Maybe it’s because I’ve been covering Microsoft professionally for over 30 years, I don’t know, but when it comes to this company and its promises for Windows, I have these out-of-body experiences sometimes, these flashes that take me out of the moment. I had one when it announced Recall at the Copilot+ launch event in May 2024, knowing that enthusiasts were going to over-react when the company dialed down the privacy concerns, as if we suddenly all just trusted it to do the right thing. And I had one when Windows head Pavan Duluri claimed last month that Microsoft would focus on the “pain points” in Windows 11 in 2026.

Here we go, I thought.

I like Pavan Duluri, and I’ll even go so far as to say that I trust the man. But I also knew from his previous hinting about the focus this year that Windows fans, buried deep in enshittification and AI bloat that’s never lived up to the hype, were going to see what they wanted to see. And that’s exactly what happened. As I discussed in Trust But Verify ⭐️, Davuluri not only ignored most of the instances of literal enshittification in Windows 11, but enthusiasts started anticipating changes that he never promised.

This is a big problem, and for once, it wasn’t because of Microsoft’s historic inability to communicate effectively. The best example, perhaps, is Copilot and AI: Most seem to have come away from this promise believing that Microsoft was going to remove most if not all Copilot and AI features from Windows 11. But what Davuluri promised was so much less. Microsoft will continue integrating Copilot and other AI features into the system, he admitted, but it would do so “more intentionally,” providing Copilot and AI experiences that are “genuinely useful and well‑crafted.” And Microsoft is not removing Copilot/AI features from Notepad, Paint, and whatever other apps; it might instead remove the Copilot icon that so many enthusiasts find so grating. For some reason.

The disconnect here is two-sided. On the one hand, we have this vocal minority of AI-denying enthusiasts who don’t represent mainstream users needs in the slightest. And on the other, we have a man, Pavan Davuluri, who wants to do the right thing for customers but must do whatever Microsoft’s senior leadership requires of him. And what they require of him hasn’t changed: Windows will continue getting new AI features this year and into the future.

With Davuluri’s promises made public, other, less-well-known members of the Windows team and other teams that are, at best, Windows-adjacent, have started engaging with customers. But they’re doing so on Twitter and other social media networks, ensuring that most of that engagement will be with the enthusiasts who, again, don’t represent mainstream users at all, and who, as importantly, Microsoft desperately needs to shut up. To Microsoft, this crowd is passionate, if you want to sound positive, but it’s really just a noisy rabble distracting from what really needs to happen. And if there’s anything worse than Microsoft being misunderstood by those who only hear what they want to hear, it’s this: Microsoft clearly communicating something that will not, and cannot, happen.

And that’s what occurred this past week when Microsoft, in the form of “principle lead architect” (the most redundant and pompous job title I’m aware of) Rudy Huyn made the fatal mistake of telling enthusiasts exactly what they wanted to hear: Microsoft will allegedly replace web-based in-box apps and experiences in Windows 11 with “100 percent native” apps and experiences. When I read that, and re-read it to make sure I wasn’t missing anything, I had that familiar out-of-body experience again.

Here we go, I thought. Again.

I want to be perfectly clear here. What Huyn is promising is impossible and will not happen. Whether he’s lying or delusional is almost beside the point. We are not on the cusp of a native app renaissance for Windows. That ship sailed decades ago.

What can happen is that Microsoft can replace some web-based experiences in Windows 11, like the recommendations in Start, which are currently based on JavaScript, with native code that may or may not be faster, we’ll see. It can fix mistakes it’s made in the past like the Frankenstein’s mash-up of WinUI 3 and legacy code that is File Explorer, one of the worst-performing apps in the box. (And an app that Huyn was responsible for screwing up, by the way.) And I guess it could completely rewrite some web apps that ship with Windows. But there are so many issues with that. Here are the two biggest.

First, app rewrites are time-consuming and expensive. In each case, Microsoft will need to weigh the benefits of even bothering, since so many of the apps and experiences in Windows 11 are ignored by so many users. Who would this even benefit?

Secondly, and most importantly, Microsoft cannot and will not rewrite at least a few of the biggest web app offenders in Windows 11. Which apps are these? The new Outlook and Clipchamp, both of which fall under the Microsoft 365 (Office) umbrella. The new Outlook is a web app by design, and the entire cross-platform extensibility model that it uses across desktop, mobile, and web relies on that technology, so it’s not changing. And Clipchamp is literally just a website. You can use it to create videos in a web browser on Chrome OS or the Mac, too. It’s not going native.

The one major web app in Windows 11 that Microsoft could turn into a native experience is Copilot (and the related Copilot 365), but, again, why? Why bother? It’s just a front-end to a website/service, and it makes sense for this app, which is also cross-platform (on the Mac) and is really just a web experience, retains that architecture. The enthusiasts clamoring for Microsoft to replace web apps with native apps are the same people who immediately uninstall Copilot anyway.

Leaving aside all the misconceptions I see in the comments about web apps, performance, resource usage, and whatever else, there are almost no “native” apps or experiences anywhere in Windows these days anyway. And web apps are no more or less removed from the truly native Win32 API than anything else one might use to create apps today. WinUI 3 and the Windows App SDK, like Windows Forms, the Windows Presentation Foundation, WinRT, the Universal Windows Platform, and whatever else came before it are all essentially the same things as React Native, the web app technology Microsoft uses: They are a thing on a thing–or a thing on a thing on a thing–meaning multiple layers of abstraction over actual native code. The people writing code to these frameworks are likely incapable of writing performant and native C/C++ Win32 code, and we would be crazy to ask them to do that. Which is fine. They’re not doing that.

So, yes, native code. Let’s get excited about something that’s not happening yet again. That’s smart. And let’s also let our hate of these nebulous things, like AI and web apps, that normal people do not care about and are not as bad as too many enthusiasts believe, warp reality so that much that we suddenly believe that this is the Bogeyman that Microsoft needs to go after to set things right.

It’s a trap.

Web apps are not holding back Windows 11, they’re not even on the map when it comes to listing out this system’s many problems.

What is on the map is Microsoft and its broader corporate strategies. Microsoft’s enshittification of Windows, its willful abuse of its own customers, is the problem. Not web apps. Not signing in with a local account. And not moving the Taskbar to the top or sides of the screen like some psychopath. The problem is Microsoft. And if you really believe that well-meaning people like Pavan Davuluri or untrustworthy hucksters like Rudy Huyn can in any way undermine Microsoft’s strategies for something as shallow and unrewarding as giving a few enthusiasts what they want, then you, too, are delusional. This is not a fantasy, it’s real life. And Microsoft has bigger concerns than us.

I know. I’m sorry.

The good news? Positive changes are coming. Maybe Huyn and whoever else can actually make the Start menu and File Explorer perform well, for a change. I think that can happen. And I think Microsoft will come through on what Davuluri promised, too. But let’s be honest with ourselves. Will that be enough? Will it ever be enough? Will Microsoft ever really stifle the droning, complaining enthusiasts?

Absolutely not.