Q&A with Ph.D. student and smartwatch designer Jun Gong

by Debora Hyemin Han | 5/25/17 2:00am

To overcome problems originating from stationary smartwatches, researchers at Dartmouth College and the University of Waterloo created a smartwatch that is able to move on its own. Jun Gong, a computer science Ph.D. student in the human computer interaction field at Dartmouth, collaborated with Dartmouth computer science professor Xing Dong Yang, graduate student at University of Illinois at Urbana-Champaign Lan Li and University of Waterloo professor Daniel Vogel to create Cito, an actuated, moveable smartwatch. Gong recently presented Cito at the Conference on Human Factors in Computing Systems, called CHI, in Denver, Colorado.

Is this an idea you had before you came to Dartmouth or something that you developed while you were here?

JG: I came up with this idea at Dartmouth because right now, as you can see, I’m wearing a smartwatch. So I’m trying to think about the shortcomings and the disadvantages of the current smartwatches and thinking about how I can improve the interactions between the smartwatches.

What do you think are some of the biggest issues with smartwatches now?

JG: I put forward seven issues in my paper, and all of these issues motivated me to design Cito. For example, when your watch is covered by your sleeve and you are cooking or your hands are dirty, you don’t want to use your hands to touch your sleeve. What if the watch could move out from under the sleeve automatically to show the incoming message, and you didn’t have to do anything and then you could use a pinching gesture to send it back [into the sleeve]?

Another example is when you’re carrying a big box and the watch face is facing downward. What if, during the carrying process, there is an incoming message? You wouldn’t be able to see the incoming message unless you put down the box. But what if the watch could crawl around the wristband and show the incoming message? These are some of the issues that made me think, “Okay, there must be some way to solve these problems.” So this is where the idea to actuate the smartwatch, to make the smartwatch move, came from.

Can you talk more about the actuation process?

JG: In my paper, I proposed five movements: linear translation, rotating, orbiting around the wristband, lifting and tilting. So we used these five movements to create some new interactions and to provide more outputs to our user.

How long did this take you?

JG: It took about five months. I spent two months thinking about why I needed an actuated watch and what the issues are for current smartwatches. I generated ideas, and I wanted to find more and more scenarios [to use this watch in] because one scenario is not enough to make a crazy watch, right?

I spent two months thinking about the scenarios, the issues and why this is helpful. Then I spent two months to implement it — to design the mechanical structure, design the whole module and figure out how to make it move. I spent two months using 3-D printed cases and designed the whole mechanical structure and chose the motors and circuits.

Then I spent one month to write the paper and to video tape the prototype.

Do you have any interest in programming the operating system for the watch?

JG: Right now, no. From my area, I’m interested in interaction, this kind of human computer interaction. So I’m more interested in how to design these interactions and how these interactions can help people, can allow people, maybe with no knowledge of device technology, to easily use this kind of device. I’m interested in that — providing more interactions, making people think this product is cool and useful. I’m not a security guy, I’m not an operating system guy, so basically we don’t focus on the implementation of the real system, the real circuit design — I’m just showing the concept and building a proof-of-concept prototype.

What was the biggest challenge you had to overcome to make the smartwatch work?

JG: I think the hardest part is the implementation — to make this idea real. Although I don’t think my implementation is professional, because I’m a computer science Ph.D. student, not a mechanical engineering student, and I didn’t know how to design this stuff before this project. I have no experience with mechanical engineering or gears or motors. I spent some time learning how to build this stuff — I’m trying to learn those mechanical structures and how to apply these mechanical structures into my prototype and how to use which kinds of motors. I spent tons of time in the implementation. And in the beginning, we didn’t just come up with five movements — we tried to add more and more.

I kept adding and adding, so I had to change the whole functionalities, change the whole structures to combine more movements. So it’s kind of like an iteration of several products.

What’s the future for Cito? Are you going to try and engineer your own smartwatch and sell it, are you going to sell the technology to companies, or what do you hope comes out of it?

JG: Right now I think this idea is still a little bit radical. But we’re just showing this concept to get other people to think, what are the issues of the current smartwatches, the fixed smartwatches? Right now, we’re not thinking about pushing this into the market or making it into a real product. We’re just showing a good concept to our users and to other companies in the industry. If they want to solve the issues, actuating the smartwatch may or may not be a good idea, but we’re using this paper and the studies to prove this idea is feasible. And many of our users like this idea; they think, “Wow, this idea is so cool.” Maybe one day other companies will implement this idea and make it a real product, but we’re not sure about that.

Do you personally have any aspirations to present this idea to smartwatch companies? What if Apple, for example, looks at your paper and agrees that current smartwatches do have this issue, and it remakes its watch to have things like lift and tilt?

JG: This could happen, but I’m more than happy to see this kind of prototype, actually. It’s kind of like seeing my idea in a real product: That’s good for me. Of course, I can’t get any money if I don’t have a patent or I don’t sell this idea, but I’d still be happy to see my idea come to reality and be a real product for our users. I think that’s pretty good for me, because for most research papers, they can’t really make their ideas into a real product. Many ideas just stop at the research stage, but all of my research projects are more realistic, to solve real problems. Some papers may not deal with real problems.

Right now I’m kind of working on the research part; I’m not trying to sell the products, sell the idea, but we will apply for patents for most of our projects to make sure the rights are preserved.

This interview has been edited and condensed for clarity and length.

Correction Appended (May 26, 2017): 

The original version of this article stated that Jun Gong was a postdoctoral candidate, when in fact he is a Ph.D. student. The article has been updated to reflect this change.