In his dissertation, Kato proposed to use photos and videos as graphical representations of the real-world input and output (I/O) data in text-based IDEs. Photos are used to capture images describing the real world. Videos are a time-series of photos, and can be used to capture movement in the real world. The core assumption of his work was that such media can be integrated into text-based IDEs and used to support Programming with Example.
The goal of his work was to examine this assumption by investigating the use of the graphical representations. He investigated the use of photos and videos as graphical representations of a program by building and evaluating experimental IDEs. The IDEs integrated graphical representations to complement textual programming, including Picode, DejaVu and Visionsketch.