July 2018

>>61926746
A sex organ?

>>61926746
looks like chrome os just for my goddamn phone, the last thing i needed

>>61926746
>>Nothing works without Google micro services, and everything depends on the Google app.
my hacked-together attempt at making it work without being able to talk to Google. It includes a crawling horror of a DNS resolver that sometimes doesn't work when I want it to and mostly but not always doesn't work when I don't want it do, with a bunch of semi-hardcoded Google domains hardwired to fail or give fake responses. There's a webserver on localhost that exists to return errors and the occasional 204 No Content to the Google requests that get pointed at it. The driver for the GPS hardware has an option added to it to always make it return the same coordinates whenever its called.

>>61927125
fpbp

What is this OS for? Is it to replace android? I thought android had a solid base with the Linux kernel etc.

File: Screenshot_20170816-034349.png (129KB, 1080x1920px) Image search: [Google]

129KB, 1080x1920px

>>61929882
>What is this OS for? Is it to replace android? I thought android had a solid base with the Linux kernel etc.

If Android was solid, my Nexus 5 hammer head would be on something newer than 3.4

>>61929994
TBF, that's surely to do with Lineage, rather than Google at this point?

Looks like I'm sticking with my Nexus 5 until mid-2018 to see what this is. Don't want to be buying an Android now if it's going to be replaced entirely.

I'm genuinely curious what Fuchsia is going to be.

>>61929882
The problem with Android is that Google can't control it. Updates are controlled by carriers and Google doesn't have the leverage force the carriers to push out updates immediately. (inb4 'just root your phone' - that's certainly an option but not user friendly, especially to the normies. Google isn't trying to appeal to /g/.)
There's also the fact that OEMs keep fucking with it and implementing weird-ass design languages that clash with Material Design, which leads to a lot of visual inconsistency and confusion.
Not to mention all the bloatware and lag.

Fuchsia aims to fix all of that. Fuchsia is sort of Google's 'do-over.' I don't know how they're going to implement it but essentially it's going to work like Windows Phone in a way; OEMs can manufacture phones, tablets and computers that use Fuchsia, but are not allowed to actually alter Fuchsia in any way, so every Fuchsia device works the way. They're also going to be a lot more tight-assed about bloatware and updates.
There's also a good chance Fuchsia will run on everything, which will allow Google to achieve MacBook/iPhone-like continuity where you use the same apps on your phone and computer.
I think Google is also going to try to kill JavaScript apps with Fuchsia as well.

I'm not looking forward to Fuchsia's unveiling because it's probably going to be chaotic. They're going to try to replace Android but the problem is that virtually no companies have any real reason or incentive to join in on this bandwagon. It's going to be really interesting to see what they're going to do with Android apps as well; will Fuchsia be backwards-compatible or will we just have to start over from scratch?

>>61929994

Why aren't you running CopperheadOS?

>>61930255
Do you expect Android Apps to run on Fuschia?

>>61930262
Nexus 5 =/= Nexus 5X

>>61930276
This one's up in the air. If you're going to make an entirely new OS from scratch, backwards-compatibility sort of defeats the purpose.
On the other hand, there are literally millions of Android apps in the Play Store, so if we just abandon all of them like that, Fuchsia will fall flat on its face.

>>61929994
Get an S8, it's on 4.4.16

File: Screenshot_20170816-042235.png (87KB, 1080x1920px) Image search: [Google]

87KB, 1080x1920px

>>61930262
>>>61929994 (You)
>Why aren't you running CopperheadOS?

I don't know much about copper head os. I just like TWRP backups and easy updates.

>>61930255
>the problem is that virtually no companies have any real reason or incentive to join in on this bandwagon.

This. No company would get in on Fuchsia if they can't tinker with it. For Fuchsia to be successful, Google would have to manufacture the hardware that comes along with the OS, sort of how Apple have done.

But for some of us, we want Google to drop Fucshia so Samsung can answer it with Tizen.

>>61930421
Actually I think companies are going at some point to jump in ( if it gets concrete ); better design, not having to develop a new design on your own, better overall user experience
But perhaps some companies will prefer continuing to develop their own custom android, like xiaomi with their miui for example

>>61926746
That logo is way too phallic. How do I send a complaint to Google?

