Share.

8 Comments

  1. **How I built it (The “Beautiful” Data part):**

    * **Data Sources:** Music data is fetched via YouTube Charts APIs. Map geometries are sourced from OpenStreetMap (Nominatim).
    * **Data Pipeline:** I built a pipeline using **Python and GeoPandas** to discover locations via autocomplete APIs and fetch chart data in parallel (using 10 parallel workers to handle the ~90k rows of data).
    * **Frontend:** The map is rendered using **MapLibre GL JS**, handling the `combined_map.geojson` and `charts_tracks.csv` files on the fly. I used topological simplification to ensure the map remains performant on web browsers.

    **Key Features:**

    * **Granularity:** Unlike most music maps that only show countries, this includes ~4,700 locations including cities and sub-regions.
    * **Beyond #1:** You can explore the full Top 20 context for each specific territory.
    * **Non-overlapping Geometry:** I used a custom global geometry that preserves local specificity while allowing instant interaction.

  2. ImNotHereToMakeBFFs on

    Interesting that the Brazilian Phonk artist (ZXKAI) is #1 all over Russia but not even charting in Brazil itself.

    One of the few actual good posts from this sub. Cool project.

  3. nice map. Italian top 3 songs are the top three position in Festival di Sanremo (Italian music biggest show of the year). The first one “per sempre sì” will go to eurovision this year.