Google’s homegrown Fuchsia operating system has taken another step closer to being broadly usable by gaining the full Google Chrome browser experience.
[…]It’s been possible to access the web in a very limited way on Fuchsia for quite some time now via the operating system’s “Simple Browser” app – which was powered by the Chromium engine under the hood. While usable, this “browser” didn’t offer the usual necessities like an address bar or tabs.
Mid last year, we reported that Google had begun efforts to bring the full Chrome browser experience to Fuchsia. As first spotted by oldschool-51 of Fuchsia’s Reddit community, these efforts have come to fruition in recent days, with Simple Browser being replaced in Fuchsia’s app list with “Chromium.”
It’s steps like these that show that Google is serious about Fuchsia. For whatever that’s worth.
Thanks Google but no thanks.
Mark my words and mark them well, all those that say “Oh Linux is gaining cuz look at Android” are gonna be in for a BIG f’in shock when Google replaces Android and ChromeOS with this completely proprietary black box OS in a few years. Gopogle has been trying for years to stick more and more of their products behind the Playstore paywall but having FOSS guts has limited this.
The fact that Google is obviously dropping some serious cash into making this a reality should be a wake up call to all those using ChromeOS and Android as Google wants to join MSFT and APPL in the locked down black box club.
bassbeast,
I agree it is a possibility, we need to tread carefully.
Fuchsia is open source.
That is not enough in 2022 anymore. At minimum device drivers needs to be open source too. For it not to be considered a black box.
In all fairness, there is not much OS in Fuschia yet. At least with Android, you can still build a functional Android OS from AOSP with things like a contacts app, a music app, a gallery etc
So, let’s wait and see…
kurkosdr,
I agree. My opinion and support for fuschia will be based on how it turns out. It could still end up being good or bad at this point. It could end up turning into a restricted ecosystem with proprietary blobs, but we don’t know this yet. If it turns out to be a very open platform that is easy to support I will welcome it. The additional competition could even put pressure on linux to finally get its house in order on ARM.
“Proprietary blobs” is a highly ambiguous term. Drivers are “proprietary blobs” that the OS vendor cannot always control (not in highly mobile gadgets were you can’t handpick Atheros WiFi cards and whatnot). Instead, this is more of a “you” kind of responsibility (pick whatever device you want with whatever license drivers you want). And then there is stuff that is part of the OS such as location services and hence is legitimately an unavoidable proprietary blob if not made available as open source.
kurkosdr,
You are right, there are multiple proprietary blobs on a typical android phone. Some of them are google’s and others are manufacturers.
I’d say google has the power to eliminate both of those assuming that were on the agenda. The manufacturers will fall into line if it’s a requirement and google brings in enough orders to make it worthwhile. Being open could be a huge selling point for people who’ve been dealing with bad manufacturer support.
That said though, I don’t feel comfortable assuming google is going to act in the interests of FOSS. After all they could open source their own proprietary blobs on android today and they haven’t. Hopefully drivers will be isolated from kernel changes at least so that users can independently update their kernels. This has been a huge headache with android!
Correct me if I’m wrong but isn’t it a BSD based license? Since BSD isn’t GPL and has no stipulations that I’m aware of for sharing the code anybody can blackbox it at any time, which is why its used for so many set top boxes here in the states.
And you wouldn’t be sinking millions of dollars and having all those devs working on this OS when you already have 2 GPL’d OSes unless there is something this can offer those can’t and the only thing I know of BSD licensed code offers over Linux is its free as in beer NOT free as in freedom.
So basically; you don’t know what Fuchsia is, but you’re certain of what it’s supposed to be.
Those are fools anyway, because Linux kernel is considered private API that userspace is not allowed to use.
https://developer.android.com/ndk/guides/stable_apis
https://developer.android.com/about/versions/nougat/android-7.0-changes#ndk-errors
They pat themselves in the back completely clueless of how Google controls Android,
But if those are fools. Not knowing how Google controls Android. Then maybe they do have a case? When suggesting Google wants even more control?
The problem is not knowing Google controls Android, rather that Linux isn’t at all relevant for Android apps, but they still try to sell Android as being a Linux success story.
Well, it is a success story regarding the adoption of the Linux kernel for the mass market, about the GNU userspace, not at all.
That is why we don’t say GNU/Linux in regard to Android. I feel that people in general are aware of that. That is people who care about this things. As for Linux without the GNU. I would still say this is/was a success story. Nothing wrong when saying that out loud. Of all the kernels it’s refreshing to know Linux is ticking in the guts of Android.
Part of my problem with the term GNU/Linux is that GNU does not matter very much at all even for desktop Linux. The application API has long been POSIX and / or Xlib and neither of those things are GNU. The GNU part is the “userland” which most users barely care about, the C library ( Glibc ), and the compiler. That is all plumbing and it can all be replaced without application developers needing to care about it hardly at all.
