Well, that was a short run. Announced with much fanfare in 2021, Microsoft has announced it’s already killing Windows Subsystem for Android, Microsoft’s solution to run Android applications on Windows 11.
Microsoft is ending support for the Windows Subsystem for Android™️ (WSA). As a result, the Amazon Appstore on Windows and all applications and games dependent on WSA will no longer be supported beginning March 5, 2025. Until then, technical support will remain available to customers.
Customers that have installed the Amazon Appstore or Android apps prior to March 5, 2024, will continue to have access to those apps through the deprecation date of March 5, 2025. Please reach out to our support team for further questions at support.microsoft.com. We are grateful for the support of our developer community and remain committed to listening to feedback as we evolve experiences.
↫ Microsoft’s Learn website
Whenever Microsoft offers a way or an API to run and/or develop applications for Windows, and it isn’t Win32, you can be certain they’re going to kill it within a few years.
Another feature nobody asked for. Wasted million$$$ for a strip down Google Play Store that nobody uses. I don’t think Google has an official Android Desktop mode available. Other than the game for PC thingy.
I actually use WSA with F-Droid in order to run AntennaPod, KOReader and NewPipe but also Shazam.
Years ago Shazam used to be available for Windows but it was abandoned.
I am disappointed Microsoft abandoned WSA and will have to either find an alternative to running those apps or alternatives to them when using Windows.
On Linux, Waydroid works fine.
It was a good idea to plug gaps in the app ecosystem of the Surface Pro line (when it comes to touch-optimized apps), but Microsoft made the mistake of not using MicroG and instead going the way of the Amazon Appstore (which isn’t much of an ecosystem to begin with).
If they had partnered with some mirror website like APKMirror and shipped MicroG with the Windows Subsystem for Android they’d have something good on their hands, since all non-paid Play Store apps would be instantly available.
And of course, the above assumes Microsoft still cares about the Surface Pro line.
kurkosdr,
While I have personal reasons for favoring MicroG, it’s what I use with lineageOS, I’ve got to be honest it’s not all smooth sailing there either. Some banking apps not work, an authenticator I needed for work had hard coded google play dependencies, another password manager didn’t work. Unfortunately when you bring this up with the developers, they typically respond by saying you need to use an official OS from the apple/google duopoly and that’s the end of it.
I’ve fought this time and time again because I really want alternatives to succeed or at least be viable, but I’ll be honest for most people it’s a lot of stress with little to show for it when network effects force you back into the duopoly anyways.
Hypothetically microsoft could put a little more pressure on companies to support microg, but mobile hasn’t worked so well for them in the past so I’m skeptical it would have made a difference. It says a lot about the consolidation of power when even a monopolist has trouble competing in such a highly vendor locked market.
Realistically speaking, Microsoft’s best option was to use MicroG and whitelist the most popular 1000 apps or so (the ones that work with MicroG anyway), for example, they could partner with APKMirror to offer “Windows Subsystem for Android verified” apps (or start their own mirror website).
True, in the past, it was possible to create an 100% compatible implementation of someone else’s API if you had enough perseverance and capital to do it. This is no longer possible, apps nowadays are tied to online services (and online DRM services) and they require “middleware” to function (with the “middleware” being tightly coupled to the online service, for example the online service can reject old versions of the “middleware” if it wants to). For example, Play Services is a “middleware”, the Steam and EA “launchers” are a “middleware” (the launcher part is a small subset of their functionality). This makes providing a 100% compatible implementation impossible because you are targeting a moving target.
Although I don’t understand what do you mean by “hard coded google play dependencies”. I understand MicroG not being an 100% compatible re-implementation of Play Services/GMS, but things should simply fail to work with MicroG, I don’t understand what “hard coded google play dependencies” are.
But in general, I am also feeling the pain of this. If an Android phone stops receiving Play Services updates, any games that rely on Play Services for DRM stop working after a while (even if you haven’t changed anything on the phone!), because the online DRM service in Google’s servers rejects the old Play Services version. It’s one of the reasons that I prefer buying games from GOG and not Steam, because GOG allows you to download a plain exe that requires just Windows, no “launchers” or crapy “middleware” to update.
