You are missing the initial point here, the accusation is that apple doesn't want the web as a viable platform to develop fully fledged cross platform applications that circumvent its own moat. You just repeat the maximalist position that the web shouldn't "extend the browser so deeply beyond the web". Many people disagree with that, its the best shot we have for true cross platform applications that would force Apple to open their platform, we say the reason apple doesn't want that is for purely profit driven reasons.
Also doesn't this argument apply to WebAssembly and other standards (like WebGPU) as well? Why should I be able to render a video game in the browser but have no way to manage input devices (like multiple controllers) to make it a suitable platform for fully-fledged video games in the future? Like I understand why Apple (OR Microsoft for that matter) doesn't want Stadia-like services to suddenly run on all their devices without any cut in monetization, so naturally they sabotage such efforts in Safari...
Like here is my own reason to support that effort: I use Linux, and there are a ton of proprietary applications that if there were developed on the web platform would become accessible to me.
I understand the privacy concerns with some of those APIs, but the argument isn't that the user shouldn't have agency over those features, do you see how that's a separate conversation?
> Many people disagree with that, its the best shot we have for true cross platform applications that would force Apple to open their platform, we say the reason apple doesn't want that is for purely profit driven reasons.
And equally, Google wants their platform to replace OSes for purely profit-driven reasons of their own. I'm not saying this as some terrible indictment. They're all corporations, I don't expect anything else from them.
> Also doesn't this argument apply to WebAssembly and other standards (like WebGPU) as well?
This is a fairly consistent strain of criticism on WASM posts here on HN. The death of software as we know it; all of computing as a service, forever.
> Like here is my own reason to support that effort: I use Linux, and there are a ton of proprietary applications that if there were developed on the web platform would become accessible to me.
What makes you think future thin-client 'PC's will even be able to run anything other than a browser shell? The current requirements for Windows/iPhone/Android will just be replaced by a requirement to run an approved, TPM-clad Chromebook, cryptographically certified to be running no software other than Chrome. The ultimate in Secure Boot. But of course this will provide no actual security to end users, in a world where websites can write to your keyboard firmware. Security for corporate IP, maybe.
I genuinely think you're letting your dislike of Apple - on which I hear you, Apple's no angel - cloud your judgment here. The future Google's building is pretty dark.
But it feels like a slippery slope argument to me, like how do you go so quickly from standardized web APIs to your ChromeOS/TCPA dystopia? I think that's quite the leap to make, especially considering we already have those locked tight ecosystems right now with iOS, Android, Windows, Mac OS to varying levels of degree anyway. If proprietary applications go that route they would remain unaccessible to me, it wouldn't really change anything. But right now I can use things like figma and countless other apps because it runs on the web, it would otherwise never have been possible.
> in a world where websites can write to your keyboard firmware
again I would never relinquish control over the decision if it "can", I've been using Linux for 20+ years I already go to significant lengths to remain in control over my computing. But the fact that I can develop a cross-platform app that talks to some USB device directly via some standard web API, has benefits that outweigh the costs, that's just pragmatism.
Also if you asked me I would personally rather see Apple, Google, Microsoft and every other orphan crushing machine/publicly traded company shut down and all it's CEOs and shareholders hanging from street lamps, but personal politics aside, I'm only speaking of within the dystopia we already have anyway.
I don't think your stated interest in "remain[ing] in control over [your] computing" is at all compatible with the specific moves in tech that you're advocating for.
There's an obvious long-term trend for software to migrate from on-disk to cloud-based: from everything SaaS, through gaming, to Office suites. One common thread with all this software is that you have absolutely no control over it, which is - of course - the aim. No piracy, no unauthorised mods. Walled gardens forever in every direction.
Equally, there's a simultaneous long-term trend for 'attestation', ID checks, 'digital ID' documents, etc. Already, many bank apps refuse to run on Android devices that aren't Google-blessed and kissed. This trend is only accelerating.
In a world where paying taxes online - or accessing a bank account, or running an office suite - requires a Trusted Device™, what good is OSS? What good is software freedom? Running a Linux computer unable to interact with basic aspects of the modern world will quickly become as quaint as trying to do your day job on an Amiga.
Also doesn't this argument apply to WebAssembly and other standards (like WebGPU) as well? Why should I be able to render a video game in the browser but have no way to manage input devices (like multiple controllers) to make it a suitable platform for fully-fledged video games in the future? Like I understand why Apple (OR Microsoft for that matter) doesn't want Stadia-like services to suddenly run on all their devices without any cut in monetization, so naturally they sabotage such efforts in Safari...
Like here is my own reason to support that effort: I use Linux, and there are a ton of proprietary applications that if there were developed on the web platform would become accessible to me.
I understand the privacy concerns with some of those APIs, but the argument isn't that the user shouldn't have agency over those features, do you see how that's a separate conversation?
So now: Why are you against it?