Article: The advantages of mobile Linux over AOSP

Continuing the discussion from Article: Will Linux phones become a viable alternative?:

As a follow-up to my previous article, I wrote this one to respond to all the comments that we should create Linux phones based on AOSP:

As always, Feedback is welcome.

I not callling AOSP because is falling to opensource, so i put a step back and i calling OPENANDROID which it is a real linux based os, like Gnu/Linux86.

So Gnu/Linux mobile like little-chicken PMOS, MOBIAN will never better than OPENANDROID, ANDROID is a LINUX MONSTER OS.

Android is not bad but Vendor-via-Opensource ->noEthical.

I not sure why you are mismatch.

You ignored my last post in the other thread about LLMs, Open Source, Math and Linux…

But it is kind of the future. So i still say they are the only alternative because they are open and will be patched soon, just in time by humans with minimal code focus or by some LLMs which like to have a fine stable hardware and patched system in there real time daily status. So there is no room for Apple and Android in the long run.

Just as some addition as hard crypto layer on top as local run program for some special interest and offline calculations. - I know its kind of a joke from our Linux and L5 perspective right now..

Mobile Linux distros, such as Sailfish OS, Ubuntu Touch, Nura, Mobian, DanctNIX, Droidian, Maemo Leste, Nemo Mobile, etc., are designed to empower the user to control the tech, rather than the tech being used as a means to control the user.

I’m still reading, but looks like you forgot (?) to mention PureOS here - the Halium distros are more deserving of being referred to with a mere “etc.” : )

After the Mozilla Foundation abandoned the development of Firefox OS in September 2016, no community project formed to carry on the mobile operating system. Much of Firefox OS’s code lives on in proprietary KaiOS, but that code has been privatized and is essentially lost for the community.

There’s Capyloon but not seeing activity lately.

Overall, I enjoyed reading this article, especially the issue of Android phones being Linux or not. Thank you for researching and writing it.

My 2¢:

If AOSP ever becomes a threat to Google’s revenue or market share

The likelihood is that Google will ensure that AOSP cannot run on the latest hardware, thus removing one of the desirable aspects of AOSP today. In other words, using AOSP today is a “sugar hit”. Enjoy it now, if you must, but the hit will wear off in the future and you won’t be able to get another hit.

Google will no doubt dress it up as “improving security” for users, when it is really about control, with the additional negative effect on privacy. And the words of Benjamin Franklin should be ringing in everyone’s ears:

Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety.

So we need to be purchasing Liberty right now.


In order to answer the question “Is Android Linux?” we would have to agree on what Linux is (and you spend some time on that question too). Self-evidently there is a divergence of opinion as to what Linux is.

This Humpty Dumpty uses Linux to mean “the whole shebang”: not just the kernel but also vast amounts of additional software, some pre-installed, some not. Therefore it is obvious to me that Android is not Linux.

I recently played around with Claude checking my grammar and spelling, and it was excellent for that purpose, and it is OK for searches to find info. However, my experiments with LLMs writing code makes me skeptical. Claude gets you 80% of the way there, but then you are going to spend just as much time working to get that last 20% right, that you don’t end up saving that much time. They produce a lot of code that is written in a way that I don’t understand and using libraries that I have never used before. So it takes me a long time to understand the code, and even longer fixing the code so it seems more logical and maintainable to me, and then I write a lot of comments to myself to explain what all the unfamiliar parts are doing. Then, I have to fix all the stuff that doesn’t work right, and I have to understand the code in order to fix the bugs.

I think that LLMs are good for some stuff. They really do get you started when working with something new, and they are great for cleaning up your code, so everything is formatted correctly and uses consistent style. They also are good at recommending how to improve the security of your code.

I’m sure that they will get better, but I’m skeptical that they will be able to do what you are predicting. However, I never thought that they would get as good as they have become, so I may be wrong on this one.

