Day 002 — Music Charts

Personal music charts spanning 20 years of listening history (2005-2026).

Charts

• Heart Top 100 — Songs ranked by intentional plays, penalizing instant skips and algorithmic inflation. Anomaly detection normalizes years with high skip rates.
• Algo Top 100 — Raw play count ranking with correction column showing where each song lands in the organic scoring.
• Decade Charts — Top 100 songs and albums per decade, styled to match the era: classic billboard (pre-2000), web 1.0 (2000s), flat design (2010s), Spotify-style (2020s).
• Yearly Charts — Year-end top songs and albums for each year, automatically styled by era.
• Newcomer Top 20 — Artists and tracks first heard in the last 12 months.

Use with your own data

You can generate your own charts from your personal listening history.

Prerequisites

• Python 3.10+
• pip install duckdb requests

1. Get your data

Spotify: Request your Extended Streaming History at spotify.com/account/privacy. Place the Streaming_History_Audio_*.json files in raw-data/spotify/.

Last.fm / ListenBrainz: If you have a Last.fm account, you can import your scrobble history into ListenBrainz (there's a built-in Last.fm importer). Then export your listens and place the monthly JSONL files in raw-data/listenbrainz/listens/YEAR/MONTH.jsonl. You can use Spotify, ListenBrainz, or both.

2. Optional: API enrichment

Create a .env file for additional metadata:

LASTFM_API_KEY=your_key_here
LASTFM_SHARED_SECRET=your_secret_here
SPOTIFY_CLIENT_ID=your_id_here
SPOTIFY_CLIENT_SECRET=your_secret_here

Get a free Last.fm API key at last.fm/api/account/create. Get Spotify credentials at developer.spotify.com.

Then run the enrichment scripts (optional, improves genre charts):

python scripts/ingest.py                # raw data → DuckDB
python scripts/enrich_musicbrainz.py    # genres, tags, labels (~1h)
python scripts/enrich_wikidata.py       # coordinates, founding years (~2min)
python scripts/enrich_lastfm.py         # global popularity, similar artists
python scripts/enrich_spotify.py        # artist genres, track popularity

3. Generate charts

python scripts/export_billboard_data.py   # → charts/*.json

The charts/ directory now contains the JSON data files. Open index.html in a browser (via any local HTTP server) to view your charts.

Data normalization

Edit scripts/normalize.py to customize:

• EXCLUDE_ARTISTS — audiobooks, podcasts, non-music to filter out
• ARTIST_MERGES — fix variant spellings (e.g., "José González" / "Jose Gonzalez")
• TRACK_RENAMES — fix mis-scrobbles and merge duplicate tracks

Stack

• Vanilla JS + Google Fonts for themed chart rendering
• Pre-computed JSON data (~444 KB total)
• DuckDB + Python for local data processing
• Static deployment on Cloudflare Pages