Google revealed earlier this year that it’s planning to support Windows applications on Chromebooks thanks to a partnership with Parallels. It’s a collaboration that will see a full version of Windows boot inside Chrome OS, providing businesses the option to run existing desktop apps on Google’s range of lightweight Chromebook devices. In an exclusive interview with The Verge, Google is now detailing how and why Windows apps are arriving on Chrome OS.
Google wants to give you access to Windows apps when you really need them, as a hop in and out experience. “The analogy I give is that yes, the world is all state of the art and Dolby Atmos home theaters, but every once in a while you do have that old wedding video on a VHS that you need to get to,” says Cyrus Mistry, group product manager for Chrome OS. “We want to make sure you have that option [for Windows apps] as well… so that every once in a while you’ll be able to get that when you need it, but we don’t want that to be the world you’re living in.”
This feels very much like a stopgap measure designed specifically for enterprises relying on old internal Win32 applications. For employees of such companies, Chromebooks – or anything that isn’t Windows – simply isn’t an option, but this might fix that.
Still, I doubt this will perform great.
Google is a bit foolish when it comes to such things. They are big enough and would basically need to fully support and endorse the GNU/Linux ecosystem (native apps), and they would likely become the most popular GNU/Linux supplier with high market shares overall, when it comes to desktop OS. They are likely so caught up with the idea everything must be on-line, they are throwing away this opportunity and instead offer only subpar solutions in this area.
Because embracing GNU Linux involves embracing the repository/dependency hell system and embracing GPL on things other than the kernel. Google is determined to not go down that path.
In 2020 there is enough solutions available, for distributing GNU/Linux apps in a more traditional or modern way. Google for sure has enough resources, to think of additional ways, if needed. I am not sure, and are you sure, on what issues you see with GPL? On an average GNU/Linux distribution people can run proprietary applications just fine, kernel in Android is GPL, that doesn’t limit anyone all that much, regarding drivers. Sooner or later, they will come to their senses.
GPL is a commercially bad license for an OS. It doesn’t allow for the “proprietary customisation” model that Android follows and it doesn’t allow for the “no customisation” model that Chrome OS and Windows follow. It only allows for non-proprietary give-back customisation. That something neither Google nor the Samsungs and LGs want. Why should Google or any OEM care for the communitah? Even Android is semi-open-source (permissive) to allow for proprietary customisation, not for the communitah.
The kernel is some low-level plumbing nobody cares about. It’s mostly drivers down there. They will just fork it and add their drivers on top and that’s it.
PS: And then there is the repository/dependency hell issue when talking about the GNU stack…
Ever heard of Red Hat? Doesn’t sound all that bad, commercially speaking. As for installing packages, they can use solutions, such as AppImage, Snap or Flatpak, or invent their own. Anyway, we are more or less waiting for them to start providing native apps, web browser in full screen mode unfrotunetly won’t do.
Google doesn’t build native apps besides Chrome and a sync tool for Google Drive. They’re a web company, like everyone else these days.
Native Apps are dead; there is only Web.
Maybe a few years back that was hyped up to such extents, some even believed it. Android and iOS have native apps and that isn’t going anywhere. Every once in a while Google is bringing Windows and sometimes Linux apps to Chromebook. The problem is they use subpar solutions. Do it the right way and get over it, a web browser in kiosk mode isn’t enough and it won’t cut it.
“…Parallels will be limited to what the company refers to as “for power usage” Chromebooks. These typically ship with Intel’s Core i5 or Core i7 processors, and 8GB of RAM for devices with a fan or 16GB of RAM for fanless models”
What does the presence of a fan have anything to do with RAM?
Likely a space constraint due to an overfancy design.
emphyrio,
If the fan is actually needed, which I suspect it is, that means the 16GB fanless models would implicitly suffer from bad thermal and throttling. This is the first time I’ve heard about this and I wonder if it’s actually true? That does indeed sound like a bad design. Proper cooling is needs to be a high priority, otherwise you end up with laptops that can’t perform to spec, much like macbooks that notoriously thermal throttle under load.
I’m really surprised they didn’t just upgrade from 8GB modules to 16GB modules as the obvious solution. No need to manufacture a different fanless configuration to get 16GB of ram. All else being equal, 2-channel ram would be better, but taking away a fan is likely to kill performance more than single channel ram anyways.
The number of page faults going on, maybe?
Bringing Windows Apps to Chromebooks by running “a full version of Windows boot inside Chrome OS”.
Why then not running direct Windows as the host system?
As “bringing Windows Apps to Chromebooks” I would assume something like WINE / CodeWeavers CrossOver / Valve Proton.
Why isn’t Google working together with CodeWeavers, to bring a WINE fork perfect integrated in Chromebook, by supporting WINE and OpenSource, like CodeWeavers and Valve doing it?
They stated it themselves when they said that they don’t want the users to have a too great experience running Windows programs, as they WANT the Windows experience on Chromebooks to be subpar like watching a VHS tape on a 4k monitor.
Besides, why buy an i7 with 16GB of RAM (fan or fan-less) and run Chrome OS on it if you actually need Windows programs? Buy a similarly specced Windows PC and run whatever Google webapps you need in a browser.
Because running Windows as the host system means tolerating all the overhead and slowness of running Windows as the host system.
Why not WINE? Because when users are sold on compatibility, they expect compatibility, not “sometimes it works ok 0but some other times it doesn’t”. It’s the reason nobody considers the Xbox 360 a true Xbox replacement and kept both boxes under the TV for as long as their need to run Xbox games existed.
Basically, the effort it would take to make WINE sellable, devoting entire teams to implement the missing APIs and such, is much more than bundling Parallels. Developer talent is scarce.
Because https://www.codeweavers.com/products/crossover-chromeos has been in Beta for the last 3 years and although it might be good enough for consumers it is no where near good enough for Enterprise. Google is basically saying “if you still need Windows you can add Windows. You need to bring your own Windows license, have a high-end ChromeOS Enterprise license and device and add a heavy virtualization package and huge Windows installation so you can run your must-have applications”. Google also didn’t want to put much effort into this, so they just partnered with Parallel and they are guessing that their customers are not capable of setting up their own application virtualization infrastructure
Basically this is Google saying ” if you really have to ….now you can….but don’t”
Why write both a Windows and an OS/2 application, when the Windows application will run well on both?
Because the OS/2 application allows for a better native feel and allows using native functionality. Not that OS/2 was ever significant enough to warrant its own developer interest, being more expensive and worse in terms of requirements than Windows. Win 3.1 compatibility is what saved OS/2 from becoming yet another “alien” OS like BeOS or the various x86 Unixes (all faring much worse than OS/2)