I've been writing and speaking about the future of the Windows desktop for seventeen years now, and there are literally hundreds of articles and several books which lay out the vision in painstaking detail. That said, I still run into people who aren't familiar with my world view, and so I need a single (and simple) place to point them rather than saying "read these fifteen articles and this book." So here's my attempt to boil down my future Windows vision into a few bullet points.
The future of the Microsoft Windows desktop (as written in 2013)
- VDI is not the future of Windows. VDI is cool, but not game changing. VDI is just a form factor change. When you logon to a VDI desktop, you have the same Windows "stuff" that you've always had. (A start button, the registry, DLLs, application conflicts, antivirus, etc.) VDI is an important part of a complete Windows desktop strategy, but it's no more "strategic" than the move from desktop computers to laptops was 20 years ago.
- In the past, a well-managed desktop was worth a lot, since every enterprise app was a Windows desktop application. (So in 1995, a perfectly-managed Windows desktop meant that 100% of your apps were perfectly-managed too.) But in today's world, Windows desktop applications only make up the minority of apps that users need. (The rest are web apps, SaaS, mobile apps, etc.) So in 2013, a perfectly-managed Windows desktop only covers, what, 25% of the end user things you need to manage for your users? The other 75% need to be managed elsewhere.
- For the past 20 years, the Microsoft Windows desktop was the "face" of IT. It was everything the user needed, including their apps, data, settings, and background pictures. In that world, the Windows desktop was a monolithic "brick." You either had it or you didn't. There was no way to break up that brick into smaller pieces. But today that's different. Instead of user files being part of the Windows Brick in My Documents, we can put them in a new file-syncing tool (Dropbox, Box, ShareFile, SkyDrive, etc.) that works inside the desktop brick while also working on mobile devices, the web, home computers, etc. The same is true for applications. SaaS and web apps are mere URLs rather than Win32 apps that have to be "installed" into the Windows Brick. User settings are abstracted out to something from AppSense or RES rather than being saved in Windows user profiles. Moving forward, we will break the old monolithic Windows Brick into its core components that we can mix & match as needed on different devices.
- The Windows "desktop" may soon die, at least in the context of something that IT departments have to manage. In the future IT will not want to manage all the "gunk" of the Windows desktop. Instead IT will manage the components that make up today's desktop (applications, data, settings, etc.), and those managed components will be able to coalesce on phones, tablets, Macs, and unmanaged Windows devices.
- Windows desktop applications will be around forever, though there will be fewer of them with each passing year. But even ten years from now we'll have some old Windows desktop applications here and there that we need to provide to users. But at that point it won't be worth managing all the "gunk" of Windows just to give the few users who need it access to those apps. Instead we'll just deliver the Windows app in some kind of simple container or via the web. In this world Windows is just middleware for old Windows desktop apps.
- Finally, as a side note, Windows Store Apps (the new name for the Windows 8 touch-based apps for Windows) do NOT represent the future of Windows in the enterprise. Some people think, "Hey, since all the old apps are Windows, these new Windows Store Apps are just new touch versions of the traditional Windows Desktop Applications, right?" Wrong. Microsoft has many restrictions about how Windows Store apps work and what APIs they can use, so you essentially have to start from scratch to build a touch-based scaled down version of your traditional Windows desktop application for touch. And let's be honest—if you're going to do that, you're going to write an iPad app, not a Windows Store App. (And if you're going to do it for two platforms, you're going to do it for iOS and Android, at which point you'll have covered 99% of the world and writing the Windows Store version won't matter to you.) This doesn't mean that Windows touch-based devices won't be popular, since Microsoft doesn't sell Windows 7 anymore. It just means that Windows Store Apps don't represent the future of traditional Windows desktop applications.