Hacker News Reader

That AR will be visual first is the eventual obvious thing, but the soft start is audio based. Siri and transparency mode on airpods is the unacknowledged AR present that hasn’t been fully unveiled.

I’ve an iOS app I wrote a while ago which provides the functionality of game audio middleware (something like FMOD I guess) basically for sounds placed at locations in the real world: tracks the users’ location and orientation relative to all the sound sources and spatializes the audio based on that. You can make and layer together sounds that follow coordinate paths, sounds that ‘wake up’ at/after times, are triggered/stopped by proximity, do sample accurate playback at different pitches/volumes for each ‘grain’/chunk (as far as possible on iOS devices), randomise order in a couple of different ways/follow looping playlists for variation and so on. It uses standard ideas nicked from game audio to save on the download/disk/memory size of sounds (though there is obv. trade off with CPU/RAM and perhaps other OS stuff going on I don’t understand). It’s basically kind of an amateurish game audio engine overlaid on the ‘real world’. It’s likely got loads of holes I haven’t detected/am unaware of in the code, probably poorly structured/commented/badly named functions/variables and so on as I am primarily a guitar teacher/music producer/sound designer for my ‘day’ job and it’s the first thing I’ve written in Swift/really just an experiment to see if I could get it working, but I’m kind of wondering what to do with it/what it might be useful for… I made a fully functioning demo. that is a historical recreation of an area of the city I live in with lots of evolving/generative sounds in it: eg. horse’n’cart trotting past, blacksmith, pubs, trigger zones for a narrator voice/music playing back ‘2D’ etc etc …I don’t really know what to do with it now… I think it might be sim. to ‘Microsoft Soundscape’, but I’ve not actually tried their app yet… Do you think anyone might be interested if I put it on Github? (…never used Github, though I used SVN/other CMS as a sound designer on games before… though I know it has a social component…) I guess it’s trivial for most ‘proper’ programmers to make something sim. as it just uses the Apple native API/swift functions (sorry if this is not the correct terminology)/as low level as they go on the sound stuff anyhow, nothing custom?… The sound ‘definitions’ are all XML/.plist basically at the moment, so it’s not user-friendly for sound designers/musos/general public, though I could make some sort of graphical front end that generates the xml for it and uploads sounds etc, though I know nothing about network coding and it looks scary!! I guess Apple could come out with some superior system tomorrow… Anyhow, if anyone reading this is at all interested/has any advice (even if it is to get knotted! :P) that would be gratefully received… Apologies for the overly long and spammy post…

Sounds cool to me. The ability to select your preferred ‘layer’ or collection of sounds for that environment would be good. I imagine a Wikipedia for real world locations could resonate with a lot of people.

Thanks! Yes, you could basically select from various downloadable ‘audio experiences’/layers for that area… The localized user-contribution based audio wiki/encyclopedia type-thing is an interesting idea... Even people telling stories or guiding people around an area… I did think about how users might be able to upload content to a location for others (like an audio grafitti-ish type thing), but I imagine the moderation/networking/storage part of it would be tricky to get right…

Yeah definitely put it up online with a license to get folks to look or contribute to it!

Thanks Karrot_Kream - much appreciated! …I’ll try and do some research about the best way to go about that/tidy it up some/update it for iOS 14 first… (think it was last built for 13)

Reading your HN profile, sound like you might find office hours.global interesting…

IMO this is insanely underrated—people think of AR visual, but it could easily be “a voice in your head”, which AirPods come very close to accomplishing.

FB has 10,000 ppl working on VR and AR. Yes, they are serious about it.

https://www.theverge.com/2021/3/12/22326875/facebook-reality...

> the Oculus platform they have to build up from scratch

It's Android, right?

It's Android.

I think it's not as much the 'OS-platform' as it is the 'ecosystem-platform' the poster was thinking about.

Well, it's easy to get excited fantasizing about the future, but your predictions are about as likely to occur as the flying cars we're flying around right now.

In fact none of the major platforms we use today was just predicted and constructed. They've evolved and shown their benefits naturally and not entirely in expected ways. No one actually planned the web to be an application platform. It was a university paper exchange program. And the Internet before it was a military communication network.

Your predictions about the grand future of AR remind me of the excitement around VRML couple of decades ago.

"Looking at stale 2D web pages will be a funny anachronism of our time" we thought. Turns out making existing content more fancy in 3D wasn't that useful, it actually was more cumbersome both to create and to use, so 3D web pages died before they even had a true chance to live.

You can see the direction things are heading.

