Could A New Linux Base For Tablets/Smartphones Succeed In 2017?

[duby229]

Unless you standardize also libraries and other system stuff, or you ship applications and their own libs together, it's not gonna work.
That's what they are doing here. Making a standard base OS

Good point, but version control solves that completely. A pretty basic version controlled runtime would do it.

[Jumbotron]

The end of convergence and the Unity Desktop along with MIR on Ubuntu should make it CRYSTAL clear.....that the answer is an unqualified NO.

Shuttleworth sank 10's of millions, perhaps more, on this dream. But everyone including him forgot this bit of computing history and fact.

We have been collectively computing with 2 computing platforms for the last 35 years. Windows and Mac. And recently Android
Now....the three companies ( Microsoft, Apple, Google ) that produce these OS's and computing experiences are the 3 richest corporations that the world has ever seen. They have MORE resources and MORE money than all of humanity has been able to hoard for the last 25,000 years.

People have used both these desktop OS's and their mobile counterparts for decades. They are used to their paradigms. Because of this one simple fact. One GUI experience per OS and one GUI experience for their mobile counterpart.

People SAY they want choice.....but they are made to be LIARS. We have known this for DECADES from brand loyalty studies and what is seen in the real world. We as a species have SOOoooo much to think about on any given day that we really don't WANT all this choice if it means having to make yet ONE MORE DECISION. And to learn something new....since everyday brings something new for us to overcome is a hard thing. We really don't WANT to do it. And when we have a chance to NOT learn something new...( like a new OS and/or GUI paradigm ) we JUMP at the chance to NOT learn something new.

Linux fragmentation will FOREVER keep it marginalized. Do I use Fedora or Suse or Ubuntu or Debian or Xubuntu or Mate or Cinnamon or Budgie or Elementary or Arch or Gentoo or Slack or GNOME or KDE or Unity forks.....etc...etc...etc....ad infinitum.....ad nauseum.

Between Apple and Google ( Android ) 95% of the mobile world ( tablets and phones ) has been conquered. You will not break this....even with a FREE OS.....if you do not have...

1: ONE SINGLE GUI and UX. Period....end of story.
2: iOS and Android apps that have been perfectly ported to this mythical Linux stack based mobile OS. You quite LITERALLY have to have 1 billion Linux ready apps based on their iOS and Android versions at system launch when this OS is rolled out on DAY 1.

Apart from this.....forget about it. I would say go ahead with this project if for only the experiance that is gained on proper package management which we STILL don't have on Linux after 25+ years. And even there you have useless and needless choice. AppImage.....Snaps....Flatpak. This is bullshit. And this is why Linux will FOREVER be marginalized to the phreaks and geeks.

And I consider myself one. But....this phreak and geek is tired of the immaturity of a lot of the Linux and FOSS community. Which is why I am VERY happy that the CORPORATE Linux world has decided to standardize on ONE Linux GUI and UX. Namely GNOME. This is a BIG step forward into lessening the Linux fragmentation that is HARMING the acceptance and adoption of Linux into the wider computing public.

This is a baby step to be sure. But perhaps with the work on this new mobile stack people will realize that having one unified and standardized stack and packaging system is CRUCIAL to extending Linux's marketshare in both the desktop and mobile world.

[Cerberus]

The should choose snaps for package format and make it convergent, as for UI I dont care too much whether its Unity 8 on Wayland, some KDE UI or something else, just DONT make it a clone of Android visually, Unity 8 was very nice on the phone, slick and intuitive, and didnt look like a clone of other operating systems. Running Android applications would be a bonus, but not absolutely necessary for me, if they can provide at least native Telegram I am okay with that. Common platform doesn't make much sense if they are all going to slap wildly different UIs etc on it. They should pool resources and work on a single project, then it might have a chance of gaining users at least among Linux users. I would buy a convergent phone, but I dont want five different projects doing the same thing.

[Cerberus]

Between Apple and Google ( Android ) 95% of the mobile world ( tablets and phones ) has been conquered. You will not break this....even with a FREE OS.....if you do not have...

1: ONE SINGLE GUI and UX. Period....end of story.
2: iOS and Android apps that have been perfectly ported to this mythical Linux stack based mobile OS. You quite LITERALLY have to have 1 billion Linux ready apps based on their iOS and Android versions at system launch when this OS is rolled out on DAY 1.

They are not planning to take on Google and Apple in any way, they are attempting to create a niche mobile OS alternative.

[Sonadow]

The end of convergence and the Unity Desktop along with MIR on Ubuntu should make it CRYSTAL clear.....that the answer is an unqualified NO.

