Open Game Data

Open Game Data is an open-source, community-maintained pipeline for collecting, processing, and sharing gameplay telemetry from educational games. The project provides logging libraries for Unity and JavaScript games, a shared event-data standard, and automated pipelines that move raw play events into analysis-ready features — turning millions of daily gameplay events into a public resource researchers can use to study how kids learn through games.

The project is a collaboration between CMU and the Field Day Lab at UW–Madison, with a growing network of contributors spanning educational game development, games user research (via ACM CHI PLAY), and learning engineering (via CMU’s LearnLab). That breadth is deliberate: the project’s data practices and infrastructure are shaped by the range of communities that need to use them.

Alongside the technical work, I host Open Game Data Open Office Hours, an informal recurring forum where researchers and practitioners share in-progress work and talk through the practical, often messy realities of working with game data. Sessions prioritize tacit knowledge over polished results, and each one is recorded and shared, gradually building a collective knowledge base for the community.

The project’s original goal was a common data standard; through years of workshops and community discussion, that’s grown into a broader ambition to foster an ecosystem where shared approaches to data, analysis pipelines, and hosting infrastructure can emerge organically, supported by lightweight infrastructure and sustained community engagement.