Let's start a new project : let's create a new mobile device powered by FreeBSD.

Status
Not open for further replies.
ziomario

ziomario

Hello Freebsd-Lovers.

I plan to buy the "PINEPHONE – Beta Edition with Convergence Package Linux SmartPhone". I would like to install FreeBSD instead of Linux on this device. I would like to create a team of developers. I can help those developers,but not deeply,because I'm a system administrator as a hobbyist. I can hack and configure a system if it does not require high technical skills. I'm not a developer. I'm not a software engineer. I would also know if I can participate by putting in some money into the project,because I know that you would take away time that you should devote to your job. I'm not able to quantify how much money this project would require to be completed,to have a clear idea if I can start or not the project itself. If the amount of money is very high for me,I would like to know if whether a crowdfunding project could be started to get the necessary money. Let me know your thoughts. thanks.
 
Pinephone started shipping in mid 2021. It is still considered experimental on ubports.
So much is missing. A list of what works and what doesn't is here: https://devices.ubuntu-touch.io/device/pinephone/
It would be a waste of time and developers to try PinePhone on FreeBSD.

We should rather help linux (ubports) get the battery last more than 24 hours, and help with video/camera instead, I feel. FreeBSD on smartphones is not happening. Sorry for being so pessimistic 😅
😊
 
I have a PinePhone and would certainly enjoy running FreeBSD on it, however the efforts needed to make this possible would be huge, and in my opinion not worth it. This device is an experimental prototype, built with components that weren't designed for smartphones in the first place. It has been already three years since the first PinePhones were delivered and although it's more or less usable as a basic smartphone now (I use it daily), there are still lots of issues, both with the available software and the hardware itself. When the whole thing will have improved, it may become interesting to port FreeBSD to it.

George said:
Pinephone started shipping in mid 2021.
Click to expand...
Already in early 2020, actually.

George said:
It is still considered experimental on ubports.
Click to expand...
UBPorts is among the least functional OS on the PinePhone as of now, and sees very little development. If you want something usable, you'll be better served by postmarketOS (Alpine Linux) or Arch Linux (unofficial port). Some will also recommend Mobian (Debian) and Manjaro, but they have been less reliable for me.
 
ok. For me its ok to change the hardware components. I'm not affiliated with the pinephone idea. We can look for the most supported hardware components for FreeBSD and we can assemble these components into a phone...or a tablet or some other mobile,interesting device. Maybe using this kind of approach is more doable ? We can keep active the final goal (to create the first mobile phone or maybe tablet or something else ?) powered by FreeBSD but let's talk about how to do this.
 
yuripv79 said:
Just... why? You surely don't expect various software vendors to start supporting yet another (amateur, at best) platform?
Click to expand...

Because it would be damn cool and fun. I don't want to look too far. It starts like an amateur project that,I think,a lot of us will love.
 
Would be nice if a group using FreeBSD would decide on, then
suggest one of each class of printer [ inket/tank, mono/color laser, thermal, multifunction... ] and put up a precise howto for each ...
 
ziomario said:
Hello Freebsd-Lovers.

I plan to buy the "PINEPHONE – Beta Edition with Convergence Package Linux SmartPhone". I would like to install FreeBSD instead of Linux on this device. I would like to create a team of developers. I can help those developers,but not deeply,because I'm a system administrator as a hobbyist. I can hack and configure a system if it does not require high technical skills. I'm not a developer. I'm not a software engineer. I would also know if I can participate by putting in some money into the project,because I know that you would take away time that you should devote to your job. I'm not able to quantify how much money this project would require to be completed,to have a clear idea if I can start or not the project itself. If the amount of money is very high for me,I would like to know if whether a crowdfunding project could be started to get the necessary money. Let me know your thoughts. thanks.
Click to expand...
Would for sure be nice.

We also want to look into this (for the current Pinephone Pro, no), but we have not yet found a developer (admittedly we have a very limited job market scope): https://lists.freebsd.org/archives/freebsd-jobs/2022-December/000017.html

Somebody has already worked on making OpenBSD boot, so it should be possible: https://research.exoticsilicon.com/series/pinephone_openbsd/part_1
 
pebkac said:
Would for sure be nice.