Still, Microsoft should have targeted partial compatibility with MicroG, simply to have something resembling a proper ecosystem.
kurkosdr,
I’m not working there any more, but it was a 2 factor authenticator required to log into a VPN. The old VPN worked just fine, but administration felt the need to outsource authentication to a cloud provider, which they did. Unfortunately this provider’s service was built around google play services for messaging. There was no way to connect directly to the VPN.
I’m sure we can agree this is objectively stupid and creates unnecessary failure modes. Regardless though it became impossible to use the VPN from a google free android device, it required google services hooked up to a google account. I ran lineage OS on my phone with microG precisely because I didn’t want my phone associated with a google account. So once they made that change, I made a big stink about it, but they didn’t care. I tried to make them provide me with a phone that could log in, but they wouldn’t do that either… it pissed me off but they forced me to personally pay for another google phone on my own just so that I could log in. Ugh.
WordPress removed some characters…
Anyway, I do agree the mobile ecosystem desperately needs more competition. I wouldn’t be surprised if we remain stuck in this duopoly long term though.
Yeah, DRM is evil. I tried replaying some of the games I owned long ago and DRM is a real issue for long term software preservation. It sucks that DRM hurts legit owners. Piracy for the win? Haha.
It’s over, we are stuck with the App Store and Play Store forever. People won’t repurchase all their App Store or Play Store apps and lose all their saved games (some of which the result of expensive in-app purchases) to go to some other OS that will do pretty much what iOS and Android do. The only niche I can think of is devices that can do both win32 apps and Android, where the Android part is a gap-filler/nice-to-have. Surface Pro is such a line of devices. The good news is that Android allows a large degree of customization without losing Play Store compatibility, OEMs have done all kinds of crazy stuff with Android (foldables, flip-phones with foldable screens, stereoscopic 3D phones etc), so you will still innovation even if you won’t see competition.
Even if you have made peace with offline DRM, online DRM is extra evil because stuff just stops working. At least with offline DRM, you can still play PS2 games or old PC games without issue if you have period-correct hardware. But with online DRM, in some cases, piracy FTW is the only option to make things work again.
Alfman,.
I don’t think this was the solution you were looking for. Microsoft, as kurkosdr suggested, implemented this API as a stopgap measure, while there was no actual use for it.
If you want to develop Windows applications there is already Win32, .Net, and UWP, among others.
If you want to use your existing apps, they would not work as you mentioned. The Android apps you purchased on Google Play would not be compatible, and Google already brings them over anyway: https://play.google.com/googleplaygames
This probably gave some high level VP lots of clout and financial bonuses, and made a few developers who used it unnecessarily wrong on their expectations.
And, they should have known better. IBM tried to do the same to Microsoft in the past, by implementing Win16 API on OS/2, which led to even less native applications on their platform.
sukru,
Hmm, I’m confused what you are referring to…aren’t we talking about supporting android software, not windows software?
Ok, but I don’t think that’s what kurkosdr was suggesting though. We’re talking about an alternative android ecosystem to google, aren’t we?
Alfman,
Sorry for the confusion. I was under the impression you were discussing Microsoft doing that alternative.
sukru,
kurkosdr suggested microsoft should have supported microg, but I don’t know if would have made a difference. Maybe it could have worked a long time ago like kurkosdr mentioned with a much more serious investment. But these days mobile is a mature market and for better or worse the incumbents are too powerful for others to reshape it.
Alfman,
I am sure at least some people in Microsoft are still sad for shutting Windows Mobile down.
However, what is done is done. And more importantly, they did the majority of the damage themselves.
Yes, it is difficult to enter as an alternate operating system, as people have vested “investment” with their years long digital purchases.
