Mishaal Rahman, who has a history of being right about Google and Android-related matters, is reporting that Google is intending to standardise its consumer operating system efforts onto a single platform: Android.
To better compete with the iPad as well as manage engineering resources more effectively, Google wants to unify its operating system efforts. Instead of merging Android and Chrome OS into a new operating system like rumors suggested in the past, however, a source told me that Google is instead working on fully migrating Chrome OS over to Android. While we don’t know what this means for the Chrome OS or Chromebook brands, we did hear that Google wants future “Chromebooks” to ship with the Android OS in the future. That’s why I believe that Google’s rumored new Pixel Laptop will run a new version of desktop Android as opposed to the Chrome OS that you’re likely familiar with.
↫ Mishaal Rahman at Android Authority
The fact both Chrome OS and Android exist, and are competing with each other in some segments – most notably tablets – hasn’t done either operating system any favours. I doubt many people even know Chrome OS tablets are a thing, and I doubt many people would say Android tablets are an objectively better choice than an iPad. I personally definitely prefer Android on tablets over iOS on tablets, but I fully recognise that for 95% of tablet buyers, the iPad is the better, and often also more affordable, choice.
Google has been struggling with Android on tablets for about as long as they’ve existed, and now it seems that the company is going to focus all of its efforts on just Android, leaving Chrome OS to slowly be consumed and replaced by it. In June, Google already announced it was going to replace both the kernel and several subsystems in Chrome OS with their Android counterparts, and now they’re also building a new version of Chrome for Android with extensions supports – to match Chrome on Chrome OS – as well as a terminal application for Android that gives access to a local Linux virtual machine, much like is available on Chrome OS.
As mentioned, laptops running Android will also be making an entrance, including a Pixel laptop straight from Google. The next big update for Android 15 contains a ton of new proper windowing features, and there’s more coming: improved keyboard and mouse support, as well as external monitors, virtual desktops, and a lot more. As anyone who has ever attempted to run Android on a desktop or laptop knows, there’s definitely a ton of work Google needs to do to make Android palatable to consumers on that front.
Of course, this being Google, any of these rumours or plans could change at any time without any sense of logic behind it, as managers fulfill their quotas, get promoted, or leave the company.
I’ve mixed feelings about this, because I’ve been enjoying ChromeOS as my “everything OS” for awhile now. It provides (a) solid driver support and power management on some good modern laptops (better than Linux) (b) a full version of Chrome that websites actually _believe_ is a desktop browser (unlike Android) (c) a “truly Linux” CLI and graphical environment that gives me access to a broad selection of dev tools and (d) Android apps if needed to provide good quality touch oriented UIs for various common applications that are not well supported on Linux.
That said I’ve always thought that Android has the potential to be a really good general purpose operating system, that has been gradually abraided by the sole focus on being a consumer oriented mobile OS. More focus on it as a desktop OS might turn this around. But I’m afraid what we’ll get is less good version of something like iPadOS, where the locked down nature and consumer focus makes it a poor environment for those of us who actually want control and flexibility in our devices.
Android has the potential to be a good general purpose embedded device OS which is because of it’s focus on being a consumer mobile OS.
If anything, the Android ecosystem is going to hold this back, like how they’ve held back Android tablets. I have no faith in 3rd party Android devs being able to handle a desktop environment. They can’t figure out how to put documents in a consistent location on a phone or how to deal with an SD card. How are they going to deal with a system with more then one disk that may or may not be a remote?
This announcement would be less humorous if Android wasn’t such a mess.
I’m not sure this will fly.
They might want to unify their base runtimes, but even that is problematic.
There is Chrome OS, with atomic updates (which is the precursor to Silverblue and other similar container based systems. It is actually the original as CoreOS was based on Gentoo / Chromium).
There is Android with multiple read write and some read only partitions.
There is even the Fuchsia, which is their ground up open source microkernel operating system.
The philosophies are different. You cannot install packages on Chrome, but you can install virtual operating systems on containers, including Android.
And that is where they will have problems. They have promised 10 years of updates to commercial customers. Do you think a large customer with hundreds of millions on the line will allow experimenting with their mission critical devices? Nope. They will either have to keep Chrome OS for at least 10 more years, or be ready to pay massive damages in some high profile civil cases.
They promised 10 years of support, not 10 years of updates. If they would stop adding new features and would only just provide security updates (like Microsoft did for more than 10 years with Windows 7) then the letter of agreement would hold.
zde,
Yes. They don’t need to release new features.
However they need to maintain the operating system for 10 years. Which means keeping up with Linux kernel and base libraries and tools. Either by heavily back porting everything to massively ancient releases. Or maintaining an up-to-date stack.
The have sold support for Android, Steam and Linux under Chrome OS. Which means they would actually need a modern kernel:
https://support.google.com/chromebook/answer/9145439?hl=en
If your organization bought Chromebooks for Android Studio / VSCode development, you’d expect to be able to run their latest versions 10 years from now after all. If the VSCode installer says your Linux is not supported it will be on Google to fix it.
