Assessing MetPy Mondays Impact on the Community to Address User Needs

MetPy is an open-source Python package focused on meteorological applications. Since 2017, the MetPy Mondays video series on the Unidata Program Center’s YouTube channel has provided weekly code tutorials showcasing functionalities of MetPy and general programming tips. Following persistent demand from series viewers, the code described in the videos was made available in a set of Jupyter notebook files in a publicly accessible Github repository. Having the MetPy Mondays codebase available also facilitates code maintenance and long-term reproducibility of the tutorials. To leverage this rich archive, the present work aims to improve navigation on the wide array of topics already covered in the tutorials while tailoring future videos ideas and approaches to community needs. Using a search of key words in the videos transcripts and manual analysis of the code snippets, the tutorials are organized and cross-referenced by identifying data types, sources, and tool or method addressed in the tutorial. The expected outcome is to simplify finding specific resources and variations in their use. From MetPy users’ surveys, we gauge MetPy Mondays’ broad relevance amongst other training modalities. To understand the audience, we analyze the user activity metrics from YouTube data API over specific time periods such as video view counts, ratings (likes), comments and subscribers gained to estimate engagement and preferences on each video. Considering the hundreds of videos in the MetPy Mondays library, this work presents insights into their evolving impact in the community. The development of a data-driven tool to improve asynchronous video learning experience is also discussed.