I agree with you completely that Linux in Android is just the kernel and that application API has nothing to do with Linux and the same applications could easily run on a different kernel. It is still HUGE that so many devices run the Linux kernel when considering the level of support for that kernel and the quality of its implementation. Losing Android to another kernel would be a huge blow to Linux.
You could say that “desktop” Linux is just as vulnerable as you can run all the same applications on FreeBSD for example and that is of course true. If the Linux kernel and the various “distributions” built around it were to become less attractive, a BSD could certainly easily step-up to displace it. For Android, Google backing another kernel would likely be enough.
Anyway, Android is certainly a huge Linux success story, it is certainly not a GNU success story, and ( other than the beneficial engineering brought to the kernel and benefits to the Linux “brand” ) success on Android does nothing to further Linux on the desktop. I think we mostly agree.
@tanishaj
If nothing else, and rest assure there is more to it, Linux uses GNU General Public License. Including in Android and hence no matter how some will or would try to minimize the importance of GNU. GNU is and was pivotal for the success of Linux. No matter on how hard the industry would like to erase that and rewritten history. This era will always be considered GNU/Linux era. Including Android. Without GNU/Linux there would be no Android. Google more or less had nothing and a decade later they still have very little in comparison to GNU/Linux. They still rely heavily on GNU/Linux. As for the “problems”. There for sure are problems with GNU/Linux. But no way the problems involved justify the 1% market shade on the desktop.
Today I learned that Google killed Google Things. Imagine having designed a POS or a hotel/in-flight entertainment system with a 10 year commercial life around it. Even it gets security updates, you still have invested in a technological dead-end.
If a Google project is not immediately successful, assume it’s dead already. But you should also avoid all 1.x versions of Google products (because Google typically screws customers that invest in 1.x versions of Google products: see Android 1.x, Google TV aka Android TV 1.x and Android Wear 1.x).
Of course, this makes people reluctant to invest in 1.x versions of Google products, which creates a chicken-and-egg situation.
Considering what partnership with Google has done to Firefox and if something like that would to happen to Linux. Then i feel we could call this being a successfully executed EEE strategy. Personally i feel that Linux is bigger then Google and it would be Google who would fall first. That is if they would really try to reduce the importance of Linux to scratch and try to force some in-house solution instead. I do wonder on why didn’t they use Firefox for this experiment? Considering they say Rust should be used everywhere and as Firefox uses Rust? Why are they avoiding and not adopting Rust in their main projects?
Rust is being used on Android.
https://source.android.com/setup/build/rust/building-rust-modules/overview?hl=en
Is that main enough?
No. That is not enough. If Google says Rust should be used in Linux. As a second language. Then Rust should be used in Fuchsia too. As a second language. And yes that includes the browser Fuchsia runs. A component such as web browser is a component that can benefit from better memory safety the most. Hence saying we have done our part and enabled other people to use Rust in relation to Android. That is not enough. As you should use Rust to build things like Fuchsia and Chromium too. Or just use Gecko as a base instead.
Fuschia’s userspace has plenty of Rust already.
I see. Then when the world will starts using more Fuschia. Then we can compare the pros and the cons of that.
This won’t be “killed by Google”. When Fuchsia is complete enough Google will ship an Android runtime with it and the users won’t notice the difference in their phones. Same with Chrome OS.
Could be but as the history shows such “revolutionary” approaches don’t always pan out.
Precisely Google's approach to Fuchsia will not be “revolutionary”, it will be “evolutionary”
I don’t consider Fuchsia to be an evolutionary step. Different languages, different device drivers model, different kernel design, different userspace, different way on how to collaborate, different … altogether. To me it’s more like Facebook -> Meta transition. That is there used to be Facebook, but now there is Meta. A vision of some future somebody is making. But is not quite there yet. At the same time people moving to TikTok. And that is telling. Even for a company as Google Fuchsia represent a big step and in my opinion success is not guaranteed. IMHO they are still reliant on Linux too much, to survive without Linux. But maybe they will try and lets see how that goes. It’s not like we can stop them. In addition Linux will still likely remain a big player if Google would move away from Linux.
Except Google has abstracted the user facing identities away from the underlying plumbing. For hardware manufacturers Fuchsia will be a different beast, but for consumers it probably is wholly invisible. Android and Chrome OS aren’t dependent on Linux. If Fuchsia gains enough capabilities, Google can just swap out the plumbing and no consumer will be the wiser and that is what matters to Google. IHVs will fall in line. The money is too good not to.
@r_a_trip
In theory yes but in practice one must be aware of the fact that Google didn’t really invent Linux, Java, Kotlin … That is it’s much harder to start something from scratch and make it a success story. Lets say for Google to push only in-house technologies such as Fuchsia, Flutter … Maybe they can pull it off or maybe they can not. We’ll see.
