Rapid prototyping Google Glass - Tom Chi

Transcriber: Andrea McDonough
Reviewer: Bedirhan Cinar

My name is Tom Chi. I spent two years of my life building the user experience team for the Google X division of Google, and it's a place I affectionately call the Department of Science Fiction because of the futuristic nature of the types of projects we took on: self-driving cars, Google Glass, and other things that you'll see soon enough.

So, for those who haven't heard of this project, this is what Google Glass looks like. It allows you to overlay digital things into your eyesight while still maintaining being part of the world. So, if I, you know, were to pull out my cell phone and look into it, I'm basically out of this world now, like, I'm in my own little cell phone-tablet world, what have you. But, Google Glass has the vision of allowing us to continue to be in the world but also have access to the digital things that we need and love.

Now, I am going to ask you a real simple question about Google Glass: how would you prototype this experience? How long do you think it would take you to make the first working version of the headset display? Okay, a little bit on the long side. The answer is one day. And here's what it looked like.

So, basically the magic piece is the coat hanger. The coat hanger, I bent it in a specific shape and the top loop goes around your neck and then the bottom loop rests against your chest and it allows me to carry a piece of plexiglass on with a little sheet protector. So these are the things you put your book reports in so they don't get wet, I literally got at the drug store. You know, have it out at the end of the plexiglass and then it gets projected onto with the pico projector that's connected to a Netbook.

And using this set-up, within one day we're already able to start having the experience of what it looks like to have digital things overlaid on your physical world, be able to move around with it, and also use the Netbook to try out tons and tons of different ideas around software.

Now, after you start getting something like that working, you know, a really important problem comes up, like you're wearing this thing on your head, it's like a pair of glasses, so you don't have a mouse or a keyboard or a touchscreen, all the ways you are used to interacting with a machine. So, we thought for a second, well, maybe we could do something like, you know, what was shown in Minority Report.

So, for folks who haven't seen that, basically Tom Cruise is manipulating software with his hands in front of his face and photos are flying over here and his email is over here and so on and so forth. So I'll ask the same question again, how long do you think it would take to have the real experience of doing something like that? Two years, OK. Somebody said one day. 45 minutes.

So here's how it looks. So you wear the thing that we saw that first time because you need some way to go project things, but what happens is we got two hairbands, which I think was the hardest part we had to do, ask people for their hairbands. But you put one hand in each hairband and attach that hairband, we tied a fishing line.

And the fishing line goes over the top of a whiteboard and then goes down to this little assembly that's taped to the floor. And what this means is every time I move my hand in any direction, it adds tension to the line and it does the following with the assembly on the floor. So, the other end of the fishing wire is attached to a chopstick and it's not because I'm Asian, there's just a cafeteria nearby, I don't just carry chopsticks on me.

But, I tied it to the end of a chopstick, I clipped it into a binder clip, and then put it over a pen, and basically what happens then is when you move your arm and it produces tension on the wire, the chopstick comes down like a lever and clicks a presentation clicker, one hand moves the presentation forward, the other hand moves the presentation backwards.

So this was built in 45 minutes and that meant shortly afterwards, we were having experiences like looking at an image gallery and saying, "next image, next image, previous image," or looking at our emails and saying, "let me click into this email, let me click reply now." And this was exactly the experience of what it was like to go control software with your hands.

And ultimately, what it taught us is we probably shouldn't have this in the product. We learned a lot of things about the social awkwardness of it and some of the ergonomic aspects of it that you couldn't have figured out ahead of just thinking about it. And, ergo the second prototyping rule, which is "doing is the best kind of thinking." They teach you to think a lot in school, but I think it is a little bit overrated.

Now last example, you know, actually Google is not the first team that's tried to go make something like this and if you search for headset display, you get tons of images of teams that have built various systems like this, but I can tell you at a glance that none of these pieces of hardware are comfortable to wear for more than 15 minutes except for maybe the helmet over there, but then you got to wear a helmet.

So, you know, how would you go figure out a way to go wear something like this comfortably? The answer is really basic materials: modeling wire, paper, clay, and using something like this is able to make something look like a pair of glasses really quickly. I cut out pieces of clay that weighed exactly the same amount as the electronic components that we were talking about putting on the device, wrapped it in paper so you didn't get clay on your face, and then taped it to the modeling wire in various places to go experiment with how a pair of glasses could fit on you.

And, we discovered something really important then. Like, if you look at this drawing on the bottom, it turns out that the weight of a pair of glasses is actually mostly perceived through how much weight is on your nose. And, it also turns out that your ears can carry a lot more weight than your nose, and that is a totally different experiment, you can ask me about that.

But, because of that fact, if you put weight behind your ears, it allows your ear to go act like the fulcrum of a lever and it then takes weight off of your nose on the front. And, actually, you can try this now, anybody with glasses, if you push very gently on the back of your glasses, you'll find, actually your glasses feel tremendously lighter.

Now, this meant that we not only discovered something interesting about how to go, you know, that's useful for developing a device like this, we actually discovered something pretty fundamental that never been discovered about glasses, period. So, if you have really heavy glasses, you could do this and you would be more comfortable.

Now, the last point I want to make is about two types of learning because through the process of rapid prototyping, you are able to learn very quickly. It's a very specific type of learning. The type of learning that you usually learn in school I call book learning. It comes from what humanity already knows and it's a necessary foundation for you guys to go and explore the world.

But there is a totally different type of learning, which I call expansive learning, and this is the learning you do on behalf of humanity. Right? You are creating something new, you are expanding into the possibilities, and you're building the sphere of human knowledge in that process.

And, we think about these things and as soon as you hear like, ok, the infinite realm of possibilities beyond the sphere of human knowledge, you might be thinking there's the scientists at the Large Hadron Collider who have these amazing instruments, like that's their job, right?

But the truth is that this action is available to all of us, you know, it's not just for the scientists, it's also for the poet or the songwriter that expresses an emotion for the first time in a unique way. It's also for the person that has an amazing business idea that they're certain could help millions of lives.

And, it's the realm of using paper, clay, and tape in order to go find a new insight in an ancient technology. So now that you know a lot about rapid prototyping, I'm excited to see what you do with it. Thank you.