Michaela R. Reisinger, Johann Schrammel, Stefan Suette, Peter Fröhlich
pp. 226 – 245, download
(https://doi.org/10.55612/s-5002-045-010)
Abstract
End-user programming concepts are increasingly employed in smart home research to address the growing complexity of controlling smart home environments. Different approaches and visual styles of end-user programming have been proposed and implemented within this context. Smart home control does however not only necessitate end-user programming but also understanding and modifying existing program structures. In this study, we compare three different approaches regarding their suitability for this application context with a specific focus on intelligibility and modification performance. We conducted an empirical study with 39 users performing three types of tasks (understanding, configuring, and programming), using three different approaches for end-user programming (form-filling, data-flow, and grid-canvas). The results of our study found no significant differences regarding the intelligibility of the three different implementations but clear differences in the subjective preference of users as well as configuration and programming performance.
Keywords: end-user programming, end-user development, smart home, home automation, smart environments, rule-based processing, trigger-action programming