File: f49.png (478KB, 680x689px) Image search: [Google]

478KB, 680x689px

>google fuchsia
>google fuchs ia
>google fucks you
yeah, not gonna fall for this

File: Screenshot_20170816-132239.png (132KB, 1080x1920px) Image search: [Google]

132KB, 1080x1920px

>>61929994
Android N doesn't officially support SD800, all ports are hacks. It's time to upgrade.

>>61929882
It's to replace both android and ChromeOS so it can run on desktop and mobile. It will basically be windows 10 just open source.

Is this going to be the ultimate botnet or what?

>>61930911
idiot, orchid = vagina

>>61931548
since when?

File: googlefuschiaglass.png (296KB, 542x371px) Image search: [Google]

296KB, 542x371px

>>61931500
This is what I'm gonna be concerned about. Unfortunately unless you buy Apple's shitty hardware you don't get their actually-okay software. But if you buy the good Android hardware you get google's shitty spying software.

The consumer just cannot. win.

>>61932183
>Apple
>okay software
This is bait?

File: desperation2.jpg (45KB, 580x369px) Image search: [Google]

45KB, 580x369px

>>61926746
>yet another kernel fork by jewgle with their branding and walled garden slapped on top
>https://github.com/littlekernel/lk/wiki/Introduction

>>61932318
Of course not, OS X is great.

>>61932318
I said it was okay, I didn't say it was _good_.
Still have to admit the anti-fragmentation that iOS has compared with Android makes it a stronger operating system (less confused UI design, less chance of "rogue" processes and apps under a closely scrutinised walled garden), but with the App Store instead of Play it's a weaker for the diversity of apps you can use. I know it's a "muh games" thing but I can't go and buy an iPhone, get PPSSPP and play PSP games with my Dualshock 4 controller paired and clamped to the phone without a whole lot of jailbreaking. Android lets you do that easily, with the risk of not every app / mod of the OS being carefully checked for shittiness.

What was ever wrong with Symbian. It just worked with no need to constantly all home. Freakin' easy access to internet has turned software devs evil I swear.

>>61932324
Fork? I thought this was entirely new. Please don't tell me it's just been a linux fork this whole time.

>>61932481
But android is better than iOS in every way, as an OS. They even fixed the java caused lag a few years back. It's only flaws are:
>carriers have a say in updates instead of manufacturers, unless you buy the phone directly from the manufacturer
>phones come locked and unrooted. That's like using Linux without sudo, or using windows without admin privileges.
Neither of these are "android's fault". It's the fucking companies that cuck users and Apple is no different.

SailfishOS and Fuchsia look more promising than android and iOS combined.

>>61932581
>unless you buy the phone directly from the manufacturer
Who the fuck doesn't do this? I thought you murricans worked even this out years ago

>>61932579
No, it's from scratch

>>61932579
It's a fork of a really small, minimal kernel that's not linux.

>>61932648
>>61932662
Then why complain? No-one would worry if they brought back some dead fucking experiment from 20 years ago.

>>61932606
I'm not an American. And I do buy directly from the manufacturer. But most people don't.

>>61932683
They wrote it in about a year, huge parts of it in Go. It's not going to be pretty.

So is Fuchsia going to be a totally new thing? Will it be able to run old Android software?

>>61932691
In Go?
Are you fucking kidding? I thought google only bothered with go for the script kiddie tier interface
designers.
>>61932706
>Maybe we'll make it posix enough that we'll be able to shimmy it in with a compatibility layer underneath android?
>Oh wait, no drivers, never mind. Maybe in 20 years...

>>61932719
That was their goal with go. Hire as many cheap code monkeys as possible and give them the highest chance for any code that they manage to compile to not be shit. So they took an existing operating system and put a thousand code monkeys on it.

>>61932706
It's not Linux based, it's not android based. It will be difficult running android apps on it unless it gets a compatibility layer, but the apps would generally run worse then. Like how shit performance was on pre-Lollipop android.
Afaik sailfish comes with android compatibility despite not being an android OS. But it is still Linux based so I'm not sure if fuchsia has a chance.

>>61932807
Didn't they say they didn't want to replace android and had other goals for it?

