Future of UWP?

Question

I'm ready to start a new project and was considering UWP. Not knowing Microsoft's commitment to it at this point makes it very difficult. I have learned some UWP and understand it's not for everybody, but I also remember when .NET first was released. There was a huge out cry and most refused to have anything to do with it. There was even a web site called NotDotNet or something like that. However, now .NET is pretty much the standard and people use it all the time. Is this how UWP will become or will Microsoft kill it off? Any thoughts or advice?

Answer 1

Unfortunately, don't expect a clear from someone from MS.

Having ported my app from WPF to UWP in the last few months, I can give you some pointers:

• there's great news, when it comes to the future of UWP, on the UI side (WinUI). There's a long way to go until it is complete. As far as I know, we've been promised Mid 2020, but we'll see
• UWP is DEFINITELY not for everybody. And it's not because of the differences from WPF - with a lot of cursing, you'll get through and adapt. It's because of the underlying WinRT - which UWP is forced to use for anything not UI related. And WinRT, to put it mildly, sucks.
• A lot of UWP's limitations come from WinRT. Fortunately, there are people at Microsoft that actually care about us, the developers, and are willing to help. So on my mind, the future is a bit gray, with some light at the end of the tunnel.
• MS will definitely not kill UWP - they are waaaay too invested in it. A lot of the Win10 UI is UWP

As long as you don't need to deal with too many files, probably you could give UWP a shot. Do expect a lot of friction at start.

Answer 2

They are constantly improving the platform, not as fast as we would like, but noticeably. UWP wasn't popular when half of users had Windows 7 on their PCs, but now, when Windows 10 is the last officially supported version of Windows, it makes sense to consider new platform for new projects. If you are going to build typical client app for some REST API and that app will not depend on COM or Win32 API (especially things like handles or filesystem access rights), UWP will be good choise and most of app's logic and layout will look like simplified WPF, so you will have an option to port app back to WPF if you want.

Of course, I am not a Microsoft employee and don't know their actual plans, this is just opinion.

Answer 3

In my opinion, what they should do is not to remove UWP, because its UI is great. They should just allow UWP apps to run outside of the sandbox, to invoke native DLLs and so on, by using a special/restricted "Capability". To protect the store from malicious apps, they could just do like they did with the broadFileSystem capability.

Answer 4

I created a very simple WPF App and tried to convert it to UWP. I was able, to but will a lot of difficulty. Some of the very simple, trivial things are either missing or changed. You have to spend a lot of time figuring out a work around. Some of the examples: There are no labels, must use TextBlock, no GroupBox, no Sytle.Triggers, no messageBoxes, had to use something different. No DataGrid, but you can install using NuGet Package, Microsoft.Toolkit.UWP.UI.Controls.DataGrid. However, not all properties are supported, for example: "ScrollViewer.CanContentScroll". Had to change an event handler.
This was a very very simple prototype App. Image if you have to convert a real application.
I would recommend staying with WPF unless you absolutely have to.

Answer 5

We can define the project as dead, the article is in Spanish, but it has enough details:

https://www.genbeta.com/desarrollo/microsoft-anuncia-que-uwp-no-se-actualizara-para-ser-compatible-winui-3-que-marca-principio-fin-para-aplicaciones-universales

But important something from it:

The session began with a quick definition of terms by Ana Wishnoff (Microsoft program manager), to clarify the confusing situation of user interface frameworks in Windows:

"WinUI is the native user interface platform for Windows 10 and Windows 11. […] WinUI 2 is the second generation of the native UX stack in Windows and is designed for UWP applications."

"WinUI 3 is a new third generation of the native Windows UX stack - it consolidates UX technologies previously built into Windows into a single decoupled framework that is included as part of the Windows App SDK (formerly known as Project Reunion)."

Personally, as a developer who published more than one app in the Microsoft Store, I'm quite skeptical to do any migration or even try it. How long is going to last WinUI 3+? Is going to survive MAUI for more than 5 years? Or would we waste our time with a future Windows 12+ and new broken promises?

Right now, there is no path to migrate UWP apps to anything else. I'm just going to keep updating my Android apps and since Windows 11 (and most likely 12+) supports Android, then, people can download them from Amazon or other platforms. If you're a new developer, I'll be quite careful with investing your time in building non-web, non-VR or non-mobile apps.