We also want to look into this (for the current Pinephone Pro, no), but we have not yet found a developer (admittedly we have a very limited job market scope): https://lists.freebsd.org/archives/freebsd-jobs/2022-December/000017.html

Somebody has already worked on making OpenBSD boot, so it should be possible: https://research.exoticsilicon.com/series/pinephone_openbsd/part_1
Click to expand...

That's so nice. I'm excited. Are those instructions useful for us ? Can we start taking some useful information there to adapt it for FreeBSD ? Is the NetBSD code enough close to the FreeBSD code ? If this is true,some of the work is already done. Or not ?
 
Back in 2015 I bought a Jolla smartphone with SailfishOS. The OS is in fact a Linux / UNIX and is rock solid. Back then Jolla had the advantage of making an OS for their own device, just like Apple does. When hard- and software are in one hand everything works. SailfishOS AFAIK now sells images for other hardware.

Making it short: maybe you can fork that OS and FreeBSD-isize it!
 
A few things;

  • How will you handle cellular communication? Getting all major telecoms to support the system for CMDA/GSM/SMS communication is going to be a challenge. Does FreeBSD even have drivers for this type of stuff?
  • How will you handle software distribution? I presume you'll have to host a repository yourself along with a way for users to administer mobile apps and system updates over-the-air; another major challenge.
  • How will the system be installed? One of the reasons why I loathe GrapheneOS is because it requires chrome for installs/updates. When the install medium should be its own cross-platform utility.
  • How will third party applications be developed? There's no truly open source toolkit out there for mobile devices. Except maybe Qt; you'll probably need a commercial license for this kind of project. Who the hell wants to write mobile apps in C++ anyways?
Also, in order for a project like this to gain any momentum; you'll have to create a community around it for people to collaborate with. Assuming you'll want to keep it open source. Making a mobile OS is ridiculously complicated.
 
Beastie7 said:
A few things;

  • How will you handle cellular communication? Getting all major telecoms to support the system for CMDA/GSM/SMS communication is going to be a challenge. Does FreeBSD even have drivers for this type of stuff?
  • How will you handle software distribution? I presume you'll have to host a repository yourself along with a way for users to administer mobile apps and system updates over-the-air; another major challenge.
  • How will the system be installed? One of the reasons why I loathe GrapheneOS is because it requires chrome for installs/updates. When the install medium should be its own cross-platform utility.
  • How will third party applications be developed? There's no truly open source toolkit out there for mobile devices. Except maybe Qt; you'll probably need a commercial license for this kind of project. Who the hell wants to write mobile apps in C++ anyways?
Also, in order for a project like this to gain any momentum; you'll have to create a community around it for people to collaborate with. Assuming you'll want to keep it open source. Making a mobile OS is ridiculously complicated.
Click to expand...

My idea is to find in this forum and why not ? on the relative freebsd ML the developers who wants to collaborate together. And at the beginning I think its better to think at this project like an amateur project realized for fun. Anyway,at this point the best alternatives to keep in consideration are :

1) a jolla smartphone (compatible with sailfishOS)

Xperia 10 III Dual SIMXQ-BT52Android 11
(12 is ok)		4.19 / 1111 / 304.4.0
Vanha Rauma
Xperia 10 II Single SIMXQ-AU51Android 11
(10 is ok)		4.14 / 10-“-4.1.0
Kvarken
Xperia 10 II Dual SIMXQ-AU52-“--“--“--“-
Xperia 10 Single SIMI3113, I3123Android 9
(do not use 10)		4.9 / 9-“-3.2.0
Torronsuo
Xperia 10 Dual SIMI4113, I4193-“--“--“--“-
Xperia 10 Plus Single SIMI3213, I3223-“--“--“--“-
Xperia 10 Plus Dual SIMI4213, I4293-“--“--“--“-
Xperia XA2 Single SIMH3113, H3123, H3133Android 94.4 / 8.1-“-3.0.0
Lemmenjoki
Xperia XA2 Dual SIMH4113, H4133-“--“--“--“-
Xperia XA2 Plus Single SIMH3413-“--“--“--“-
Xperia XA2 Plus Dual SIMH4413, H4493-“--“--“--“-
Xperia XA2 Ultra Single S.H3213, H3223-“--“--“--“-
Xperia XA2 Ultra Dual SIMH4213, H4233-“--“--“--“-
Xperia X Single SIMF5121Android 83.10 / 64.4.4 / 192.1.2
Kiiminkijoki
Xperia X Dual SIMF5122-“--“--“--“-

Sailfish OS can be ported to (almost) any device which currently runs Android.