>>61932852
They literally said it will replace both android and ChromeOS, and will be the next multiplatform OS to compete against windows 10

>>61932807
>but the apps would generally run worse then. Like how shit performance was on pre-Lollipop android.
How did you come up with this nonsense?

It's built around a microkernel.
Microkernels are the best systems available.
http://blog.darknedgy.net/technology/2016/01/01/0/
It's going to run laps around leenoks.

>>61930276
yeah actually, they can easily make an ART kernel module for fuchsia to ease the transition, and in the mean time give a really easy toolkit to convert apps to fuchsia native (using something like flutter, which already exists and works with android and iOS)

>>61933023
>not exokernels
Nah.

>>61933126 >>61933107
>falling for the exokernel meme
kek

>>61933046
>kernel module
They don't need kernel modules. They have a pure microkernel architecture.
>art
Is just a glorified java runtime. It could be ported to fuchsia just fine.

>>61933013
What nonsense?

>>61933183
>but the apps would generally run worse then. Like how shit performance was on pre-Lollipop android.

>>61933196
What's nonsensical about that statement? Emulating android apps would definitely have shit performance than a natively installed app.

>>61933227
>Emulating
That's what the core of the nonsense is.
How the hell did you come to believe emulation is needed?

>google switching to their own OS to avoid GPLs with using Linux for Android

Will this throw back Linux (the kernel) to "obscurity"?

>>61933023
>Microkernels are the best systems available.
HAHAHA! Oh wait you are serious? let me laugh even harder.
HAHAHAHAHAHAHAHAHAHAHA!!!

>>61933481
>doesn't reinforce his present beliefs
>dismisses whole article without reading
Way to go, Anon!

>>61933445
it's not. Google is not such a great contributor as people tend to think.

>>61933491
Microkernels are OOP of Kernels.
Everyone thinks it's great idea at first but then they realise how shit it is.

>>61930421
>Samsung can answer it with Tizen
>Tizen

Are you fucking kidding?
https://arstechnica.co.uk/gadgets/2017/04/samsungs-tizen-is-riddled-with-security-flaws-amateurishly-written/

>Presenting at Kaspersky Lab's Security Analyst Summit and speaking to Motherboard, Neiderman had little positive to say about the state of Tizen's code. "It may be the worst code I've ever seen," Neiderman said. "Everything you can do wrong there, they do it."

>While much of the code is inherited from Tizen's Intel and Samsung predecessor projects, Neiderman says that most of the flaws he found were in the newer code. Buffer overflows are widespread due to issues such as the improper use of the strcpy() function in C—a notoriously dangerous function with risks that are well known to experienced C and C++ programmers. These risks lead many developers to use alternative functions entirely, but not so the Tizen developers: Neiderman says that Samsung is "using it everywhere."

>Samsung's code also failed to use SSL in a consistent way, transferring even sensitive data in the clear.

>>61933391
It won't be run natively, hence it's emulated.

they should have let us do the logo

>>61933499
I mean usage wise. A great part if not the biggest part of devices running Linux in any form are Android phones/tablets.

>>61933516
If that was true, nobody'd be using them.
But the article shows that's not the case.
>>61933530
>It won't be run natively
How is a java program more native in Linux than, say, in Windows?

>>61933542
>If that was true, nobody'd be using them.
Oh yea, nobody uses OOP at all, it's not most popular way to write code.
Are you retarded?

File: blackberry-10.png (537KB, 800x799px) Image search: [Google]

537KB, 800x799px

>Runs QNX RTOS microkernel
>Super stable and great battery life
>Has the ability to install Android applications
What went so wrong?

>>61933556
Why do you think microkernels are so popular where they are?

>>61933542
See dalvik vs ART. That's obviously what I meant. Stop baiting (you)s

>>61933569
Because they are abstract like OOP languages.
How many times do I have to say LIKE OOP for you to understand my point?

>>61933582
>See dalvik vs ART. That's obviously what I meant. Stop baiting (you)s
They're both glorified java runtimes.
Why do you think it's not possible to port them or to build a better one on fuchsia?

>>61933540
there's probably as many servers, routers, and other embedded devices running linux as their are phones/tablets

>>61933583
>How many times do I have to say LIKE OOP for you to understand my point?
I don't see the analogy. Analogies only work if the other party agrees to them.

