« inQuire | all research | Controversy Visualization »


Amusica is a socially interactive song selection system designed for individuals and groups in bars or nightclubs. It allows patrons to submit a playlist of songs to a communal pool and to vote on others—selections throughout the night. It enables the sharing of personal media among friends and strangers in public spaces, creating new social bonds, strengthening existing relationships, and broadening musical horizons.

January 2003 to May 2003

Graduate Design Studio II
School of Design, Carnegie Mellon University
[This research project was a collaborative effort between myself and five fellow classmates.]

[— presentation —] 6MB (PDF)
[— research video —] 115MB (AVI)
[— scenario —] 140MB (AVI)
[— project website —]

At home, technology helps us take control of our music—an unlimited selection available from many sources, musical tastes becoming increasingly eclectic, and new technology facilitates musical preference sharing. But when we go out at night jukeboxes and DJs take over bars offer a limited selection of music, music selections reflect just one person's taste, and there tend to be very few opportunities for social interaction through music selection.

In this class project sponsored by Microsoft, we developed a design solution focused on enhancing an environment such as a bar or nightclub that enables customers to contribute their musical tastes without the limitation of a jukebox selection or a DJ’s musical preferences. The result is a broader array of musical options and more opportunities for customers to learn about new music that provides a medium for self-expression among customers and retains the excitement and newness of existing environments.

Once we were given our theme—sharing personal media—all of the team members sketched out a variety of ideas and then selected our direction. Once we narrowed our topic to music sharing, and decided that it would be a replacement for the current jukebox and DJ experience, we faced a a number of challenging problems. One of the most pressing issues was the development of a system that would enable participants to contribute music, regardless of musical taste, and still be heard. We knew that voting would be the mechanism, but one of my most important contributions was a system that ensured that every participant had a fair chance at having their music heard. My solution was a tiered system that required participants to vote on 3 songs (one each from three sets of three songs), and from that vote 5 songs would be selected for the next playlist. The top 3 vote getters each received a slot, the fourth slot went to highest ranking second-place vote getter from each set, and the fifth slot went to the losing song recommended by the person who had gone the longest without hearing one of their songs played. This system dictated the form factors required for the handheld and tabletop devices, each needing to enable the selection of 3 songs, one each from three sets of three songs.