And my experience with Google tells me they are terrible with maintenance. Basically no engineer would want to be assigned to that task.
This will be an interesting roller coaster.
Google is not responsible for what is going on inside of your Linux container. It’s just a VM. You can run modern Linux OS in VMWare released 10 years ago, no problem. Similarly with ChromeOS: Google would fix security issues with VM and everything beyond that is not their responsibility.
zde,
In a way they do…
Once again, they sold Chromebooks with the implied (or explicit) understanding that productivity applications like Xcode, or gaming platforms like Steam would be available.
I’m not sure either of those will continue to run on a 10 year old kernel.
Okay, thinking back…
It seems like this is another “VP” taking in reigns.
We have seen this happen many times. A new VP is hired to direct a project, and they decide to “deprecate the old way” without giving enough time to build up a proper replacement. For one reason or another Google higher management does not question this, even though it has failed many times.
Not only this affects the morale negatively, it also has a trust deprecating issues for consumers and partners.
Google Talk -> Hangouts -> { Allo, Duo }, Hangouts Chat -> Chat
Google Checkout -> Google Wallet -> Android Wallet -> Android Pay -> Google Pay (might have missed a few)
Android Things -> Works with Nest -> Matter
Google Play Music -> YouTube Music
Google Chrome Apps -> Progressive Apps
Chrome Extensions -> Manifest v3
Google Cloud Print -> ???
And many, many, more.
In all of these instances the “new and shiny” was never a proper replacement of the older one. Heck, many years later, I still miss the features from “Inbox by Gmail”, which was supposed to be properly rolled into mainline Gmail.
This never misses to disappoint.
The “Chrome Extensions -> Manifest v3” thing was Google making proper adblockers impossible on Chrome (basically the “switch” part of the “bait and switch”, let’s not forget people moved from Firefox to Chrome because it supposed ad-blockers just like Firefox), so it’s a bit different. I agree with the rest.
kurkosdr,
Yes, of course that is true.
Yet, I still see it as a “VP decision”. I’m not sure having the “side effect of weakening ad blockers” was a goal of engineers working on Chrome.
What they should do instead is to turn Chrome OS into a fully fledged desktop operating system and to compete with Windows.
How would that make it more profitable?
ChromeOS problem was always that it’s a tiny, niche, OS that demands about as much resources to develop and support it as huge and profitable Android.
Replacing ChromeOS with Android plus proper browser on top is no brainer, really.
The only big question, that I have no idea how to answer, is: why this is happening in year 2024, and not in year 2014, when this should have happened?
What kept ChromeOS alive for so long?
Android is horrible outside of phones, and ChromeOS is much better suited to desktop, laptop, and tablet form factors.
ChromeOS was also the Google Workspace thin client. Management by Google Workspace admins was a big selling point.
Why does it matter? It’s simple math!
ChromeOS sells around 20 millions devices per year (except for short temporary jump in COVID times).
That’s approximately the same number of sales Windows Phone got when it was discontinued.
But it’s complexity and thus money needed to continue to develop it is comparable to Android.
It just doesn’t make any sense to invest in it if you don’t share the codebase.
If Google may spend 10% of development efforts by adding specific things that desktop needs to Android – then it could make some sense, if that’s not possible – then ChromeOS should just be discontinued, but why spending 10x more per user without any hope of ever reaching parity?
You couldn’t fix that by trying to squeeze 10x more from each user, math just doesn’t add up!
Historically speaking a mobile operating system never really worked on desktop and vice versa. So if Google believes they can simply install Android on a desktop computer and expect them to succeed. Then they are naive.
I don’t think Google plans to just install Android on desktop without suitably altering it (like many Chineese companies do). This just wouldn’t work.
Google can spent ⅒ of what they spend on Android development to make Android better for desktop and that’s not a small sum.
But, for years, they tried to support completely independent OS which would require 10x more spending per user (if we assume that development of any single feature requires comparable investment).
That math doesn’t make much sense and I couldn’t see what have been driving them.
P.S. It’s the same story as with UNIX workstations: they sure were cool. Seriously cool. But small number of sales per year could never justify development of OS comparable in capabilities to Windows or ChromeOS thus they are dead now. Why should ChromeOS be any different?
Google has been trying to make Android tablets happen for years, but it hasn’t happened.
ChromeOS had a lot of “free” work done, so I imagine it was fairly cheap to keep running. Chrome was already paid for, and quite a bit of the Linux stuff was already paid for, by others, as well. The GUI was custom, but there were rumors they were moving to Wayland. ChromeOS was about to get even cheaper.
I would imagine Android is much more expensive to maintain considering Google writes the entire stack.
ChromeOS makes sense because the world is a different place. The web browser is the common runtime environment, and native apps using native toolkits aren’t the priority for companies anymore.
I’d like a FirefoxOS. There are many days I don’t fire up a native applications.
It’s true that people no longer use native toolkits to do GUI. That’s really bad for users, but that’s the truth.
However companies ONLY use web technologies for GUI which makes abominations like ChromeOS or FirefoxOS are stillborn. And you are correct: ChromeOS is exactly as dead as FirefoxOS, and for the exact same reason.
