Music is intrinsically linked to our social lives. As more music becomes available through streaming services, deciding what music is appropriate for social events becomes increasingly challenging and nuanced. While prior work has considered the social role of music and the creation of music playlists for user contexts, how individuals utilize music to create social contexts is an area that has largely gone unexplored. To investigate this topic, we created and evaluated a prototype music recommender system called Reflektor. Reflektor interactively visualizes users' chat conversations to generate music playlists. Our analysis of user conversations with Reflektor uncovered distinct strategies participants use to create the ambiance and conduct for social contexts. Our findings help to illuminate mismatches in the way metadata and recommendation systems align with user strategies to create social context. We elaborate on these strategies and discuss design implications for future collaborative music recommender systems.

Context has long been considered an important component of design, but as technology becomes more capable of inferring the user's behavior and environment, what constitutes context has become an increasingly pressing concern to designers. Although design frameworks and models have been proposed for context-aware computing systems, there has not yet been research that focuses on understanding context empirically from the perspective of the designer. To address this, we present an analysis of 11 in-depth interviews we conducted with designers of a variety of context-aware systems. Our analysis of the artifacts and interviews reveal five concerns designers address in their work. Furthermore, we present a process model that illustrates how context-aware system designers address these concerns. Our findings demonstrate the central role that designers' views of context plays in (a) framing a design space, (b) encoding the relevant features of context, (c) unifying possible solutions within that design space, and (d) evaluating designs. These findings suggest that context is a dynamic concept that evolves over the course of a design project, generally from a more phenomenological perspective toward a positivist interpretation. This, and the process by which it occurs, contributes insight into context-aware design with implications for both academics and practitioners.

Generating potential design ideas through ideation often benefits from the spontaneity of random ideas. Having potential users participate in this process can be beneficial, but is often difficult to implement. We present a new method for generating design ideas with potential users. The method uses scenarios with missing words, which potential users fill in to generate ideas for features and attributes of new technology designs, similar to the children's game of Mad Libs. We developed three different formats of DesignLibs, including 1) "Mad Libs-style": blanks presented before seeing the scenario, 2) "Fill-in-the-Blanks": blanks presented within the context of the scenario, and 3) "Q&A": blanks presented as questions and answers. We found that Design-Libs generated a number of new ideas, with the Fill-in-the-Blanks method providing the highest ratings for usefulness, feasibility, and diversity of answers. All three formats provided equal ratings for creativity.

Sleep is a basic physiological process essential for good health. However, 40 million people in the U.S. are diagnosed with sleep disorders, with many more undiagnosed. To help address this problem, we developed an application, ShutEye, which provides a peripheral display on the wall-paper of the user's mobile phone to promote awareness about recommended activities that promote good sleep quality. Based on preferences about the user's desired bed-time and activities' for example - consuming caffeine or performing vigorous exercise - ShutEye displays guidance about when engaging in those activities is likely to affect sleep without requiring any explicit interaction from the user. In this paper, we describe ShutEye and results from a four-week field study with 12 participants. Results indicate that a simple, recommendation-based peripheral display can be a very low-effort but still effective method for improving awareness of healthy sleep habits. We also provide recommendations about designing peripheral displays and extend insights for designing health-based mobile applications.