One of Us: A Multiplayer Web-based Game for Digital Evidence Acquisition of Scripts through Crowdsourcing

Digital devices are an integral part of our lives. Through these devices, people produce and save personal data, with or without their explicit awareness. This personal digital information has been exploited by companies, but users find it hard to access and search in a uniform way, due to the heterogeneity, fragmentation of data and non-uniform access interface. By integrating and organizing this information into common kinds of everyday episodes ("scripts") that people engage in, we can help users recall and explore forgotten details of their past. However, being able to recognize such episodes in the user's personal digital information requires not only script knowledge (e.g., the steps/actions in the script), but also explicit knowledge about the digital traces potentially left behind by each of the actions. In this paper, we present "One Of Us", a web-based multiplayer game, which collects descriptions of different kinds of personal digital traces, by having players identify the digital traces that might be produced by each of the actions in a given script. We report on the results of an experimental study, which gives evidence that our game is i) enjoyable, ii) accounts for uncommon answers, iii) validates and assesses knowledge by having the players vote on other's responses - thus not requiring a second round of quality assessment, and iv) dynamically acquires new pieces of information.