>>61933622
Well it shows that you don't understand how Microkernels work and you should go read about them in more than one source and actually learn how do they work.

>>61933653
>unironically believes millions of lines of buggy code running in supervisor mode is the best we can do
>refuses to read article
>http://blog.darknedgy.net/technology/2016/01/01/0/
>that argues the opposite point
Way to go, Anon-kun.

>>61933664
Again, go and learn how Microkernels work, this will open you to more arguments other than what you read from that particular blog.
There is no right or wrong way of doing kernels, there is how ever wrong ways of presenting your argument, and only knowing a biased opinion is one of them

>>61933681 >>61933653 >>61933516
>http://blog.darknedgy.net/technology/2016/01/01/0/
>Microkernel hatred is a peculiar phenomenon. Sheltered users who have never had any background in much beyond Windows and some flavor of free monolithic Unix, will, despite a general apathy or ignorance in the relevant subjects, have strong opinions on the allegedly dreadful performance and impracticality of “microkernels,” however they define the term (and we shall see that a lot of people have some baffling impressions of what a microkernel is supposed to be).
Holy shit, is that article's intro accurate.

>>61933708
>I argue with you
>You don;t understand me
>I advice you to learn about the subject
>You go full fucking retard

Stop, you are embracing yourself.

>>61933717
>hating this hard on an article he hasn't read
Anon, you can do better than this.
>Stop, you are embracing yourself.
Much embrancing.

>>61926746
RIP Linux on Android.
Android's kernel will be replaced with that thing in under 2 years.

>>61933681
>Again, go and learn how Microkernels work, this will open you to more arguments other than what you read from that particular blog.
You have no real answer for this.
>unironically believes millions of lines of buggy code running in supervisor mode is the best we can do
Do you, anon?

>>61933755
I didn't hate on the article, I hated on your knowledge of the subject, its very biased and not very informed.

>>61933781
Anon-kun, you seem very knowledgeable in the subject, and I'm honestly curious about this. Why do you
>unironically believes millions of lines of buggy code running in supervisor mode is the best we can do
?

>>61933766
So the Linux development will die before windows 10 dies. Fucking lel.

>>61933818
you do not understand that Monolith is and what Mico kernel is to ask question.

Unfortunately for you even micro is using buggy code running in supervisor mode.

i didnt know Fred Fuchs is developing an OS

>>61932742

Honestly not a bad idea for a language. Java and C# was the same idea, but they didn't really execute as well.

>>61933890
>Unfortunately for you even micro is using buggy code running in supervisor mode.
Great, find me a bug in the current verified version of seL4.
protip: You can't.

>>61933916
https://github.com/seL4/seL4/issues/53

Just go to their github, bug reports are everywhere. But you miss the point of why sel4 is "bug free"

again showing your lack of knowledge anything computers.

>>61933766
it will if all manufactures will be okay with Google fucking them in the ass.
>>61933862
Why the death of Android will stop Linux development? I think it will just stop Android development. And why do you compare a desktop os like win10 with Android?

>>61933974
>in the verified version
Try again, anon-kun.

>>61933990
every bug in github is part of verified version today. Again showing your general lack of understanding of software.

I just want a new desktop OS that is actually good to step up its game. Something to finally remove Windows from it's throne. Also remove POSIX and UNIX/UNIX-likes.

>>61933974
Do you unironically believe it's not easier to keep 5k lines of code bug-free than to keep 1m lines of code bug-free?
>>61933996
Definitely not the code you linked to.
I do not know much about your general understanding of software, but your lack of understanding of seL4's development process runs deep and fundamental.

>>61931254
how do you go from fuchs to fucks ?

>>61934018
>Also remove POSIX and UNIX/UNIX-likes
Why? Unix-like systems are very flexible. They can be anything you want them to be.

>>61934053
Ch frequently makes a k sound. Chrome is a google related example.

>>61933566
>Has the ability to install Android applications
this feature came like right when they decided to abandon the project entirely. If they'd stuck to their guns then they would've had decent marketshare.

>>61934323
Sometimes you just have rest to start over.

>>61934323 >>61934430
Any system that's not trying for a niche does have to at least do POSIX.
Else it's a non-starter.
Minix3, Genode and Fuchsia all do provide some degree of POSIX compatibility.
Even HelenOS, which is trying to specifically reinvent that wheel does have some provision for actually running POSIX software.

