A Way Forward for the Microsoft Store and Windows Apps (Premium)

While the Universal Windows Platform (UWP) has failed, there is a single spark of hope for the Microsoft Store in Windows 10. And I think this success points to a way forward for the Microsoft Store and for Windows apps.

I am referring, of course, to the Desktop Bridge.

If you think about Microsoft’s long-term strategy for Windows at a high-level, there are a couple of key takeaways. Microsoft wishes to move past the legacy Win32 core of Windows because of security and reliability issues. And, it wishes to modernize Windows so that it behaves more like mobile platforms and can be serviced like an online service.

Initiatives like S mode (originally called Windows 10 S) and Windows 10 on ARM speak to these ambitions, but they impose too many limitations. In S mode, no Windows desktop applications or drivers are available, essentially cutting off users from the things that make Windows, well, Windows. And Windows 10 on ARM is even worse, with its performance problems and inability to run 64-bit desktop or Store applications.

Anyone rational would look at Microsoft’s goals for Windows and see them as logical and well-intentioned. But this same audience should likewise shun S mode and Windows 10 on ARM because they impose the wrong compromises.

Worse, the new Windows platform, UWP, is terrible. This mobile platform was originally designed for a type of hybrid PC that never took off and for integration with a smartphone family that failed in the market. UWP is a mobile apps platform and the Microsoft Store is a mobile app store. This is not what Windows users need.

Which brings me to the way forward.

Microsoft’s long-term strategy is contradicted by the reality of today: UWP, S mode, and Windows 10 on ARM are all failures. There are few compelling apps in the Microsoft Store in Windows 10, but if there were one group of apps that we might describe as being of routinely high quality, it is those applications (and games) that are really desktop applications, not mobile apps.

They are Desktop Bridge apps. And, on the gaming side, Xbox Live-compatible Xbox Play Anywhere titles that have absolutely nothing to do with UWP either.

Since the gaming piece is pretty obvious, let’s focus on apps.

Microsoft wants to deprecate and, as much as is possible, eliminate desktop application support in Windows. It wants to do this, again, because these applications are insecure and unreliable, and because they can impact the performance of your PC in a variety of ways. (Including the arbitrary ability to add running tasks at PC startup.)

But desktop applications are hugely popular. And while Microsoft is, perhaps understandably, not sharing any data about the applications—or types of applications—that people are really using with Windows 10, I think it’s fair to say that usage patterns haven’t changed. They’re using desktop applications. Almost universally.

Microsoft should run with that.

A mobile apps platform makes no sense on Windows. And Windows users still run legacy desktop applications. Microsoft already has technology—the Desktop Bridge—that lets the developers of legacy desktop applications “containerize” those apps, which brings with it a number of improvements.

The Desktop Bridge (basically) sandboxes the application from the system, creating a virtual file system and other resources for the application and “fooling” it into believing it is running normally. It allows the developer to easily add modern (and UWP-based) app features, like notifications and live tile support. And it lets the app be delivered through the Microsoft Store which, despite its faults, is at least understood to be a safe place to find apps.

The problem with Desktop Bridge is that it requires the developer to buy-in. A user cannot take an arbitrary desktop application, run it against a wizard, and have a contained Desktop Bridge app emerge on the other side.

That would be a nice capability. And it would go a long way to fixing my problems with S mode, in particular. But better still would be an automated capability, where any downloaded/installed desktop application would simply be automatically wrapped in a Desktop Bridge container on the fly. This would let users continue using the applications that they really want to use. But to do so in a way that is consistent with Microsoft’s need to modernize Windows. This system would fulfill real-world backward compatible needs without compromising the integrity of today’s modern Windows 10 platform.

What about the perceived need for Microsoft to create a modern apps platform for the future? That’s sort of tangential to this discussion, but I’ve long argued that Microsoft needs to make UWP capable of creating more professional, “command-dense” applications. That work could certainly continue.

But I find myself not really caring much. What I am concerned with is recent Microsoft efforts to hobble the desktop applications platform that defines Windows and that its customers expect to use. Put simply, Microsoft should do what its users need, and not force users to do what it wants.

Who knows? There may be some future Microsoft platform that only runs UWP apps. But that platform is not Windows, not even close.