Final wearable: 1UP

1UP is a small clip-on device that tracks your physical activity throughout the day. It is designed around the fundamental tenet that most people don't want to quantify themselves to obtain data, they want to know what they should do to be healthy.

1UP is a simple indicator of activity through the use of a Bluetooth LE link to your gyroscope/accelerometer-enabled mobile device. At any given time it reports one of three states:

  1. Low activitity level (flashing blue arrow only)
  2. You've met your daily activity average (alternating blue and green arrows)
  3. You are well over your daily activity average (flashing green arrow only)

With the generous assistance of Mr. Fletcher Bach, I etched my first custom circuit. It has two LEDs and two 220 Ω resistors. I surface mounted the components for a thinner profile. There are also three holes for pins to connect to the TinyDuino that I used as the microcontroller.

After cutting the excess wafer, I soldered the TinyDuino to the board. It didn't work at first because I had put both LEDs backwards, but after troubleshooting this it works perfectly.

The enclosure is a small acrylic box with a cover of two halves of acrylic sandwiching a strip of wood.

Key challenges encountered:
  • Form factor (difficult to make device as thin as I had wanted)
  • Charging/power (could not figure out a good way to make the battery removable)
  • Bluetooth integration (the development of the BLE link was far too ambitious for only a few weeks)
  • How difficult it really is to design a wearable deice that not only works but is fashionable/aesthetically pleasing
After creating this device, I finally understand why the concept of wearables has been so slow to take off. It's an incredibly hard design and development process that requires knowledge in several very different fields. I now have more of a respect for why "failed" wearables fail.

Final: 1UP wearable activity monitor

My final project is a small wearable device that encourages physical activity by monitoring the number of steps you've taken during the day and abstracting this information into either an "upvote" or "downvote," similar to the voting apparatus used on Reddit.

 If the green arrow is lit, your physical activity is trending positively, and if the red arrow is lit, you need to get some exercise. The device will use your daily average number of steps to determine whether you should be upvoted or downvoted.

The biggest technical challenge is that I have been unable to obtain the Blend Micro or a Bluetooth LE TinyShield, as one is out of stock and the other takes weeks to ship. Even if I had ordered several weeks ago, it would have only arrived next week. I am exploring alternatively options, including using a TinyDuino to demo the device as if it was paired with my iPhone.

Today before class I am going to fabricate the acrylic pieces, and mock up the other components to get a sense of its size.


For my final wearable, I am considering a simple display of my daily steps. My phone is constantly tracking my steps through an internal pedometer + accelerometer. But I have no easy way of sharing the information about my activity with anyone in public, besides showing them the screen of my Health app. Instead, it would be nice to have a meter displayed publicly on my body so that anyone could be aware of my physical activity that day.

I imagine a simple LED meter for this prototype, perhaps ranging from blue (low activity) to red (high activity). The meter would be connected to a Bluetooth Low Energy (BLE) device, possibly the Blend Micro, which in turn would interface with my iPhone in order to read my step data using the HealthKit API provided by Apple.

Having such a device would eliminate my need to check my Health app several times a day to monitor my activity level, but it would also allow my activity to generate a conversation with anyone who noticed it. Yesterday I walked 8.89 miles in the course of various errands and travel to and from school. I wanted to bring it up but it felt a little braggy. And maybe this wearable would be braggy as well, but I think it could provide a way to reinforce healthy behavior while sharing something about my life with others in a passive way.

The Uncrosser

In order to determine a way to make people in close proximity more intimate, I imagined a social situation in which nonverbal communication would be of great importance: a party, crowded with people who you both like and dislike. I also used The Nonverbal Dictionary to refresh my familiarity with the nonverbal methods I learned in Speech and Debate class in high school.

The arm-cross is a classic nonverbal cue that is "unconsciously used to alleviate anxiety and social stress." A study on college students even found that "women use open arm positions with men they like, but cross-arms with men they dislike." The study concluded that men show no difference in this regard, but I beg to differ, as I know I cross my arms when I'm talking to people I dislike.

Since the goal of this week's project was to increase intimacy, it would make sense that a reduction in arm crossing could facilitate more intimacy. My design prototype is a shirt with a pressure sensor band underneath the fabric across the chest, which would detect the presence of the wearer's arms in a crossed position.

When pressure is detected, a light shock is applied to the wearer's arms, causing them to immediately release their standoffish posture. A few times of this shock therapy is all that's required for a more intimate social experience. The results can be seen below:

A new model for healthcare

I enjoyed this week's article, "Reframing health to embrace design of our own well-being".  It is a call to action for the redesign of every aspect of the healthcare system to allow those who we today call patients to become tomorrow's users of a participatorily-designed structure focused on true well-being, not just being free of disease. Or in the words of the the World Health Organization, health is "a state of complete physical, mental, and social well-being."

I agree that we need to empower people to take control of their own healthcare. There is a recognizable trend in this direction that I believe started with the rise of the Web – simply having access to more information turned a lot of people off to the traditional model and on to the DIY, holistic side of things. There are merits to both strategies, of course, but we need a merger of the two to really get the most advantage.

The article is also inspiring because it brings the concepts of good design to an area which might not seem hospitable to design practice. Though we usually think of good design in the form of objects or structures, design thinking applies to pretty much every construct in the world, tangible or otherwise. This is why Dubberly's brief introductory paragraph is so inspiring – "wicked problems can be 'solved' only by reframing, by providing a new way of understanding."

PanHandler - a Tone.js plugin

For my midterm I developed PanHandler, a Tone.js plugin that pans an arbitrary number of inputs in interesting, random ways. To showcase its abilities I built a simple visualization which can be controlled with the keyboard.

I love music that takes advantage of aural space so I wanted to create a utensil for panning multiple inputs at the same time. PanHandler has two main functions that handle panning:

  • shuffle() - randomizes the pans across the aural space (e.g. a random value between 0 and 1)
  • flip() - reverses the pans and their respective magnitudes across the aural space (e.g. 0.9 becomes 0.1, 0.6 becomes 0.4, etc.)
Additionally, each of these methods can take an argument specifying the time to ramp the pan (e.g. shuffle(1) would randomize the pans of all of the inputs over 1 second).

In order to visualize the inputs moving across the aural space, I created grayscale bars for each one (in the demo there are four inputs, so there are four bars). These represent each distinct sound/note. As you use PanHandler to manipulate the pans, you can simultaneously see and hear the movement of the sound.

PanHandler code: GitHub

Final project: GSM paper fortune teller

Jia, Jason, and I created a GSM paper fortune teller that sends you a text message containing a fortune. Inspired by grade school fortune telling mechanisms, this project uses a GSM module, and Arduino, and conductive paint to put a new twist on a classic form.

Jia and Jason handled most of the fabrication, while I wrote most of the code to handle the GSM module's interaction with the Arduino. We used a software serial connection to communicate with the module, and used a series of basic commands to initiate the sending of an SMS message when one of the black symbols is squeezed by the user's fingers.

One of the key things to remember is to send the AT+CMGF=1 command anytime the GSM module has restarted. This command sets the module's SMS mode to text, allowing the subsequent command to send the message, AT+CMGS, to function properly. If you don't do this, you will be unable to send messages at all.