Meeting in the notebook: a notebook-based environment for micro-submissions in data science collaborations
This addresses collaboration challenges for data scientists, but it is incremental as it builds on existing notebook and version control concepts.
The paper tackles the problem of integrating software engineering tools into notebook-based data science workflows by introducing Assemblé, a JupyterLab environment that allows code fragments to be contributed as pull requests, with a user study involving 23 data scientists.
Developers in data science and other domains frequently use computational notebooks to create exploratory analyses and prototype models. However, they often struggle to incorporate existing software engineering tooling into these notebook-based workflows, leading to fragile development processes. We introduce Assemblé, a new development environment for collaborative data science projects, in which promising code fragments of data science pipelines can be contributed as pull requests to an upstream repository entirely from within JupyterLab, abstracting away low-level version control tool usage. We describe the design and implementation of Assemblé and report on a user study of 23 data scientists.