Looking for a desktop environment suitable for use as a phone interface in FreeBSD installed on a phone.

Older Motorola, Samsung and Pixel phones are well known and fully hacked. I remember old LG phones where you can dual boot Windows CE and Android. Unfortunately, such hardware is way too old for these kids that need 8K display resolution and 150 Megapixel octa lens camera. Which brings us to original conclusion. Its completely pointless and useless. In this very case, there is nothing to learn and nothing to gain.

But even if you install FreeBSD there, does one have the driver for the modem? For GPS? For touch screen? And all the hardware there?

That is why I insist in building from android.

I would not say pointless, a mobile unix-like computer in which one have the control may have many uses, even if it is not usable as modern smartphone.
 
The problem here is not technical at all. I'm sure ZioMario will overcome all technical problems and finally succeed, whichever route he takes. The problem is that his phone will be useless because it won't have any ecosystem.

Take banks, for instance. There's a trend among them to discontinue their web banking in favor of their banking app. If you have a smartphone, you absolutely want it to run your bank's app, but your bank only provides it for iOS and Android.

Building a phone OS is challenging, but given enough time and dedication, a single person can do it. What a single person will never be able to do is build an ecosystem - only third parties can do that, it if there's a return on investment.
 
But even if you install FreeBSD there, does one have the driver for the modem? For GPS? For touch screen? And all the hardware there?

That is why I insist in building from android.

I would not say pointless, a mobile unix-like computer in which one have the control may have many uses, even if it is not usable as modern smartphone.

The touch screen already works in my scenario without any special configuration. That's because I've chosen a good display. In addition I'm using one USB physical mini keyboard. I will not implement the GPS. I don't need any special driver to enable the LTE / 4G + 4 ports HUB modem USB. So. I don't need any Android.
At the moment is already recognized by FreeBSD. One later step is to write a decent DialPad for placing and receiving phone calls. But since this is a phone,I can't stop here. I need a full desktop manager that looks like perfect to be used on a phone,that will add a Contact list and a Call history which naturally covers both dialed and received calls. Yeah,I'm enjoying very much,I will make a decent final product.
 
The problem here is not technical at all. I'm sure ZioMario will overcome all technical problems and finally succeed, whichever route he takes. The problem is that his phone will be useless because it won't have any ecosystem.

Take banks, for instance. There's a trend among them to discontinue their web banking in favor of their banking app. If you have a smartphone, you absolutely want it to run your bank's app, but your bank only provides it for iOS and Android.

Half True. For the banks,yes,I'm out. I can't do nothing to support home banking. But I will try to improve the Linuxulator for a better Linux compatibility and I will try to port some software structure from Android to FreeBSD,with a fresh reimplementation,for supporting a limited number of Android applications. As alternative,I could try to virtualize Android using qemu+bhyve.
 
I don't need any special driver to enable the LTE / 4G + 4 ports HUB modem USB
You mean, the drivers of FreeBSD work on the LTE / 4G modems?

Every modem need a special driver, it is a headache to find a modem that works with an OS other than the supported ones by the manufacturer.
 
You mean, the drivers of FreeBSD work on the LTE / 4G modems?

Every modem need a special driver, it is a headache to find a modem that works with an OS other than the supported ones by the manufacturer.

Code:
ugen0.2: <USB 2.0 Hub Terminus Technology Inc.> at usbus0, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=SAVE (100mA)
ugen0.3: <EG800Z-EU Quectel Wireless Solutions Co., Ltd.> at usbus0, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=ON (
200mA)
 
You unlock the bootloader. The pre-OS firmware has enough functionality to write OS partitions from USB commands (via the fastboot command). That's about it.

If you are able to unlock the bootloader,that's ok...another project that I have planned in the future is about to install FreeBSD on the BlackBerry PassPort Q30. In this case the bootloader can be unlocked only applying an hardware mod.

I've sent it to the lab to apply the mod :

WhatsApp Image 2026-08-18 at 22.20.09.jpeg


If I will be lucky and the mod will be succesfully,it will run FreeBSD,including the internal modem that I will try to make work porting the driver from Linux.
 
FirefoxOS/KaiOS are mobile phone operating systems. They use the Linux kernel and include graphical environments. I thought perhaps you could port the UI to work for your project.

These projects are very old. I'm not sure if they worth. Maybe it's a good idea to choose one option in the list of options that I showed on the first message.
 
What are we talking about here, vibe coding a functional mobile OS?

Lmao.

Also Zio, when you say "I will try to improve, I did, I built" is it you or the LLM? LLMs do not allow power users to become software architects regardless of what your initial results might look like.
 
What are we talking about here, vibe coding a functional mobile OS?

Lmao.

Also Zio, when you say "I will try to improve, I did, I built" is it you or the LLM? LLMs do not allow power users to become software architects regardless of what your initial results might look like.

Your question does not make sense for me. I'm focused to the result. Claude helps me doing what I want and I use it. That's all.
 
You can say what you want, ZioMario's threads always gather attention 😅

What I want is not the attention. I wanted to talk about what I have asked on the first message. No one after the first message replied in a coherent manner (except few users), and after having tried to get you back to the topic,this one :


I failed and I've decided to go with the flow—that is, your comments. I certainly can't force you to discuss what I want. I suspect that all of your attention is not focused on the projects,but a lot of you are obsessed by AI,so,it's hard for you to keep the focus of your attention where it counts more.
 
