I was wondering why Apple rolled out Safari Web Apps in Sonoma. There's no shortage of tools to turn web apps into desktop versions—Fluid, Applicationize, Nativefier, Unite, Electron, etc—but they don't offer a true native experience and haven't gained much traction.
Apple has had a similar feature on iOS which is presumably extremely underutilized. Historically, Apple has pushed developers towards native apps and has been hesitant or slow to support web technologies like WebRTC, PWAs, and WebAssembly in WebKit. I don't think this shift is due to a newfound belief in the potential of web apps, nor is it likely motivated by regulatory factors.
I think part of the goal is to offer a cross surface feature that aims to fill the app void in Vision Pro's upcoming launch. Big names like Google, YouTube, and Facebook, along with many workplace apps, probably won't go native early for Vision Pro.
Many startups are also unlikely to secure funding targeting the Vision Pro launch timeline, resulting in a potentially sparse app landscape. However, many web apps will probably work pretty well as Safari Web Apps (SWAs). Lately, I've been enjoying using Threads this way on iPadOS, so I suppose SWAs are a practical fallback if / when big-name app developers skip the Vision Pro launch.
While converting iOS/iPadOS apps to visionOS is an option, that's up to the developer to support and not all developers will go that route, particularly if they see the platform as a competitor and / or distraction.
Initially, it seemed Apple was late to the game with this feature, but now it looks more like they're just early to the Vision Pro launch.