Earlier quoted context omitted.
In light of the current top comment thread [1], I'm glad the state of browser APIs is holding back this particular "innovation". And I wonder if the friction of distributing and installing desktop apps is actually a good thing, even though it's affecting my current (non-AccessKit) project. It means you have to be really sure that what you're doing deserves to exist outside the good constraints of the browser. I'm not…
I mean, sure, I personally wouldn't care for this feature, but that's because I'm a bloke and don't derive much joy from shopping. My girlfriend though absolutely loves window shopping. I think this sort of feature would be far more appealing to her than me. For better or worse we can't know how popular this sort of thing would be, because their project failed for tech reasons before even getting to the point of doin…
It would be better, IMO, if the web could get back to having multiple browsers that are serious about acting as user agents for a variety of users. Then, freed of the expectation that the browser must render a pixel-perfect presentation of the UI as it was designed, some browsers could provide a VR interface to the lovely accessible hypertext that sites should be providing as their one and only version, while other browsers provide a TTS-optimized UI, a Braille-optimized UI, a UI optimized for large fonts, etc.
Put another way, I think the web community should redouble its efforts to achieve the things that Hixie apparently gave up on. https://news.ycombinator.com/item?id=34622514