01 July 2026
One of our first big post-migration improvements is to update our data visualization and mapping libraries. We’re now using Leaflet and Vega-Altair to display data.
What's New
More flexible data visualization tools will enable better performance and more creativity
Over the past few years, including during the migration, we’ve been using Plotly, a library that offers both mapping and data visualization. But as we’ve started to plan future features for Pulse, we needed something more specialized, which has meant choosing different libraries for each function.
So, for maps, we’ve chosen Leaflet, a leading open source project for geodata and interactive maps. It has support for tile layers, vector layers, markers, pop-ups, image overlays, and GeoJSON.
And we’re now using Vega-Altair, a Python library for data visualization. It’s built on top of the Vega-Lite JSON grammar. Being native Python makes it easy to integrate with our data pipelines, and transform that data into the Vega-Lite syntax for display. This also opens up other opportunities for improved accessibility for our data through plugins like Olli.
Both of them will help us improve performance, since the combined size of both, along with associated plugins and custom code, is on average 50% smaller than Plotly.