The issue is people cant talk to you about how to implement any of your ideas

By discussing the code and the technical aspects
because you are vibe coding with Claude Code and arent a developer

Thats not a criticism of you
Its just the reality of vibe coding
 
The issue is people cant talk to you about how to implement any of your ideas

By discussing the code and the technical aspects
because you are vibe coding with Claude Code and arent a developer

Thats not a criticism of you
Its just the reality of vibe coding

Let's read again the first post. To reply to the question I posted you don't need to be a developer, because the topic is very light. Furthermore, I wasn't a programmer even before AI came between us,when every post of mine received a lot of replies. So the reason isn't me because I'm not a developer, but your obsession for AI. If I hadn't said I use AI (lying), you would have continued to reply to my technical questions,as always. I think you should ask yourself why you hate so much a tool and those users who use it so much, even though it produces a lot of good code and projects that without it wouldn't see the light even in 20 years. The first concern has been that the post looked like it was written by an AI, not to read its content at first glance.
 
Its not just choosing a "desktop" for the looks

But the technical considerations on how you would get it working on Freebsd

You would need to know what needs to be ported for the "desktop"
and how to get it working with the hardware

Because no one has your phone hardware
its makes recommending a "desktop" for it pure guess work
 
ugen0.2: <USB 2.0 Hub Terminus Technology Inc.> at usbus0, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=SAVE (100mA) ugen0.3: <EG800Z-EU Quectel Wireless Solutions Co., Ltd.> at usbus0, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=ON ( 200mA)
But that still does not mean having a driver.
 
You would need to know what needs to be ported for the "desktop"
and how to get it working with the hardware

Bro,the discussion is not even started.

Because no one has your phone hardware

Are you sure ? I've informed that the soc that I'm using is the Radxa Cubie a7z :

 
There are 2 things

1) Phone operating system
2) Desktop "theme" for a phone

1 ) So your either trying to port a Linux/Android operating system and merge it with Freebsd
to provide the functionality for making calls and send texts

Different phone operating system only work on particular hardware
so the issue is then getting that phone os to work with the hardware you are using


2) Or you are looking for a GUI / Desktop / Theme to put on top of Freebsd
and Freebsd provides all the functionality for making calls and sending text

However the issue is that desktop themes
will be designed to work another operating underneath like linux or android

So you will then need to wire up the gui / desktop
to the Freebsd functionality for making calls and texts

To be able to recommend a front end "desktop"
you would need to what the drivers and code are used on freebsd to provide the functionality

And then knowledge of the front end "desktop" and what needs to be ported
then how to wire it up to freebsd

Saying you have a Radxa Cubie a7z
wouldnt be enough information to recommend a front end

Realistically people would need access to the hardware
to be able to determine what would actually work as a front end
 
But that still does not mean having a driver.

I don't need to write or port any driver.

At most, I need this:

1. Kernel: a table, not a driver. If the interfaces are vendor-specific, add the VID/PID to usbdevs plus one line U3G_DEV(QUECTEL, EG800Z, 0) in u3g.c and rebuild u3g.ko. Ten lines of data, zero logic. If instead it exposes CDC-ACM, umodem(4) attaches on its own and you don't even touch those. Either way you get /dev/cuaU0.x.

2. Calls: AT over a tty. ATD<number>; to dial, RING + ATA to answer, ATH to hang up. There's nothing to write in the kernel: it's cu -l /dev/cuaU0.2.

3. Audio: a userspace daemon. If AT+QPCMV=1,0 responds, the PCM comes out of a second tty. Read/write those buffers and pipe them to /dev/dsp. A few hundred lines, all in userspace.

The only scenario that would require real kernel work is if the module exposed only UAC and snd_uaudio(4) failed to attach to it — but then it would be a fix to an existing driver, not a new driver.
 
There are 2 things

1) Phone operating system
2) Desktop "theme" for a phone

1 ) So your either trying to port a Linux/Android operating system and merge it with Freebsd
to provide the functionality for making calls and send texts

Different phone operating system only work on particular hardware
so the issue is then getting that phone os to work with the hardware you are using


2) Or you are looking for a GUI / Desktop / Theme to put on top of Freebsd
and Freebsd provides all the functionality for making calls and sending text

However the issue is that desktop themes
will be designed to work another operating underneath like linux or android

So you will then need to wire up the gui / desktop
to the Freebsd functionality for making calls and texts

To be able to recommend a front end "desktop"
you would need to what the drivers and code are used on freebsd to provide the functionality

And then knowledge of the front end "desktop" and what needs to be ported
then how to wire it up to freebsd

Saying you have a Radxa Cubie a7z
wouldnt be enough information to recommend a front end

Realistically people would need access to the hardware
to be able to determine what would actually work as a front end

If you can't provide certainties, you can create hypotheses and make assumptions, basing your reasoning on what I tell you and what you already know. No one asked anything. It's so clear that there was a real lack of will here, because as I told you, many are busy criticizing me because I use an AI.
 
If you can't provide certainties, you can create hypotheses and make assumptions

I did list a few desktops / operating systems
but its impossible to know if any of them would work

That could only be done by you actually testing them on the hardware

busy criticizing me for using an AI.

Its fine if you want to use AI

But the drawback is that you may have difficulties have technical discussions
Because you arent a developer
 
I want a FreeBSD phone that just has dial pad with speakers and a microphone. No need for all these browsers and screens and touch screens and things.
 
Back
Top