Todos for Google Glass
This project started off with a simple goal statement: can we create a task management ("todos") app for Google Glass"? To accomplish this, we first chose to learn more about the behaviour of "todo"-taking along with its various nuances, review past work in this field, and then proceed to designing and testing out our concepts.
This was a group project done as part of the Mobile and Ubiquitous Computing course at Georgia Tech, and I played the roles of UX Researcher and UX Designer.
To learn more about todos and todo-taking, we did several distinct steps:
- Survey: We conducted a survey to find out the various task management tools people used already, to see if there were any patterns that emerged. The results of this survey was used in the later portions of our research.
- Literature Review: We reviewed of past research on the subject of note taking and task management.
- Interviews: We then proceeded to conduct our own user study consisting of 12 semi-structured interviews to understand the domain of task management and the behaviour of todo-taking in more depth. We used the results of our survey and literature review to hone down the areas to focus on. We reviewed all the interview transcripts to extract common themes from them, and collated them together to get an aggregated understanding of task-management habits.
The research taught us that todo-taking behaviour encompassed more than just what we traditionally call "todos". We also learned a lot about the various ways people took todos, and also the variety of media they used to do so.
Variety of "Todos"
From our study, we learned quickly that what starts out as a todo-management behaviour quickly branches out into something that is more broad. We found interview participants talking about related note-taking activites which were not todo tasks, but something allied to it. Past literature has called it "notes" or "information scraps".
The kinds of notes that people often took fell in the following categories:
- Short term todos: This consisted of items like errands, shopping lists, short-term task lists etc.
- Reminders: These were exactly like short term todos, but with a deadline.
- Long term todos: These were longer term life goals related to future planning, and did not get mixed up with their short-term cousins.
- Saved information: These were handily saved pieces of information that was often required, but were not already memorized or easily memorizeable. Eg, account numbers, SSN, passwords etc.
- Reference material: These were links to "howto"s, recipes, manuals etc.
Types of "todos"
Variety of Media
How and where people jotted down information was also quite varied - it varied between people, and between the kind of information being jotted down. Here are some interesting patterns we saw:
- Handy medium: Most of the time, notes were taken using the medium that afforded the quickest way forward - on "something handy", like a piece of paper, a post-it, or even on their own palm.
- No structure: The notes were largely unorganized and unstructured.
- Purpose and longevity: The medium used to take the notes was driven by its purpose, longevity and context of use. Eg, shopping lists might be stuck on a fridge, passwords and accounts may be saved in a text file on a computer, and resources may be saved as bookmarks.
- Very visible: Notes were placed in a way that would be visible so as to remind them. "Out of sight, out of mind" was a big pattern. Placing the notes strategically (eg, a post-it on car keys) resulted in automatic "notification" like behaviour.
We brought all our findings from the interviews to our design board, and explored numerous ideas about which direction to take the note-taking app in. We realized most clearly that the space of "information scraps management" is very vast and varied, and we need to consciously focus our design to be specifically targeted to todos and reminders, and avoid the space of larger unsctuctured note taking, and long-term planning.
Some design desicions drawn from our background research were:
- To allow quick and hassle-free todo-taking, we designed the app to allow users add new notes using their voice.
- The app would processes the note's content and add contextual information like location/time, and shows reminders based on the contextual meta-data. This made the todos more useful.
- The todos would be organized as a simple reorderable list with no heirarchy
- The app would allow users to attach images to their todos in order to add more information
We came up with detailed user flows for accomplishing tasks that we thought would matter most, like taking a note, viewing their notes, and reordering them.
Wireframes with user flows
Prototyping and Testing: Iteration 1
We built a prototype of our app and had people try it out and complete tasks like taking notes, deleting a note, reordering notes, etc. To avoid having to build the actual app on Google Glass, we implemented a web-based prototype that would allow us to conduct "wizard-of-oz" testing.
The prototype requires test administrator, a laptop and a projector to run. The test user would wear an inactive Google Glass and sit in front of the projector, while Google Glass "cards" would get projected to the top right corner of the projector to the display of the actual device. When the user issues voice commands, or performs actions on the device, the test administrator can update the display based on the user's actions. This allowed us to run a "wizard-of-oz" style user test to identify some initial quirks with our task flows. The prototype we implemented can be found online.
A testing session in progressView Prototype
Prototyping and Test: Iteration 2
After some initial testing with the web based prototype, we built a real prototype of our app and tested it with users using the actual app with a real Google Glass. We assigned them a few tasks like creating a new todo, adding a picture to it, marking a todo as complete, or reordering todos in the app. 9 out of 10 testers successfully completed all the test tasks assigned to them.