Hi All, I wanted to share the following white paper <https://nbgallery.github.io/health_paper.html> (abstract below) which details our team's work to systematically measure the code health of Jupyter notebooks from within an nbgallery instance. As background, nbgallery <https://github.com/nbgallery/nbgallery>is an enterprise Jupyter notebook sharing and collaboration platform developed within the Department of Defense. Our team operates in a unique environment which requires us to take some interesting and unconventional approaches to evaluating code health. For instance, traditional unit testing in our environment is a challenge since we use dynamic data sets with row-level security (so the data landscape shifts by the day and by the user), and many of our notebook contributors are not typical software developers with experience or interest in developing unit tests. We think some of our practical approaches arising from evaluating code health in a complex notebook environment might extend to projects like JupyterHub and Binder.
Please check out the paper <https://nbgallery.github.io/health_paper.html> and let us know if you have any questions/comments. Thanks! Dave *Systems that support user-developed code are faced with a key challenge: understanding the health of that code, which we define as the expectation that existing code will function properly in the current environment. The growing popularity of Jupyter notebooks has led to the development of publishing and execution platforms such as the open-source nbgallery <https://github.com/nbgallery/nbgallery> project. Users of nbgallery would like to understand when they can expect a notebook to work, and notebook authors may wish to monitor the execution of their code and be informed of errors. This paper describes our initial efforts to measure code health in a corpus of notebooks within an instance of nbgallery. Our vision is that this work will help address problems that arise from user-developed code and motivate further study in systems beyond Jupyter and nbgallery.* -- You received this message because you are subscribed to the Google Groups "Project Jupyter" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jupyter/c755afc5-30ad-4add-8df1-d1a790f0ee5b%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
