A promising approach that addresses some worker output issues examines the way that workers do their work rather than the output itself, using machine learning and/or visualization to predict the quality of a worker’s output from their behavior [119,120]
This process improvement idea has some interesting design implications for improving temporal qualities of SBTF data: • How is the volunteer thinking about time? • Where does temporality enter into the data collection workflow? • What metadata do they rely on? • What is their temporal sensemaking approach?