They were getting there, and ChromeOS was competing with Windows. It even had an Office version.
I mean proper competition, using the whole GNU/Linux stack, like a normal GNU/Linux desktop distribution and utilizing the whole ecosystem build around it. Other people have already developed it for them including modern approach to packaging. Now it’s on them, just like they have done with Android, to take all that and profit. As lets be realistic, without GNU/Linux and wider ecosystem accompanying it they have nothing. A crippled down version will never work on desktop, or forcing a limited mobile OS on desktop and expecting it to work. That is so much on what some exec would do, lacking understanding.
“The whole GNU/Linux stack” sucks. There are no SDK and while Flatkpacks and SNAPs promise something that may resemble it… one would need to invest billions to even achieve basic requirement of any App developer: ability to create binary which works on 5-10 years old device and also can use latest and greatest bells and whistles.
Android have, macOS have it, Windows have it… GNU/Linux doesn’t have it. Valve, with Proton, comes close – but to extend it to something resembling general-purpose OS (and not just something capable of running games and nothing else) would be both costly and risky.
And since one may already run Android apps on ChromeOS… going with full Android is no brainer.
Whether geeks would like it or not is irrelevant, geeks are never enough to sustain popular OS.
This is why people pay Red Hat for RHEL or run Debian.
Not having an SDK is kind of the point of a baseless system, and creating a base is why Linux distros exist, especially point-in-time distros. Debian 12 or RHEL 9 is the SDK.
It’s not every ChromeOS device can do this There were restrictions last time I checked, and it was mostly the expensive models last time I checked.
ChromeOS had gained quite a few features over the years. There were rumors it was going to become a full fledged Linux OS. Switching to Wayland from their custom display server was something which had bubbled up in the rumor mill.
Quite a bit of the power of my Fedora installs gets wasted because I do nothing by run web browsers some days.
I have to imagine ChromeOS was cheap to produce since most of the OS was pulling in Chrome or some other Linux tools. Unlike Android which is mostly Google.
The Android ecosystem pretty much ignoring tablets doesn’t bode well for this effort. Fuchsia with Flutter was probably the better path forward for both.
I have looked on numbers. No, they weren’t even close.
There was jump to almost 40 millions devices sold per year in 2021, but then it went back to 20 million like they had for many years.
And ChromeOS is not MacOS, Chromebooks are cheap, it’s not as if they could compensate for the lack of volume with price.
The only way for them to make the whole thing sustainable is to reduce development costs. And that means using Android as base.
That was the move I expected 10 years ago, and while the fact that Google finally does it is nice, I still couldn’t understand why they have waited for so long.
Looking at the numbers, macOS doesn’t compete either. Moving 20million units is quite a few units, and Chromebooks were being picked up by people.
I’m pretty sure ChromeOS development is cheaper then Android development. Especially since ChromeOS is reusing parts developed by others, unlike Android.
Android didn’t suit the form factors ChromeOS does, and ChromeOS was built as a companion device to Google Workspace.
ChromeOS was built because laptop makers wanted to have some alternative to Windows to have a negotiation power with Microsoft.
Apple sells about the same number of devices as ChromeOS sells – but these are much more expensive and they bring money to Apple (and even then Apple tries to unify macOS and iOS codebase as much as possible).
ChromeOS doesn’t bring money to the Google from hardware sales and only brings money indirectly.
And the question is not whether ChromeOS development is cheaper than Android development, but whether it made sense to develop Android for desktop to be used in ChromeOS (which Google did starting from year 2016) and also develop something besides it.
Developing something is cheaper than not developing something thus the answer is total no-brainer.
ChromeOS fate was obvious 10 years ago, the only question is why Google decided to first put Android on top of ChromeOS and create an ugly and resource-hungry monster (not a good fit for cheap laptops, that’s for sure!) and then took many years to finally ditch the useless outer layer.
zde,
It absolutely did.
People can’t develop applications for Android on Android. They need something like Windows, Linux, or macOS, but if Google baked the ability into ChromeOS people wouldn’t have to leave the Google ecosystem. If they were going to add Android emulation for developers, they might as well make it so Android apps run as well.
The scope of ChromeOS has expanded over the years. It’s been on the path to being a full OS for a few years.
You don’t need Android support in ChromeOS to develop Android app. And, worse yet, you couldn’t use it to develop Android apps either!
Sure, few years AFTER Google added support of Android apps to ChromeOS they ALSO have added support for Linux – which made it possible, finally, to develop these Android apps, but, unlike Android, support for THAT one is not automatically added by default, is entirely optional and NOT covered by service contracts (but Android support is covered).
At some point heavy drugs that you definitely need to invent such strange theories may rot your brain. You really should stop using them.
This will be fun. 🙂
Is this an open admission Web apps are failures, and native applications are better, even if the apps are ultimately Java based?
THAT particular admission happened many years ago, back when, in 2016, Google Play arrived on ChromeOS.
Now it’s time to remove the cruft, apparently, and make Android apps the only apps supported in ChromeOS.