With the growing adoption of smart home technologies, inhabitants are faced with the challenge of making sense of the data that their homes can collect to configure automated behaviors that benefit their routines. Current commercial smart home interfaces usually provide information on individual devices instead of a more comprehensive overview of a home's behavior. To reduce the complexity of smart home data and integrate it better into inhabitants' lives, we turned to the familiar metaphor of a calendar and developed our smart home interface Casalendar. In order to investigate the concept and evaluate our goals to facilitate the understanding of smart home data, we created a prototype that we installed in two commercial smart homes for a month. The results we present in this paper are based on our analysis of user data from questionnaires, semi-structured interviews, participant-driven audio and screenshot feedback as well as logged interactions with our system. Our findings exposed advantages and disadvantages of this metaphor, emerging usage patterns, privacy concerns and challenges for information visualization. We further report on implications for design and open challenges we revealed through this work.
ABSTRACTWith the growing adoption of smart home technologies, inhabitants are increasingly faced with the challenge of making sense of the data that their homes can collect to configure functions that benefit their routines. Yet, current commercial smart home interfaces usually consist of categorical menus or spatial maps that are disconnected from the tool that people use most to capture their routines and coordinate with others -calendars. To reduce the complexity of smart home data and integrate it better into inhabitants' lives, we turned to the familiar metaphor of a calendar and developed our smart home interface Casalendar. In order to investigate the concept and evaluate our goals to facilitate the understanding of smart home data, we created a prototype that we installed in two commercial smart homes for a month. The results we present in this paper are based on our analysis of user data from questionnaires, semi-structured interviews, participant-driven audio and screenshot feedback as well as logged interactions with our system. Our findings exposed advantages and disadvantages of this metaphor, emerging usage patterns, privacy concerns and challenges for information visualization. We further report on implications for design and open challenges we revealed through this work.