build of Android, with lack of standards in mobile platforms. Every device support is reverse engineering effort. See https://wiki.postmarketos.org/wiki/Devices for the status of this effort.
There's also a person working on a Fairphone 5 support and I think someone was going to work on a PinePhone port
Contributions are always welome, we need more devices!
Are these devices so different that none of the testing and development work is transferable to the newer devices?
And, reversing the question, if one was to be the owner of a such a Xiaomi device, what can be done to help them being supported?
Though it would probably be trivial to just copy what the non-atomic Fedora Mobility does for pinephones, I might as well do this in the future and just ask someone with a pinephone to test
Can this run Distrobox https://distrobox.it/ ?
It would be cool to have say debian packages or even arch packages while the base os is rock-solid in an Mobile Device while being pure Linux!
I hope more devices can be added along the way. Can the community have one cheap device and make it the standard de-facto where affordability can thus be achieved?
To me, I don't really care about specs of such phone. I just want a linux phone which can just run good enough and have such features
- Cheap to Buy
- Easy to boot Linux in
- Has calling functionalities
- 5G support (sort of optional but would be nice to have)
- Software Side: Flatpak is good enough for me with distrobox on the side. It can finally be an os where you don't need java sprinkled everywhere for it to work :D
Is there any particular phone brand which is available now for this? or can the community take such feedback and look at any such hardware and do the following and then be able to give a definitive answer to my query.
Maybe even in future, you can make a single guide for people like me who just want linux phones for affordable-pricing where I can read it and run this project!
I have been thinking more and more about using handhelds for such purpose or a rasp pi phone (there is a guy I know who is working on it) & I hope he succeeds as well.
Recently, a rasp pi laptop was launched, jeff geerling made a vid about it but because of the inflation of prices, that idea is sort of dusty right now (also with the lack of sleep mode)
So I feel like for affordable Linux phones, right now with the current AI bubble. Projects like these who can bring Stable Linux into Android phones are the best and this is a unique concept of adding atomic in Android phones which is so cool!
I really want this, I can't explain how much if it can be brought to a level which is simple. For some reason Rooting phones feels hell even though I have played/tinkered with so many os in PC's.
I wish you good luck in your project & if you have any suggestions for a cheap phone then let me know as in the future, I would love to playtest fedora atomic a lot.
I have used dumb phones a lot and to be honest the reason why I don't like SmartPhones is that I find them to be restricting & at times addicting. Its too easy to scroll youtube shorts in phone then try to create an application which is helpful unless you write it in java and just a very lack of customization. I want phones where I can develop applications in any lang I want without too much PITA. Linux is it!
Also maybe this can also be used waydroid to run android applications natively too. It would be interesting to have that possibility as well where I might still need android apps :D
- only works with very few phone models
- battery doesn’t last long
- bad UI with tiny elements.
- not managing a smooth refresh rate
- no apps
That‘s the pattern we‘ve seen over and over again. The only approach that has worked better is to base things on AOSP.
If I were to do a Linux Phone platform, I'd be targeting feature phone levels of functionality to begin with, with a focus on battery life and actually working telephony. I'd be aggressively throwing Wayland/GTK and all that nonsense in the bin just to get something basic working well. Draw straight to the framebuffer if you have to. This doesn't help with the app problem, but it sets a tide mark for quality & performance, and it can be iterated on.
That said, if you want to start without the entire Linux desktop stack, you can, and there's even a project that already does something like that IIUC: https://sr.ht/~mil/framebufferphone/
In AOSP, not even the call app is usable nowadays. AOSP doesn't have any functional UI anymore.
So depends what's your definition of a phone stack, if it includes a call app, an agenda or a keyboard, AOSP doesn't fit it.
Yes? That's literally the sales pitch on the website. Am I missing something?
Quote from https://bootc-dev.github.io/ tells me that bootc is using OCI as a delivery format for bootable images.
Transactional, in-place operating system updates using OCI/Docker container images.
Motivation The original Docker container model of using "layers" to model applications has been extremely successful. This project aims to apply the same technique for bootable host systems - using standard OCI/Docker containers as a transport and delivery format for base operating system updates