User:Tomasic/Sample page

Tiramisu Transit (called "Tiramisu" for short) was a multi-year research project, conducted at Carnegie Mellon University, focused on providing Accessibility to bus transportation. The application is named for the dessert Tiramisu which translates to English as ``pick me up``.

History
Tiramisu started with research funding from the United States Department of Education through the National Institute on Disability and Rehabilitation Research in 2008. Later funding was from the National Institute on Disability, Independent Living, and Rehabilitation Research. NIDILRR is a Center within the Administration for Community Living (ACL), Department of Health and Human Services (HHS).

The original idea for the project focused on providing a service for users to report accessibility problems. However, early interviews with participants revealed a need to know when transit vehicles were too crowded to board and to know accurate transit arrival times. The project then pivoted to a crowdsourcing approach. Tiramisu began as a research project and evolved into a long-running transit information app used by everyday riders. Over the 14 year life of the project, it continued to be utilized as a research testbed until it was shut down in 2022. Over 120 researchers contributed to the project, including undergraduates, master students, PhD students, post-doctoral students, staff, and faculty

User Experience
The Tiramisu mobile application provided scheduled and real-time information on buses to the user for the city of Pittsburgh.

In the first version, real-time information was crowdsourced. Upon boarding a bus, the user would select, from a pull-down menu, the bus they had just boarded, and the mobile application would use the user's mobile phone location as a proxy for the location of the bus. The user would also indicate their destination stop. The app would warn the user when their stop was approaching and the app would stop using the phone as a proxy for the bus location when the user arrived at their stop. This information was integrated into a real-time bus arrival prediction algorithm as part of the information system, so other users would received updated bus arrival information. In addition, users provided information about how crowded the bus was as a real-time indication of bus load. The system also provided historical information about buses arrival time and bus load based on route, day of week, and time of day.

The second version of the user experience provided some social network style communication. The third version of the interface no longer emphasized crowdsourcing, since the Pittsburgh transit agency (PAAC) provided real-time bus location information. This new version focused on using machine learning to provide personalized information, with the goal of reducing user interaction effort (e.g., streamlined information, less scrolling, etc).

Research
The Tiramisu app served both as a functional mobile application and as a platform for research experiments. The first version contained an experiment examining different methods to motivate users to contribute crowdsourced data. Every user (technically, the device of the user) was randomly placed a user into one of four conditions. Each condition provided a small but crucially different user experience. The experiment showed that the condition where users are required to contribute data as a method to avoid free-riding is a successful strategy to generate more data, but also led to higher user abandonment of the application. That is, the users who continued to use the application under this condition generated more overall data compared to the condition where free-riders were permitted without limitation.

The third version of Tiramisu contained a longitudinal study of the effects of an adaptive interface. This interface was designed to observe user behavior in the application and then change the user experience in response to the observed behavior. In particular, the application monitored user transit information search behavior, modeled the buses examined by the user, and then subsequently automatically presented this information when the user next used the app. The idea of adaptive user interfaces was later expanded into a general study of the impact of this new user experience material on wireframing  design activities.

Awards
Tiramisu Transit has received several awards along with research performed as part of the project.