5 ms·

> SwiftUI [...] doesn’t feel like an AR platform component (Where’s the third dimension?). I'm unfamiliar with SwiftUI, but fwiw, I found the extensions needed

> SwiftUI [...] doesn’t feel like an AR platform component (Where’s the third dimension?). I'm unfamiliar with SwiftUI, but fwiw, I found the extensions needed for CSS3D to support AR to be surprisingly small. A few years back, having a custom browser-based VR stack with passthrough AR, I sketched a talk for the BostonVR meetup, to give around April Fools. It would have purported to be an introductory onboarding demo, of newly available CSS3D extensions for AR. With support for placement in realspace, billboards, HUD overlays, integrated multiple displays, position aware and 3D displays. The extension needed was surprisingly minor. The talk would have been basically "introductory CSS positioning, in a slightly enriched context", which I expected to be quite believable, followed by a "surprise! - the making of the demo spike". I don't quite remember why it didn't happen. > 2(.5)D experiences Shallow-3D UIs seemingly have a lot of potential, but don't get much discussion. Meta: I wish HN discussions around AR were of higher quality. It's be clear for years that Apple was pursing this. Being unfamiliar with the Apple ecosystem, I'd have liked to see more discussion of what those pieces and their characteristics might suggest about the future. Or for instance, of whether Apple is doing any CEP complex event processing, or retraction of app state changes, to support input with diverse latencies.

grork 5y ago
Shallow 3D definitely has a lot of potential. Not everything needs to be ILM effects - even in the long term. Sometimes buttons & text are all you need.
- mncharity 5y ago
  > Sometimes buttons & text are all you need. And most displays have some privileged focus depth(s). Vergence-accommodation conflict, and eye strain from long use, can be reduced by staying close to that. Ahem, and as my eyes have grown old, the annulus where I can focus well has narrowed. And maybe, it might be possible to have text in layers, close enough to avoid vergence-accommodation conflict eye strain, but far enough to be tolerably readable, and ideally to allow eye tracking to determine which layer is being read. I was layering a keyboard-cam-with-annotations over desktop for a bit, to see the kbd in an HMD, but also to explore "graphic artist looking at screen containing processed view of tablet input"-like UX for software dev. The layered clutter was somewhat annoying, so I'm unclear on viability, but it did seem intriguing. Also, "shallow" still permits some depth. A cm or few at 50 cm (laptop screen) seemed comfortable for me (for some limited frequency of focus switching, and I've only a little experience, and ymmv). That's depth enough for more than buttons and text, but still a very different design regime than "fill the room with XR UI". ILM effects, hmm... someday I'd like to find a stereo drone straight-down view of beach surf, shallow but 3D, to use as desktop background. :) Maybe.

grork 5y ago

Shallow 3D definitely has a lot of potential. Not everything needs to be ILM effects - even in the long term. Sometimes buttons & text are all you need.

mncharity 5y ago
> Sometimes buttons & text are all you need. And most displays have some privileged focus depth(s). Vergence-accommodation conflict, and eye strain from long use, can be reduced by staying close to that. Ahem, and as my eyes have grown old, the annulus where I can focus well has narrowed. And maybe, it might be possible to have text in layers, close enough to avoid vergence-accommodation conflict eye strain, but far enough to be tolerably readable, and ideally to allow eye tracking to determine which layer is being read. I was layering a keyboard-cam-with-annotations over desktop for a bit, to see the kbd in an HMD, but also to explore "graphic artist looking at screen containing processed view of tablet input"-like UX for software dev. The layered clutter was somewhat annoying, so I'm unclear on viability, but it did seem intriguing. Also, "shallow" still permits some depth. A cm or few at 50 cm (laptop screen) seemed comfortable for me (for some limited frequency of focus switching, and I've only a little experience, and ymmv). That's depth enough for more than buttons and text, but still a very different design regime than "fill the room with XR UI". ILM effects, hmm... someday I'd like to find a stereo drone straight-down view of beach surf, shallow but 3D, to use as desktop background. :) Maybe.