Geck,
I’m sure you will disagree, but I think you are underestimating google and overestimating android’s dependency on linux. The userspace has been fairly well abstracted from the kernel. They don’t have to start from scratch either, once they have a working kernel, I don’t think it would take much effort to switch android userspace to it. A small team could do it easily. They could even call it “android” and users probably wouldn’t notice the switch at all. It really depends where google wants to go with it and whether they want to differentiate fuchsia from android or slide it in as a direct substitution.
I have said it before but I think the strategy win Fuchsia is the following:
!) Replace the Linux kernel. Make the old APIs whole compatible.
2) Make the new Fuchsia APIs (in C++, Dart,, Go, Rust, etc) available. Make them fully inteoperable with the old. Perhaps even make the old a thin layer over them.
3) Make the Fuchsia APIs the preferred APIs. Make them more featureful.
4) Make the old thing optional. Make it possible to ship new products without the Android Runtime & co.
@Alfman
If Google really has a plan to try to execute an EEE strategy. Regarding Linux on mobile. Then not much we can’t do about it anyway. Somehow i feel that we will manage as we always have. With GNU/Linux. Maybe there will be more GNU/Linux phones after. Maybe indeed people will become a bit smarter then their phones. Who knows.
Geck,
“Embrace, extend, extinguish” doesn’t really fit to be honest. Goggle didn’t buy android, build up linux marketshare more than any other company has in a bid to kill it decades later. I think a potential shift to Fuchsia will have more pragmatic motives. For instance, dealing with long term inaction on driver decoupling, which has given android a bad name in terms of support for decades.
It’s too early to say whether fuchsia will actually deliver a worthy successor, and I have to withhold judgement until I see where it’s going. But I applaud google’s decision to try something rather than just waiting out driver problems that the linux community has shown zero interest in solving for decades. If new competition puts a flame under leaders who’ve become overconfident and take their success for granted then all the better…. The threat of migrations to a rival FOSS OS could be what the linux project needs for leaders to finally take the fingers out of their ears and listen to our needs. I’m sure this sound harsh, but I say it as a linux user and developer who wants linux to genuinely improve to succeed instead of just being stagnant and succeeding as an incumbent.
@Alfman
You don’t need to buy something and to try to execute an EEE strategy against it. As for the fears. This is what basically the gist of the concerns is. For some. Will Google manage to execute a successful EEE strategy regarding Linux on mobile or not. But i do agree that most people don’t care about that like at all.
Geck,
Google has done more to promote linux kernel adoption by the masses than anyone else for over a decade. You may be annoyed if google ends up switching, but it does not fit the motives for “embrace extend extinguish”.
@Alfman
GNU/Linux has done much more for Google than Google has done for Linux.
Geck,
Who knows where things would be at if google had invested in something other than android, but I kind of doubt that android would have survived as an independent entity without google. Virtually all of linux’s mobile market share is thanks to android. Of course there are other projects like maemo.org, but they never even registered on the marketshare scale. I understand your ill-will for google if they replace linux with fuchsia, but realistically I think it would have been very difficult to build up linux’s mobile marketshare without the help of google. Who else would have been willing and able to do it?
Google has probably had it with the Linux kernel’s unapologetic server centered design. Combined that with an environment where almost all apps are already written in a cross platform language like java, experience making them work on Chrome OS, getting tired of the driver bs, etc. It should not be hard for them to make the switch. It simply takes about 8 years to make a new commercial OS, so visible progress is slow.
@dark2
So you say they are fed up with Linux but you do acknowledge it took them at least a decade to catch up? And during that decade of catching up, they being fed up, having a dominant market share otherwise only possible for them in some wild dream? I guess in the end it’s rather easy to live a full-assed life. Isn’t it?
They did what every other company would do. They used the cheapest, easiest option until it was no longer sustainable. It certainly did not take them a decade to catch up to anyone. Apple started smartphones, and Google caught up within a year or two. The problem is after years of non-stable driver ABIs, hardware vendors that refuse to update their drivers, maintaining a separate code base for the kernel, and other things; Google has most certainly come to a conclusion that it’s cheaper to build their own replacement for the Linux kernel in their products. Google made Android popular by providing an SDK and decent 3rd party software support, and that’s something Linux and the Linux desktop just have no interest in. All google has to do is port their userland to their new kernel, and then make sure it passes the Google Play test suite.
@dark2
Is that really Google take on it or is that your take on it based on a lot of assumptions?
Conspiracy theory 2312313123 – Google is secretly supporting Android on Windows to have a large scale test bed of independent (of Linux) Android implementation and sort out all the kinks before switching to Fuchsia.