The iPhone was a UI step change improvement over previous 'smart phones' and the app ecosystem came out of that.

The ground work being set in the OP's post is about getting things ready for hardware that can then take advantage of it.

It's possible to make predictions based on trends and the capability of hardware that becomes possible when it previously wasn't: https://www.youtube.com/watch?v=sTdWQAKzESA (Also see: Douglas Englebart's the Mother of All Demos: https://en.wikipedia.org/wiki/The_Mother_of_All_Demos). Xerox PARC too - computing history is filled with examples of people pulling the future down because they recognized what was possible.

Just because 3D websites are a bad UI doesn't mean looking at little hand-held glass displays is the best one. Likely in AR we'd still pull up flat 2D websites a lot of the time, you just wouldn't need to pull out a little glass display to do it.

Michael Abrash used to have a blog post about the hardest AR problems when he was at Valve (he's at Oculus now): drawing black, and latency - the latter is mostly a hardware constraint - I'm not sure if anyone has solved the former (the magic leap sucked).

If the hardware is possible, the UI benefits seem big.

I don't know if you remember Google Glass was a thing, and turned out that wearing glasses on your face 16 hours a day is far more annoying than those "little hand-held glass display" you're trying way too hard to be dismissive of.

Google Glass didn't die because it wasn't AR, it died because wearing glasses all the time wasn't practical.

Not practical technologically in terms of battery life, weight, and not practical in terms of simply that you don't need to interact with some digital UI every waking moment of your life.

Also while in our imagination we can conjure up virtual displays in AR and use them for complex UIs, actually waving your hands in empty air, aside from being super weird, is also super inconvenient, compared to handheld multitouch glass.

You're not making AR predictions based on "current trends". You think you are. Instead you're trying to draw a straight line from the present reality to your favorite sci-fi movies that have shaped your idea of what the future is going to be like, while also skipping over all the pesky details that can trip up that idea from concept to realization.

In other words, same reason why everyone was dead set flying cars are coming. And yeah, the generic "no one believed in XEROX PARX and no one believed in trains and car engines, no one believed in airplanes" argument was brought up about flying cars. Turns out that this argument is not an automatic win for believing whatever you wanna believe is coming.

Just because someone didn't believe in airplanes doesn't mean I can't roll my eyes at predictions that faster-than-light travel is just around the corner.

Google Glass died because the hardware sucked, the UI and utility were not there. The general magic device was also a failure, but mobile computing is obviously not.

The timing has to be right and the hardware has to be possible - if you're too early it won't work.

Flying cars is a bad comparison - they mostly don't exist in widespread use because of reasons not related to computing. Risk, fuel, control, etc. - even then rich people do have helicopters (though that's mostly different).

Pointing out failed predictions does not imply that all predictions are similarly wrong. In computing - the examples I showed (and there are others) are more relevant.

> Google Glass died because the hardware sucked, the UI and utility were not there.

And nothing has changed about that.

> Pointing out failed predictions does not imply that all predictions are similarly wrong. In computing - the examples I showed (and there are others) are more relevant.

Most predictions are actually wrong. Let's see the hardware that "doesn't suck", let's see the UI and utility that "are there" and then I'll tell you if we have a winner or not.

Right now we have nothing except bold fantasies powered by sci-fi movies full of cheap hologram VFX.

> And nothing has changed about that.

Yet - it's a prediction based on the capability of future hardware that seems plausible.

> Let's see the hardware that "doesn't suck", let's see the UI and utility that "are there" and then I'll tell you if we have a winner or not.

Yeah sure, it's way easier to make predictions in hindsight after other people have already built it. Even then - when the iPhone launched in 2007 it was largely panned in a similar way to what you're doing now.

Making accurate predictions is hard - I agree with that. If you dismiss everything you'll be right a lot of the time, but you'll also miss every big and interesting change until someone else builds it.

> it's a prediction based on the capability of future hardware that seems plausible.

And nuclear fusion reactors are only 20 years away. That's a concrete goal and it's more likely than some vague new "metaverse". I promise, you'll still be posting on HN when anything resembling a shadow of this is released.

I remember the one time Sergei Brin said helicopters are surprisingly affordable. Sure if you are like 9th richest in the world that might be true.

Helicopters are surprisingly affordable in that you'd be surprised how cheap a bare-bones helicopter like you'll find all over the outback is. It's still shitloads, but it's less than you think.

> wearing glasses on your face 16 hours a day

I do this every day of my life.

Me too, but I've never posted photos of myself wearing them in the shower.