This is my submission to the d3-parade hackathon https://d3js.community/d3-parade-2021.
We celebrate 10 years of D3 by looking back at the wonderful bl.ocks created over the past decade and create interesting mosaics from these visualizations.
npm run serve
The dataset used is via https://observablehq.com/@enjalot/blockbuilder-search-data?ui=classic, specifically the blocks-min dataset (https://storage.googleapis.com/bb-search-data/parsed/blocks-min.json).
There are a little over 40k bl.ocks in the dataset, of which the ones with no thumbnails are filtered out in the preprocessing, leaving about 10k bl.ocks that are used in this application.
Scrapping all available bl.ocks and save their output as images would be a nice to have, a full completed dataset.