BEST HARDWARE HACK FOR PENNAPPS 2015

The digital world is not designed and built for our grandparents. Many of them are struggling to keep up, so as to keep us in touch. Things are not easy for them.

Project Overview

Mago is a tangible "App store" designed for the non tech-savvy. Utilizing a Pixy camera (CMUcam5), a Intel Edison board, and a thermal printer, Mago allows users to receive real time information, connect with family - performing a wide range of actions they were not able to digitally.

Date: Jan 16 - 18, 2015
Location: University of Pennsylvania, PA
Duration: 36 hours

{ 224 }

The total number of projects submitted to PennApps 2015 Hackathon

Step #1. Define persona and challenge
With our focus set on empowering the technologically underserved population, we created a brief persona to help identify problems and potential solutions.
Some key characteristics of the persona include:
  • Uninterested in modern technology
  • Wish to stay in touch with family who are far away
  • Independent, not looking for help unless absolutely necessary
    • Our challenge here is to construct a product that is intuitive enough to be used and liked by Phyllis.
Step #2. Gather constraints
Sitting in a national hackathon, we listed out the critical constraints for the project, so as to properly define scope.
  • Time (36 hours) Little time for research & Construct MVP (minimum viable product) and add features to it
  • Limited hardware Gathering what we brought and scavenged from PennApps gadgets, we ended with: 1 Intel Edison board, 2 Arduinos, 1 Leap motion, 1 thermal printer, 1 Pixy camera, 1 radio module, 2 step motors
  • Skills set We are a team with mixed talents of design, hardware and software. Our technical strength is web development, but relatively weak on mobile development.
Step #3. Brainstorm and narrow down use scenarios
Through a quick brainstorm session, taking into account the constraints, we decided to build a system with colored cards as input, and printed receipts as output, utilizing most of our gadgets. A couple of simple use scenarios, which might be viable, were discussed next. Several that addressed the pain points of the persona were eventually picked:

- Check today's weather in the morning

- Print a map for the place to go

- Interact with children through Facebook

- Print a list of favorite contacts

- Send emergency SMS to predefined contacts

- Send text message to others

- Take a selfie and post on Instagram

- Print out reminders for today

- Scan a medicine and print out its name

- Track medicine use for today

- Buy products on Amazon

- Order pizza for visiting grandchildren

- Print out the amount of exercise done today

- Check her own stock prices

- Get latest updates and news for her neighborhood

- Listen to radio

- Print a random recipe for lunch

- Take a photo and have questions answered online

Step #4. Translating for technical implementation
With scenarios defined, we started translating them into technical details. Our principle was to make each component as modularized as possible, making collaboration all the more efficient.
As Pixy camera picks up the color of input card (each color represents a service above) and feeds the info into Edison, a corresponding service (app) residing on the Heroku server is notified. The service retrieves data from the external provider, and the server rasterizes data into printable bitmap image, sends back to Edison and the thermal printer in turn.
The whole process is of surprisingly low delay - from the moment a colored card is scanned, it takes 0.5 - 1 second for the printer to start printing.

... it's about designer's willingness and enthusiam for working within constraints. Constraints of price, of size, of strength, of balance, of surface, of time and so forth.    - Charles Eames

Project Outcome

Mago supports 6 unique applications that bring great convenience to elderly's daily life.
The colored cards require no previous training - one simple wave at the camera would do the trick.
We are also considering making the system self-sustainable: why not print out information that could be scanned later?