>>61927287
This is what I imagined it would look like, engineered for tablets so everything is a hamburger menu and impossible to navigate with.

I also imagine it's just the beginning of the walled garden, after they con everybody into using Fuchsia it will begin to lock itself down and heavily filter everything you do and see "This pdf does not have a DRM signature, you may not read it"

>>61934659
>I also imagine it's just the beginning of the walled garden, after they con everybody into using Fuchsia it will begin to lock itself down and heavily filter everything you do and see "This pdf does not have a DRM signature, you may not read it"
Fuchsia is open source like Linux. If they wanted to do that, they could just do it on Linux. I don't get your argument.

>>61934784
>BSD 3-clause
Nah, it's nothing like Linux

>>61934839
It's open source all the same.
More free than Linux, though.

>>61934851
>BSD
>More free than GPL
Delusion is hell of a drug.

>>61934860
It has more freedom. You're free to make a blob out of it, distribute it and keep your modified source to yourself.

>>61934871
And the end user won't be free. So it's shit.

>>61934871
>It's more free
>To be closed source

Yea, thanks for proving my point

>>61934893 >>61934900
Allow me to interject...
GNOO!

>>61933542
no one uses micro kernels
Windows is a monolithic kernel
XNU is a monolithic kernel (hybrid kernel is a marketing term)
Linux is a monolithic kernel
FreeBSD is a monolithic kernel
OpenBSD is a monolithic kernel
only netbsd is microkernel but they are cool dudes so no one in the industry likes them.

>>61934977
>only netbsd is microkernel but they are cool dudes so no one in the industry likes them.
Netbsd isn't a microkernel -_-
I'm pretty sure of that one, there's patches from me in netbsd mainline.
However, Minix3 is a microkernel-based replacement (microkernel + system servers in userspace) of netbsd's kernel.

>>61934621
Supporting POSIX in a new Operating System not based on UNIX is in my opinion pretty bad, this would limit yourself to all kind of crap from the middle ages that have no place in a new and modern system.

>>61934977
>Windows is a monolithic kernel
hybrid kernel

>>61935433
Supporting doesn't have to mean building the whole system around it.
Minix3 has chosen to go that way (implement posix-compatible system with new design). HelenOS, Genode and Fuchsia have chosen to just support POSIX by having some compatibility glue.

>>61935433
>Supporting POSIX in a new Operating System not based on UNIX is in my opinion pretty bad
>not wanting to be able to run the millions of applications and libraries that require POSIX

>>61934018
I'd rather keep windows on its trhone if google is the one taking its place

>>61933566

fucking millennials don't care about superior software, they just want to use snapchat to send dogface photos.

>>61935817
almost nothing of value will be lost if it doesn't support POSIX. And the things of value that do get lost can just be rewritten.
>>61935896
I agree with you though, sadly we will never get a good os.

>>61935896
I'd rather windows fails no matter what.

It's taken a long time, but ambitious microkernel-based systems are finally popping up left and right.
Finally the time to leave the labs and take flight.

>>61926746
>lincucks/androidtards opposing innovation again
>>61930255
> MacBook/iPhone-like continuity where you use the same apps on your phone and computer
you dont actually run the same apps on macbook/iphone

>>61933566

BBOS10 was an absolute masterpiece

>>61926746
https://en.wikipedia.org/wiki/Google_Fuchsia
So, they've finally gotten done with hacking their acquired operating that wasn't meant for embedded systems or meant for Real-time interaction... Now they're leveraging Real-time Open source linux kernel components

Took them long enough.
> mfw when brainlets never understood that Android's UI was laggy because android isn't a RTOS and was hacked into one.

>>61936046
Unlike Magenta/Fuchsia/seL4/genode/minix3/helenos.
QNX and BBOS are not open source (OSI or GNU definitions) So we can't fork it. It's lost.
At least, a lot is known about it (lots of papers published), and the source code has been viewable at times, so it's not 100% a waste.
We can do better these days anyway; The seL4 microkernel and the genode system are much better designs. Third generation microkernel systems (2nd is L4 era, 3rd is L4-like with capabilities, not to be confused with that capabilities thing Linux has) are superior in every way.