Shuttleworth sank 10's of millions, perhaps more, on this dream. But everyone including him forgot this bit of computing history and fact.

We have been collectively computing with 2 computing platforms for the last 35 years. Windows and Mac. And recently Android
Now....the three companies ( Microsoft, Apple, Google ) that produce these OS's and computing experiences are the 3 richest corporations that the world has ever seen. They have MORE resources and MORE money than all of humanity has been able to hoard for the last 25,000 years.

People have used both these desktop OS's and their mobile counterparts for decades. They are used to their paradigms. Because of this one simple fact. One GUI experience per OS and one GUI experience for their mobile counterpart.

People SAY they want choice.....but they are made to be LIARS. We have known this for DECADES from brand loyalty studies and what is seen in the real world. We as a species have SOOoooo much to think about on any given day that we really don't WANT all this choice if it means having to make yet ONE MORE DECISION. And to learn something new....since everyday brings something new for us to overcome is a hard thing. We really don't WANT to do it. And when we have a chance to NOT learn something new...( like a new OS and/or GUI paradigm ) we JUMP at the chance to NOT learn something new.

Linux fragmentation will FOREVER keep it marginalized. Do I use Fedora or Suse or Ubuntu or Debian or Xubuntu or Mate or Cinnamon or Budgie or Elementary or Arch or Gentoo or Slack or GNOME or KDE or Unity forks.....etc...etc...etc....ad infinitum.....ad nauseum.

Between Apple and Google ( Android ) 95% of the mobile world ( tablets and phones ) has been conquered. You will not break this....even with a FREE OS.....if you do not have...

1: ONE SINGLE GUI and UX. Period....end of story.
2: iOS and Android apps that have been perfectly ported to this mythical Linux stack based mobile OS. You quite LITERALLY have to have 1 billion Linux ready apps based on their iOS and Android versions at system launch when this OS is rolled out on DAY 1.

Apart from this.....forget about it. I would say go ahead with this project if for only the experiance that is gained on proper package management which we STILL don't have on Linux after 25+ years. And even there you have useless and needless choice. AppImage.....Snaps....Flatpak. This is bullshit. And this is why Linux will FOREVER be marginalized to the phreaks and geeks.

And I consider myself one. But....this phreak and geek is tired of the immaturity of a lot of the Linux and FOSS community. Which is why I am VERY happy that the CORPORATE Linux world has decided to standardize on ONE Linux GUI and UX. Namely GNOME. This is a BIG step forward into lessening the Linux fragmentation that is HARMING the acceptance and adoption of Linux into the wider computing public.

This is a baby step to be sure. But perhaps with the work on this new mobile stack people will realize that having one unified and standardized stack and packaging system is CRUCIAL to extending Linux's marketshare in both the desktop and mobile world.

I would have to say no.
It's not clear what "succeed" means, but the market has already sent a strong signal it doesn't need yet another mobile OS.
Windows Phone 7 (and later) were generally reviewed favourably and still tanked. Joola is supposed to be pretty good and still going nowhere.
So what's the proposition here? Common platform? That's what AOSP is supposed to be. Don't like Google's approach of closing as much as possible? There's intel's Yoctos project. But neither has yielded any groundbreaking mobile OS so far.

For a new mobile OS to succeed, it needs to do two (and only two) things:
1. Enable users to do something no other mobile OS can do.
2. Convince hardware manufacturers to write drivers for the new OS.
Sadly, neither is a task an open source team can see through. Maybe if they get lucky the can solve #1, but no way they can approach #2.

Windows is actually a very sad case. Before iOS, they had the defacto mobile OS in Windows Mobile and Windows Pocket PC. Even Nokia's attempt at a smartphone could not unseat Windows Mobile back then.

And if Microsoft with all its cash in the bank could not turn Windows Phone 7, 8, and Windows Mobile 10 into a success, there is no chance an unheard-of vendor can do so.

And as someone has already mentioned, consumers are actually idiot liars. They demand and insist on choice, but back down and complain when they need to make a decision. That's why the world can only accommodate no more than two of anything (especially global / defacto standards), with the rest being nothing more than fringe options for the adventurous. The mobile OS has already been set with Android being the clear winner and iOS the runner up, and Windows Mobile at a very distant third that is close to disappearing to irrelevancy.

And even within the Android space, the only real vendor is still Samsung, with Sony, LG and HTC playing catchup. Only in Asiapac do other brands like Xiaomi, Oppo, Oneplus and Lenovo have a sporting chance in the market as well.

