Thanks in particular to the great German podcast Das Coronavirus-Update from NDR with two fantastic virologists (Christian Drosten and Sandra Ciesek) I gained a better understanding of the pandemic and this coronavirus (SARS-CoV-2) and learned quite a lot new stuff. Maybe because of this, I felt for a while that there was a lack here (in Québec/Montréal) about how the data is presented/visualized. But I had this realization “late” (Fall 2020) and didn’t have past data so I didn’t want to start something from scratch.
One day I randomly came across a repository on GitHub of a dashboard for Montreal. Unfortunately, the website hadn’t been updated in a while so I reached out to Jeremy Moreau who created it. We had a chat and I started helping him adding the data, adjusting to new data formats and automating the data retrieval and processing. Adding the missing data from the past was possible thanks to the Wayback Machine and (for the most part) to the hard work by Jean-Paul Soucy archiving all Canada-wide data for the COVID-19 Canada Open Data Working Group.
Besides getting the data up-to-date we also improved the mini boxes on top and updated some of the charts. For example, the map of Montréal boroughs now is coloured based on the 7-day incidence per 100 000 people (the number of new cases in the last 7 days adjusted for population size). Due to changes in the pandemic situation, we also added charts for vaccination and variants of concern.
A few months before I had started using pandas for a client project. Great timing since this was already in use here. So I could focus on wrangling all the different data formats. I hadn’t used Plotly/Dash before but I was able to quickly figure things out and make contributions to the different charts.
The dashboard is available in French, English, Mandarin and Spanish. And it is open source so contributions are more than welcome.