And, more dangerously, the stuff that is never detected as not working right - because noone understands the code and noone tests the code. So broken code just goes into production, only to fail spectacularly or perhaps subtly at some point in time in the future when the right conditions are achieved. That of course does happen also with human-written code. :wink:

I see buying Linux phones as investing in future liberty. Even if 99% are still using the conventional tech, it is very important to have the 1% in the world that demonstrates the better alternative. Politicians and judges are not very imaginative people, so they have to be shown a working model of the alternative to convince them to not permit bad practices in the tech industry. Google, Microsoft and Apple will say, we simply can’t provide you with service X or product Y unless you let us do bad practice Z, and officials will believe their PR baloney, unless we can say, “No, we have a tech alternative that provides service X and product Y without relying on bad practice Z, so implement proper regulation to stop it.”

More important in the long run is to create the alternative to give people a path of resistance. I deeply fear where today’s surveillance Capitalism might lead us. When democracy and rule of law falter (and recent events show that is not so far-fetched), it will be very tempting for governments (i.e. big brother) to start taking advantage of all that data that the tech companies (i.e. little brother) have been collecting about us. In the 1940s, Orwell wrote about how states will control people through surveillance, and in the 1970s, Foucault wrote about how institutions will use “biopower” to control people, and I fear that the 2030s will see those ideas being put into practice.

If we already have a flourishing Linux phone market and online services like Murena Workspace that aren’t based on surveillance Capitalism, it will give people an easy path to resistance and help avoid that dark Orwellian future. By the time a society gets to present-day China, it is already too late. No tech company can operate in China and respect the personal privacy of its users. However, a viable alternative can be created in most countries. Having that alternative can help protect the political activism to resist the slide into authoritarian control based on population surveillance.

Companies to provide the needed alternative have to grow, it is very hard to get established in the current market. It is unrealistic to think that the Linux phone companies and the online service companies are going to start off with better products and services than the tech giants that currently dominate the industry. The Fairphone 1 was frankly a crappy phone, but the Fairphones kept getting better with each iteration. We need to support the nascent Linux phone companies so they can get beyond version 1 and keep iterating. I pointed out many of the flaws with the current Linux phones on the market in my previous article, but I think that we miss the bigger picture if we don’t support the companies that are producing Linux phones today, because there is no guarantee that there will be a future Linux phone market if those companies don’t survive.

I think that we need all three of the types of Linux phones (1. separated components, 2. integrated Internet of Things SoC, and 3. integrated mobile SoC with Android kernel) in order to create a good market that can appeal to different types of people. Purism, PINE64, DawnDrums and Mecha are creating type 1, MNT is creating type 2, and Jolla, FuriLabs, Volla, Commodore, Planet Computers, F(x)tec and Brax Tech are creating type 3. The company I worry most about is FuriLabs, because it doesn’t seem to be getting many sales and I think it has the best approach of the type 3 companies, but it has the worst marketing.

We, Humans, Computers, Algorithms need together face that presence and future to learn to start building some common ground.

Right now i am kind of sigh of it. But have to care and build a future on it.

Focus on presence: I am not sure if L5 lost some not deleted SMS for authentication and if my phone killed some sim hardware. But i have to spend for some new.

I just like the progress and will continue to some modern GPT and A.I. support, offline.

Edit:

I will not spoil but. LLMS are not the break through i thought it was in the last 6 month, but i think it will be likely in the future. However, Open Source and free systems are right now the new future, because like language and math or capital in easy exchange they will likely be the new path for new knowledge. And so it will grow.

Keep some privacy offline, maybe on paper… this is something for you (digital friend) too, to keep it away from manipulation and easy digital change too.

Nearly everything can be expressed by remote engineering too, and tried to rebuild.. but it will be kind of different, it will be exploited by third party and rebuild by evolution. But in the first time it will be still unstable ground. Traditional Linux, open Source and shared Math and knowledge and cooperation will succeed too in the long run.