Zhuoshi Xie is a

Sidetrack An app that assists in meandering

Sidetrack, an independent project, is an app that helps people discover new things in their surroundings, resulting from personal desires for tools of navigation in cities.

Interface demo. Best consumed with sound.

Research on Navigation

People navigating cities

How do people navigate cities? How does this differ between locals and tourists? How well do people understand the cities they navigate? What tools do people use? What methods do people prefer? What factors do people consider?

Interview notes.

I interviewed 10 participants, and the key takeaways were:

  • Locals don't necessarily understand a city better than tourists. Often tourists will do research before they get to a city, and have a good mental model of it.
  • People usually use a combination of transportation methods, and have one or two they don't like. The one that they don't like differs.
  • One big unsolved challenge for navigation is figuring out the side of street.
  • Almost everybody relies heavily on Google Maps.
  • Many factors affect the transportation people choose:
    • Time of day — cars and cabs are more popular at night
    • Existence of others — whether or not splitting a cab is efficient, whether or not everyone has cash, the age of some members
    • Cost — monthly passes, cheapest options
    • Current location and proximity to major stations
    • Weather and time
    • Intent to exercise
    • Having to change methods of transportation
    • Reliability of data / availability of real-time data

Navigation apps

Informal competitive analysis led to discovering that there are a lot of existing options for navigation, many of which fill the gaps that Google Maps doesn't fill. Combined with the interview research, I realized that there was little use in creating yet another app for navigation, especially when people tended to stick to Google Maps even when there are so many other options available, some of which are better than Google Maps in certain situations.

Research on Exploration

Earlier interviews on navigation revealed that people don't necessarily always follow the map. I then wanted to probe further into how people veered off their routes, and how they explored cities.

Exploring cities

What prompts exploration? What guides exploration after starting? When do people explore? How do people explore?

  • People differ in how they prefer to explore. Some will never do it with friends, others are scared to explore alone.
  • Exploration is largely the result of free time, such as when they don't have a destination to hurry to, or they have time before needing to be at the next destination.
  • Exploration is guided by “interested” things, but the definition of what is interesting differs per person.
  • The intent to explore needs to be there first.

There seemed to be a need for something that would prompt people to explore at appropriate times, and to guide them in discovering what they find interesting. That is the route I went in.

Exploration apps

I found that there were a lot of apps that provided information on what to do and where to go, as well as what certain landmarks are, but none that assisted in discovery, in finding that sense of delight when you discover something for yourself.

A brainstorm on what the app should do.

Defining the app

Sidetrack aims to find routes based on the user's preferences on transportation and their definition of what's interesting, learning from a combination of user profile data and other app users' route data.

The intent is not to help the user be efficient and get things done, but to help them explore their surroundings. One challenge is to design the interface in a way that helps users be on the app as little as possible—the goal is to get people to discover things on their own, not to hand them all the information.

Assumptions about the audience / use cases

  • People who need a reminder to explore when the opportunity is right
  • People who need recommendations for where to go
  • People who are waiting and have time to kill
  • People who are going somewhere and have time to kill

Situations when people might open the app

  • Already exploring and want assistance
  • Doesn't need to go anywhere and wants to explore
  • Going somewhere familiar and wants a new route
  • In a new area and wants to take a good route
  • Doesn't want to explore
  • Responding to a notification
Rough wireframes and notes
Interaction flow

Exploration interests

Different people are drawn to different things when exploring. In order to build personalized routes where the user is likely to find something they like, there is a settings section on the app where they indicate their exploration interests. While asking users to self-identify interests may not be the most accurate way to gather this information, it's the most efficient way to do so, and the options can match with the data that the system has. In addition, personalized interests would not be the only factor in producing routes; aggregate user data would also be considered.

I sent a survey to find out what people might be interested in, and formed this list based on their responses.

Visual language

Thinking about the goals of the app, getting people to explore and experience the real world, I wanted the visual language to support those goals—to almost elicit curiosity. At the same time, I wanted a flat, minimal aesthetic, just enough to communicate the information, but not too much for someone to want to stay in the app constantly.

Design Testing

When you hand people a route that goes off the quickest path, and they're in the mood for exploration, what do they do? How much do they veer off the route? I did a bit of research to find out.

I asked for volunteers, made routes based on where they needed to go, asked them to let their interests lead them, then followed them to see what they would do.

KC was making a trip to a store that was a 15 minute bus ride away. I gave her a path where she would get off a few stops earlier and walk by a park and cemetery to arrive at the destination. Here are some of the things I learned and/or noticed:

  • Inconsistencies in bus times need to be accounted for
  • The detail view should have distances to help with knowing when to expect to turn
  • She was interested in a cemetery but did not want to go around to find the entrance
  • She followed the route exactly
  • When delighted by something (she found a mausoleum), she wanted to find similar things (more mausoleums)
  • Expressed interest in taking the same route again

PZ had free time and just wanted to quickly explore her surroundings. I gave her a route that took her to unfrequented areas next to campus. Notes about her trip:

  • It wasn't clear which was starting and which was ending on the route map
  • She completely ignored the route summary
  • She followed the general direction of the route at first, but completely just let her interests take her, which took up the rest of the time
  • She was constantly trying to connect what she was seeing with what she's seen in pictures
  • She expressed enjoyment, that it felt like off campus
  • She said she was afraid of trespassing, but completely ignored a trespassing sign elsewhere

Continuing work

  • More research on how people might use this app, and how the interface should change based on their behaviors
  • Onboarding, to introduce the user to the app and to set up preliminary settings like city and interests
  • Sharing a generated route
  • Planning routes ahead of time
  • Taking advantage of different screen sizes

← Back to all projects