2) the pinephone
3) an old phone compatible with PostMarketOS
4) an old phone compatible with Ubuntu Touch (
  • Google Pixel 3a/3a XL.
  • OnePlus One.
  • Sony Xperia X (F5121 & F5122)
  • Google Nexus 5)
 
Do you have a workable 'proof of concept' that you can demonstrate? Like, do you have something we can play with and complain about?

Even Linus Torvalds famously wrote much of the initial Linux kernel all by himself before showing it off to the likes of Tanenbaum.

I see most people on the Internet get the playbook just flat out wrong - they first try to gather a crowd, and then get to work... And nothing gets off the ground.

In my family, there's a saying: "You first work for your name, and only afterwards, your name will work for you".
 
ziomario said:
I can't start a project like this by myself only.
Click to expand...
I'd start with writing a PID; Project Initiation Document. Outline some details, have something to 'sell' to others. Just throwing out ideas and expect someone else to do the actual work isn't going to win you any support.
 
Beastie7 said:
How will third party applications be developed? There's no truly open source toolkit out there for mobile devices.
Click to expand...
AOSP comes to mind... That is actually Open Source, as long as you're willing to study your ass off and actually read the source code. OP will need to have enough education to know which parts can be replaced with FreeBSD-based implementations, and to know how to recompile that enchilada and how to check that it's actually working...
 
astyle said:
Do you have a workable 'proof of concept' that you can demonstrate? Like, do you have something we can play with and complain about?

Even Linus Torvalds famously wrote much of the initial Linux kernel all by himself before showing it off to the likes of Tanenbaum.

I see most people on the Internet get the playbook just flat out wrong - they first try to gather a crowd, and then get to work... And nothing gets off the ground.

In my family, there's a saying: "You first work for your name, and only afterwards, your name will work for you".
Click to expand...

At the moment I'm looking for a smartphone where I can install Linux or where where Linux is already installed. This phone should also have a good tutorial that I can follow and "make" until the end to understand at least the whole process of the Linux installation on that specific phone. In stalled on this point. I think that the choice of the right phone is the most important step before to say to have a project. And in the choice for me there is also the money component. The phone should be cheap. So,it should be a relatively old and cheap phone or a new but also cheap phone (we are at the price level of the pinephone : I can't spend more).
 
There should be a project page for this where code changes can be tracked and testers can contribute tests and bug reports etc.
 
Here's an idea. Why not use bhyve to create a simulated mobile UI for others to play with. Then once FreeBSD lands enough driver support, you can install the UI and the base system on hardware. Or go the other way with getting the base system installed first on said hardware.

Either way, we're going to need something to tinker with. Going the bhyve route would probably be easier since all the userland infrastructure is there; ie. Weston, OpenGL, DRM, Qt, etc. If you start from the bottom... have fun writing drivers. :)

Getting Weston and DRM to work on a mobile device using FreeBSD is uncharted territory. It's like trying to build a house on mars.
 
Bhyve does not work on arm,but there are some developers who said that they want to port it there soon. In this case using qemu + kvm is the only choice,so we are forced to use Linux.
 
For me is the point,since I want to make the tests and the configuration of everything directly on the hardware that I will use.
 
ziomario said:
For me is the point,since I want to make the tests and the configuration of everything directly on the hardware that I will use.
Click to expand...
Then do it.
1. Work on it
2. Show results

It was said before, but maybe not clearly enough: You won't magically find someone working on whatever you suddenly dream about. My gut feeling (and I assumed it's shared by many) is: It's certainly possible to port FreeBSD to some other hardware platform. And the result will be utterly useless.
 
Status
Not open for further replies.
Back
Top