Hacker News Reader

Comment by mkolodny | Hacker News Reader

I think this type of AR / holographic technology has many, many more potential real-world applications than VR. With VR, you're shutting yourself off from the outside world. Here, you're enhancing the outside world with technology. You still get to interact with others. Using HoloLens doesn't stop you from doing almost anything.

What I'm curious to find out is whether HoloLens will run into the same core problem as Glass. People are afraid of people wearing Glass. They're scared that they're being filmed, or worse. Unless HoloLens can avoid making you stand out - by looking like regular glasses, or even contact lenses - I'd guess that HoloLens will end up suffering the same fate as Glass.

View on HN

It's hard to understand how AR / holographic technology could help people in their day-to-day life. There are a zillion potential uses, but all of them seem extremely complicated and hard to pull off.

Take the example of fixing your car. For example, performing your own oil change, or replacing your alternator. That seems like a perfect use case for holo, right? The goggles would tell you what needs to be done and what the next step is.

But that would involve so many technical challenges that it seems very difficult. You, as the creator of FixYourCar holo app, would need need to detect what type of car the user is looking at, what part of the car they're looking at, render an overlay with the correct orientation, and so on. And at the end of all of that, it's not entirely clear that your app is more helpful to them than if they'd just look up a list of steps for fixing their car using their mobile phone.

I guess what I'm asking is, what do you think holo's killer app would be?

Imagine a car mechanic remotely helping you to do what's needed to be done. This use case is showcased in one of the videos and it is probably not as complicated as building an AI engine to help a user repair his car.

This thing is really crazy!!

Ah, yeah.. That's a good point. And the bandwidth probably exists to transmit images from the goggles to someone else in near-realtime.

Now I'm really excited about this.

This is actually a relatively silly use case for this. A lot of the actual difficult things that a mechanic can do for you usually involve more strength than you have. Or just experience in actually working with things that they can't see.

Now, a mechanic using this to "see" things that are actually in control of a remote robot? Pretty cool. Showing you the thing that is right in front of you? Cute, but ultimately silly.

Yeah, I'm just not sure that I buy the idea that there's a big market for an expert coaching you through doing repairs via AR goggles.

How exactly does this work?

You still need to pay for an expert's time -- in fact, you probably need to pay more for it, because the expert is probably faster to do a thing than to explain the thing to you and then you do it. Also, the expert now needs to be someone with these additional skills of coaching someone through an operation.

Tools are still needed -- is there actually a big market for the kind of repairs you can do with the tools that everyone has lying around at home but which is complicated enough to need hand-coaching by an expert?

I mean, maybe! Especially if you can locate the expert in some place where labor prices are much lower (so: India). But then you also need the person to buy the AR goggles. And how often does this use case come up? Is this like 3D printers where people try to sell me on the concept that I could pay hundreds or thousands of dollars for something that can make me things that cost less than $20 and which I need three of every year?

At any given time there are probably thousands of car mechanic experts sitting idle

You could draw on a much bigger pool than professionals. There are plenty of people who know a skill and don't professionally sell their services but could spare a few minutes occasionally to help someone with a problem. If you combine skill tracking with instant global availability of services, you have a lot of room for development.

Training. Emergency assistance. Military applications. Don't trap yourself by thinking that the item needs to be the be-all for the Joneses.

Yeah this could make putting Ikea furniture together alot easier.

Like that's every going to happen.

Online training though - a whole other thing. Imagine math and physics with interactive 3D visualisations.

Or for kids, a virtual version of one of those 30-in-1 electronics teaching kits.

And so on.

If this works it could be a game changer, and it could also create a whole new app industry.

I just hope the technology is non-crappy, and it doesn't get managementised into uselessness.

None of those things require the augmented reality aspect of this: you don't need to place math and physics lessons into your local environment. They'd do just as well with VR, and indeed it doesn't sound to me like they'd do MUCH worse with just a plain old screen. What is it that you're imagining we couldn't do with a tablet that has swipe gestures to rotate the demo around all axes?

An electronics teaching kit might not work on a tablet (but would in VR), and note that any kind of really fluid manipulation of a virtual environment is going to involve a whole additional technology that gives precise locations of your hands (at least). The HoloLens allows a few simple gestures, not the ability to handle virtual objects in many degrees of freedom.

I agree this is highly unlikely to be a common use case. I might use it, because I try to do most things myself, and I could often use some expert advice. And I know the people who would help me, but it'd be inconvenient to have them come all the way out here. But overall, this is a one-in-a-thousand use case.

But:

> Tools are still needed -- is there actually a big market for the kind of repairs you can do with the tools that everyone has lying around at home but which is complicated enough to need hand-coaching by an expert?

