Apollo#
This guide covers topics related to effective and inclusive collaboration.
Data science is defined by its interdisciplinarity. Our work can only reach its highest potential if there are diverse teams of people involved in designing and delivering the research or product.
There are many different skills required to work well in groups with a wide range of expertise. In this guide, we welcome contributions in developing guidance on the following (but not limited to) topics:
Designing a project that welcomes contributions
Distributed collaboration on GitHub
Reviewing team member’s contributions
Remote working
Running an inclusive event
Chairing a meeting
Defining explicit expectations
Participatory co-creation
Check out our contributing guidelines to get involved.