>>61936061
>Android's UI was laggy
Unfortunately, Linux also sucks for workstation use. Latency spikes, and multi-second I/O stalls have plagued it forever.
The good news is they recognized there's a problem (last autumn) with the I/O stalls. The bad news is there's no fix, and the problems are fundamental issues with Linux design; It's not going to be easy for them.

>>61935446
"hybrid" kernels are a marketing gimmick what they really mean is "monolithic kernel that looks like a microkernel that runs everything in ring-0" which is still monolithic becuase it's still all ring-0.

>>61937174
>"hybrid" kernels are a marketing gimmick what they really mean is "monolithic kernel that looks like a microkernel that runs everything in ring-0" which is still monolithic becuase it's still all ring-0.
Parroting Linus this hard.

>>61929994
Blame Qualcomm for that dumbass.

>>61930421
>wanting samshit anything, much less their software
jesus christ

>itt people are still believing this is more than a lk fork with google branding on top of it
fuck, is nu-/g/ bad... stop believing the lies from jewgle and building hype for them for free...

>>61936133
which is why non-cheap asses customize it.
RT-linux is a thing and is commercially available.
RT-kernel patches exist as well.

Embedded systems (at a certain scale) is dominated by linux based kernels. Google was behind the game like they always are and launched a ragtag product on a non purpose built code base. Now many years later, they've finally managed to allocate some resources to a proper code base. More than likely this project will be announced like it already has in 2016, they'll launch shitty half assed products and it will fail.

> mfw groupthink ruins your company and any product you touch
> mfw when you fired all the old people who would have told you not to do such shit
> mfw you also got rid of all of the people with dissenting opinions telling you how to make a proper product.

>>61938811
>RT-linux is a thing and is commercially available.
>RT-kernel patches exist as well.
rt-linux isn't very good. It does only offer soft-realtime.
For hard realtime, their implementation is a cop-out; Just run realtime tasks on a dedicated CPU, separate from Linux.
Serious realtime is dominated by microkernels. They make it feasible to offer actual guarantees.
An example of this is hybrid manned/autonomous military helicopters using seL4 + eChronos.

>>61938811
>> mfw groupthink ruins your company and any product you touch
Google has a serious SJW problem. I fear the consequences this will have in the world, because Google is a quite powerful company.
>they'll launch shitty half assed products and it will fail.
Fuchsia has big names working on it. Not fresh out of school kids, but experienced OS developers, including those that built the base of BeOS back in the day.
While some sort of commercial failure is feasible, I expect the system itself to become really good.

>>61933520
>tizen being this bad

We need a fork, stat!

>>61930255
>I think Google is also going to try to kill JavaScript apps with Fuchsia as well.

I'm not really sure how that would be possible or if it is even a good idea.

As long as you can load webpages inside of an app then you're going to be able to write hybrid apps. And removing the feature would be a big problem because so many apps rely on it, often not for the main UI but for certain views.

Even if it were possible I don't think it would be a good idea. I think native apps are far superior from a usability perspective (JS is slow as shit - even on basic apps the speed difference is noticeable) but there are cases where hybrid is the only viable option.

>>61926746
>>It's totally great.
>>Everything is sandboxed, you included.
>>Nothing works without Google micro services, and everything depends on the Google app.
>>Now there are many forks. Yours is the seventh.
>What does it look like?
Soooooo gay
Pink + Purple == Fuchsia (a new Operating System)
It will convert phones in iot devices
Trying to replace linux for closed source

>>61941041
>Trying to replace linux for closed source
Fuchsia is Open Source.

The magenta kernel is superb. Very fast and light. Its based on a kernel that was initially written for microcontrollers. It's even smaller than the MINIX or uClinux kernels.

File: 1389222039487.jpg (66KB, 520x363px) Image search: [Google]

66KB, 520x363px

ok but what happens wiht all the adroid phones wiht 8 gigs of RAM

>>61941158
It's not among the worst, but it's a shame they patched it together from LK, instead of using an existing solid, reliable microkernel like seL4.