Almost all repairs on appliances, cars, houses, and so on can be done with tools you have laying around the house; they don't require anything more than a hammer, drill, screwdrivers, wrench sets, etc. The only thing you're typically not going to have on hand is replacement parts, which are usually not too difficult to get your hands on and which you would have to pay for anyway.

When people do this, it's not going to be an "expert", it's going to be something along the lines of "hey dad, look at this real quick."

> But then you also need the person to buy the AR goggles.

The Ars reporting showed someone using a Surface to view and annotate the HoloLens-user's view, not another set of HoloLens. So the barrier is much lower, any Windows 10 PC should be godo enough.

Have you ever worked on a car? Tons of specialized tools can be needed for some tasks. Do you have a set of triple square bits? General purpose puller? Bearing extraction tool? How about half inch drive Tor-X sockets? 200 ft lb torque wrench? When you work on a car for fun, you find that your collection of tools balloons just for all the things on the car that require one specific tool. If you think you can do everything on the car with just a simple socket set, you'll wind up stuck and having to buy a new tool for every task.

Yes, although granted when I said "almost all repairs" I did not have in mind major automotive work, but more along the lines of general maintenance and little things going wrong. Of course if you're doing something like rebuilding the transmission you'll need more tools than the average bear.

Fair enough, but my point was that for things where you'd benefit from a mechanic walking you through a task, you'll probably need some specialty tools to do the job.

Also, apologies for my brusqueness.

I'd approach this problem differently. Not from a service industry angle but from a product vendor angle. Imagine a world where AR-glasses are widespread. I'd be pretty interested in buying the kitchen sink that comes with an AR repair guide or the furniture that comes with AR assembly instructions.

So I think the interesting market is in building the infrastructure/app that makes it easy for vendors to create the content and ship it as a (free) addon for their products.

But the hologram need not be a person! At least, it won't be once this use case makes sense. The hologram, will be an AI hologram. Just like the light switch installation in TFA, it would be silly to have a human expert show you how to install a switch, or trade out a component in your car, once an AI expert will do.

I don't think that 'strength' is the issue but better tools and a lot more experience are the core parts: I remember watching one mechanic changing the light in my car: what took me ~15 minutes (I'm not kidding) took him ~30secondes..

And that's like riding a bike: you cannot really tell someone how to do it..

It can still be very interesting in many ways!

Don't forget actually having the right tools to do the job.

If there's a market for this, why doesn't it already exist? You could take your smartphone under your car, and video chat with a mechanic anywhere in the world. The mechanic could even draw arrows or highlight areas on your video in real-time as it's looped back to your display.

Or allow them to select from a list?

That adblock thing would probably be great. Also, I ofthen wish there was an easy way all those specials in the store could be compared more easily. Usually when you work out that 2 for x offer you see they gave a generous 10% discount that made you buy an entire extra thing for no real savibg.

https://www.youtube.com/watch?v=4LE_IocFnL0

http://www.bmw.com/com/en/owners/service/augmented_reality_w...

Its not like car manufacturers dont have cad models of their own cars or anything, right?

I don't know about the consumer market, but I can think of numerous commercial applications.

Sony currently sell an HMD for surgical use, allowing for comfortable and convenient viewing of video from endoscopic cameras. A practical translucent HMD would be extremely valuable in surgical procedures guided by x-ray or ultrasonic imagery.

To give a trivial and easily-implementable example, I would have bought Google Glass without hesitation if it integrated with my electronics test equipment. Being able to view data from an oscilloscope or logic analyser without taking my eyes off the PCB would be a boon. PCBs are designed with fiducial markers as a necessary part of manufacture, and machine vision is already used extensively in many aspects of electronics manufacture and repair; It would be relatively straightforward to overlay all sorts of data that would be enormously useful to technicians and engineers.

Stereoscopic and volumetric displays are used extensively in petroleum and mining geophysics; This equipment is currently relatively niche due to extremely high cost, but could be used in a much greater range of geoinformation applications if costs fell.

Making it easier to find your car in a large parking lot would be a start.

I noticed that the promo videos all show you doing things indoors, in more or less private settings. Your living room, your kitchen, your workspace. Contrasted to initial Google Glass video (skydiving, jogging, meeting for lunch .etc) I think it's safe to say Microsoft has learned from Google's mistakes.

I noticed this too. It avoids a whole class of problems interacting with other people, and seems like a good idea marketing-wise.

I was also thinking about battery life. If it's not designed for outside, then presumably you'll be near a charger, so you're less likely to run out of charge when you need it.

From the videos HoloLens looks like being an actually well thought out product unlike Glass.

Many people would find tools like these incredibly useful at work, in the car or at home. But not in the street, at the beach or in restaurants whilst talking to other people. That's just socially awkward/insensitive.

Reply on news.ycombinator.com