The purpose of Windows Subsystem for Android was to allow Surface Pro laptops access to applications win32 and UWP/Metro don’t provide natively, it wasn’t about being a competitor to Android smartphones.
With that in mind, Windows Subsystem for Android + MicroG could have been useful in filling the app gap for the Surface Pro line even if it wouldn’t have been a complete alternative to Play Store + Play Services.
But again, this implies Microsoft still cares about the Surface Pro line, and considering they’ve missed the yearly refresh for the Surface Pro, I have my doubts. Now that desktops and laptops are selling again (you need one to work-from-home effectively), it’s entirely possible that Microsoft doesn’t care about tablets anymore and is focusing Windows exclusively on desktops and laptops again. And if this is true, they destroyed the beautiful desktop-optimized UI of Windows 7 for nothing.
kurkosdr,
Ah, I misunderstood you then. I thought you wanted MS to introduce a google app store competitor. I agree with sukru, microg doesn’t do much for microsoft or surface users without the google services & play store. The lack of a strong software market is one of the weaknesses using microg platform and I got the wrong impression that you wanted microsoft to bring that to the table.
Well, this is a big topic, haha. There were jarring UI issues with windows 8, but I believe it had less to do with demoting desktops and laptops and more to do with microsoft’s walled garden ambitions. The bad experience wasn’t an accident, it was designed to convince users that legacy apps didn’t fit in anymore. Remember their marketing leaned towards reclassifying win32 software as “legacy software” and it ran on a very awkward legacy desktop. Per microsoft, the future was “modern apps”, which in the windows 8 era happened to be locked down to microsoft’s app store. If this hadn’t been so unpopular, then today win32 apps would be deprecated outside of enterprise customers and new windows computers would be using modern apps tethered to microsoft’s walled garden.
Obviously microsoft had to backtrack due to customer uproar, but their goals for metro perfectly explain why the optimized windows 7 desktop had to die. They wanted to give “modern” software an edge even if it meant making the legacy experience more jarring.
They could introduce a limited marketplace just for the Surface Pro line in co-operation with a “mirror” website like APKMirror, but again, it would’ve been a gap-filler for the Surface Pro and not something fit for Android smartphone users.
That was the intent, but thanks to usual 2010s-era Microsoft incompetence, Windows 8 was a UI mess both for win32 apps and for Metro apps. Even when using Windows 8 as a pure tablet OS (only Metro apps), the Windows 8 Settings app would randomly drop you to the Control Panel or Explorer because there were no Metro equivalents for some of the settings and there was also no Metro equivalent for managing disk space and doing disk cleanup. Windows 10 also sucked as a tablet OS for the exact same reason (you’d think 3 years would be enough to fix the UI at least for tablet users, but nope), but it had a pre-installed start menu. tl;dr The Windows UI has been sucking since WIndows 8.1 for both tablet and desktop/laptop users. So, Microsoft’s tablet and Metro ambitions were thwarted (not that they had much chance against the dominance of the iPad and its App Store) and we are now left with the UI mess.
.NET is also entrenched enough to be considered safe, but anything else cannot be considered safe from deprecation.
kurkosdr,
Yes, .Net was the “better” Java, and it became really popular back in the day. Unfortunately I am not as enthusiastic, as in, not “evangelically trying to convert people”, but it is here to stay.
Yet, inside .Net many APIs rose and fell. ASP.Net, Razor, and whatever came after that in the web area, Xml Serialization / SOAP, later REST, and WCF for communication, Windows Forms, WPF, Silverlight, UWP.
Even though many are “backwards compatible”, or have carried parts over, there is a lot of “cruft” over there, too..
This seems obvious – Android isn’t all that useful without gapps, and they never offered even an unofficial way to get that running. That’s on top of Android’s limited use outside of phones (even on tablets). After of a few minutes of curiosity, I never touched the thing (and I did play with gapps and some other method of installing Play apps that I’ve long since forgotten about).
I’d be very surprised if they also cancelled WSL2, though the GUI thing might not survive.
I personally think WSL may take over that helm.