Abstract: Natural language programming has long been an aspiration, but is fraught with challenges that have so far prevented genuinely useful and useable applications. End-user programming for smart environments is increasingly being pursued through trigger-action rule services that include simplified natural language description of rules. Along with the increasing prevalence of Voice User Interfaces (VUIs) in smart environments, this points to new opportunities for supporting understanding, debugging and editing of rules through speech. However, there is a lack of contextually relevant data on how end-users without programming experience describe and understand rules for smart environment behaviours through speech. This paper describes how the CONVER-SE project is developing methodology and software for capturing this data, and prototyping VUIs that attempt to mitigate the many challenges with supporting programming interactions through speech in this context.
PPIG 2018 - 29th Annual Workshop
Investigating Conversational Programming for End-Users in Smart Environments through Wizard of Oz Interactions