>>61941219
I appreciate sel4 but im afraid its too opinionated for what google wants to do. LK offered them the flexibility they need

>>61941244
Probably being avoided because
>they don't want to deal with capabilities, perceiving them too complicated
>they know LK well, as they've used it in the android bootloader before
Using an older design (building a 2nd generation microkernel, instead of using a 3rd) should allow them to get it up and running earlier, which does matter from a corporate standpoint.
It doesn't mean it's better, but it DEFINITELY is easier than the state of the art.
As long as they don't make 1st gen mistakes (bloat it, see MACH, the shit OSX/IOS is based on), they're going to blow everything else out of the water in performance.

>>61941303
true, we see from gnu/turd what a shitty microkernel can do to an OS with potential. to me LK is just "good enough" as it already has some solutions for traditional microkernel problems.

maybe we should start a sel4 based distro anon :3

>>61941303
also i want to add, they might've went with lk simply because sel4 doesnt have an arm64 port yet.

>>61941244 >>61941324
>gen2
L4, Minix3, fuchsia, helenos fall into here.
>gen3
seL4, NOVA and as far as I'm aware the only free software effort to do something practical with this is Genode, which is been making slow but steady process for a few years now. The next couple releases are going to be seriously exciting.
>the turd
Has bigger issues than just being stuck with MACH. See the famous hurt critique paper. Fundamental flaws unaddressed, no sizable body of developers left, project is effectively a corpse that sometimes takes a misguided step or two. A zombie.

>>61941354
https://sel4.systems/Info/Roadmap/
They're quite close, but it's true.

>>61941361
the noux environment of genode looks quite promising. im going to contribute to sel4's arm64 efforts soon. i cant wait for a fast powerful microkernel. linux had a way too long run

>>61929779
suck my fucking dick you fucking retard

>>61932606
Gotta get the best new phone on credit!

I'm hype, bjavascript:;ut does it only support arm?

>>61926746
Ultra efficiency, just look at chrome OS for example how it runs like a dream on $200 hardware

>>61935901
Those dog face photos are legitimately horrifying

>forked bsd kernel
>not gpl
dropped

>>61929882
>>61929994
its for iot and embedded stuff and not competing with android

its also very immature.

the majority of work is done by cuks because google cant be helped to design their own OS

just like chromeOS and android (both linux distros) they needed to steal the design from someone else and build upon it

itll be trash because they cant even focus on their other projects

>google regretful of using gpl software

>scared they lost hold of android and linux to community

>fork bsd kernel so you can relicense to a more restrictive license once adopted


REEEEEEE

There's no way they don't rename this before throwing it out there
I also don't really see how they'll begin phasing this out. They're too far gone with Android to just try and replace it. It'll be Windows Phone tier, the Android market share is also far too huge

>>61943577
All they have to do is virtualize the android app ecosystem a-la-blackberry and create a transition SDK adapter. It's not rocket science, it will take a couple of year, but the transition is pretty obvious. By controlling data, they'll have a firm gain in UX efficiency and grasp on consumers.

I don't see the problem, while it's open source, signed and anyone can write its own data persister backend.

https://youtu.be/_7rRK4S9uk0
It's just Cards: the UI

File: 1492472384749.jpg (23KB, 370x278px) Image search: [Google]

23KB, 370x278px

>>61943650

>>61943215
Keep in mind while it is a bsd microkernel, it is not the bsd kernel or related to it.
>>61943650
That's just a gui. Expect Xorg, wayland and so on to work on that thing.
>>61943577
Android is just a bunch of programs and libraries running on a java runtime. There isn't much to tie it to Linux, and I expect they'll get it running on Fuchsia with little to no difficulty.

I loved my n900, after that I moved onto android, shortly dipped myself into sailfish os, then moved back to android and have been using android for years.

I'm not interested in fuchsia, at all, as a smartphone os (if that's what it's for). If it's not linux, it's not going to have wide support for hardware - you can plug in variety of shit to your android and it works. Android is starting to get quite good nowadays and I'm pretty content with it. Yeah it could be lighter on system resources but that's not strictly needed as OEMs continue to push the specs.

>>61941219
Isn't seL4 gplv2?

>google can't compete with ios because of its shitty performance and no updates
>has to make another os
Just shows how superior apple is, unlike google who rushed the os to touch phones.

File: laughing cheetah.jpg (30KB, 500x322px) Image search: [Google]

30KB, 500x322px

>>61946231
>google can't compete with ios