It does not matter if the new mobile OS is better, superior, more beautiful or more user-friendly than Android; the world has already chosen to reject it simply because it cannot get the traction needed to stay relevant. And you can bet the OEMs and vendors will not hedge their bets and waste money on supporting a new OS when they see their profits slipping year after year despite their futures are tied to Android.

[bug77]

On 2: The linux kernels hardware support gets better every single day. Plus it's been common knowledge for decades now that if you want a linux system you must choose hardware that is already supported. Basically if you take even just 2 seconds to think about it you will realize that hardware support for linux is the very best hardware support in the world, Period. Just choose supported hardware. It really is that simple.

You don't have the luxury of choosing hardware for mobile. You get to pick from Qualcomm which is open source hostile and MediaTek and others which are worse.
I'm not sure why anyone thinks in 2017 closed drivers are a competitive advantage, but that's how it is

[duby229]

You don't have the luxury of choosing hardware for mobile. You get to pick from Qualcomm which is open source hostile and MediaTek and others which are worse.
I'm not sure why anyone thinks in 2017 closed drivers are a competitive advantage, but that's how it is

In a limited sense that is true, but practically it's wrong. There are so many drivers being developed without the manufacturers support. Qualcomm is a perfect example, they are actively hostile to open source development, but regardless of that a good proportion of their hardware works surprisingly well with open source drivers. Although, I personally don't like Mediatek, the same thing is true for the majority of their hardware too.

I say screw them, they may not want you to use open source drivers, but there is nothing that can stop you from doing so.

[starshipeleven]

Good point, but version control solves that completely. A pretty basic version controlled runtime would do it.

Heh, pretty basic my ass. You need to run applications with specific library versions without other applications conflicting AND without having to change all the applications you take from upstream to work in your own system, that's not trivial.

NixOS does that (also Guix), for example, but it is not exactly mainstream.

Most seem to have gone the easier route, shipping apps + libs in the same package.

[starshipeleven]

People SAY they want choice.....but they are made to be LIARS. We have known this for DECADES from brand loyalty studies and what is seen in the real world. We as a species have SOOoooo much to think about on any given day that we really don't WANT all this choice if it means having to make yet ONE MORE DECISION. And to learn something new....since everyday brings something new for us to overcome is a hard thing. We really don't WANT to do it. And when we have a chance to NOT learn something new...( like a new OS and/or GUI paradigm ) we JUMP at the chance to NOT learn something new.

Please note: the studies tell us how the average (fucking dumbass) Joe acts, not how EVERYONE acts.

The entire existence of Linux Desktop does show that there is people that want choice.

Linux fragmentation will FOREVER keep it marginalized.

Yeah, as it caters a market niche that is different from the one of Windows or Apple users.
Is that even wrong? Linux does not need to dominate, only to get recognized enough. Apple is recognized enough and it is at like 3% at most of marketshare (PC).

Between Apple and Google ( Android ) 95% of the mobile world ( tablets and phones ) has been conquered. You will not break this....even with a FREE OS.....if you do not have...

They don't want to compete with Android, they just want to cater to the same kind of niche market that is using Linux Desktop.

1: ONE SINGLE GUI and UX. Period....end of story.

Nope. See above. The idea is to provide the same base for 2-3 different UIs.

2: iOS and Android apps that have been perfectly ported to this mythical Linux stack based mobile OS. You quite LITERALLY have to have 1 billion Linux ready apps based on their iOS and Android versions at system launch when this OS is rolled out on DAY 1.

Anbox will deal with this for Android apps, I assume this will be enough.

But perhaps with the work on this new mobile stack people will realize that having one unified and standardized stack and packaging system is CRUCIAL to extending Linux's marketshare in both the desktop and mobile world.

Unifying the stacks is not possible on Desktop as each distro has its own agenda and values and so on. If you unify stacks you basically kill them all and make a single OS.

Quite frankly, it's not better, and it won't help Linux adoption. What would help Linux adoption is some company that starts selling PCs with their Linux distro on it.

[duby229]

Heh, pretty basic my ass. You need to run applications with specific library versions without other applications conflicting AND without having to change all the applications you take from upstream to work in your own system, that's not trivial.

NixOS does that (also Guix), for example, but it is not exactly mainstream.

Most seem to have gone the easier route, shipping apps + libs in the same package.

That's true only because a version controlled runtime hasn't already been made standard. Somebody is going to have to bootstrap the process. That starts by version controlling a runtime and then getting developers to standardize on it.