Revealing information about a room with an interactive data visualization.

My Role

Development, Interface, Visualization

What it is all about

In one of our last semesters, we wanted to build a data visualization system that operates autonomously to collect and display information about the space we're working in. Our program has an open attic where lectures take place and students spend the majority of their time completing their projects. The project is supposed for students to gain more insight in the dynamic of our working space.


Making the invisible visible

When thinking about how our collected data could be displayed in a readable and aesthetic way we had to keep in mind that we not only wanted them to work as a single object but also next to each other as a weekly overview. On the individual day you can see more detailed information about each day and can try to decipher what was going on that day. Maybe it was really good weather so there were unusually few students working, or maybe finals are approaching so students are working until late at night.

Showing the bigger picture

The home screen shows an overview over the last seven days. The height of each diagram represents how many people were working there on that day, so you can easily compare them to each other. Also there is a flow of particles, that represents the students, going through the week. You can discover this data freely by scaling and rotating. From here you can select a specific day and get more detailed information about it.

Making it tangible

In using a leap motion as input we saw the possibility in enabling a more intuitive and tangible interaction than with a mouse and keyboard. The cursor suggests what you can do by using arrows to hint in a direction to move.

Where it begins

With a NodeMCU we collected temperature, brightness and noise level and saved it to a database. Every night this database a json file with the data of the last seven days is created and stored online, so we can access it with our custom software. Additionally we're processing a file that contains the sunrise and sunset data for every day of the year.


What I learned

Surprisingly enough - at least to me - we were able to complete this project in a relatively short period of about two weeks. We had split the work in such a way that everybody was able to focus on one aspect of the whole experience, and then bringing everything together. Short feedback loops and prototyping helped us move fast and efficient.