OSM Planet Torrent Location Indexer

Index OSM planet torrent by geographic location - fetch only what you need!

Problem

OSM planet file is 85 GB. You only need data for specific locations.

Solution

Map lat/lon coordinates to torrent pieces. Download only needed pieces (4 MB each) instead of entire planet.

Example: Ramanujan's Journey

From Kumbakonam (India) to Cambridge (UK):

Planet: 85 GB, 21,763 pieces
Locations → Pieces:
- Kumbakonam (10.96°N, 79.39°E) → Piece 13,668 (shard 36)
- Chennai (13.08°N, 80.27°E) → Piece 14,137 (shard 8)
- London (51.51°N, 0.13°W) → Piece 16,793 (shard 37)
- Cambridge (52.21°N, 0.12°E) → Piece 16,945 (shard 47)
- Trinity College → Piece 16,945 (shard 47)

Result: 5 pieces × 4 MB = 20 MB instead of 85 GB!

Usage

# Use default user (ramanujan)
cargo run

# Use your own user
cargo run myuser

# User locations stored in userdir/

User Locations

Create userdir/yourname.json:

{
  "user": "yourname",
  "wikidata_user": "Q12345",
  "osm_user": "your_osm_username",
  "locations": [
    {
      "name": "YourCity",
      "lat": 40.7128,
      "lon": -74.0060,
      "wikidata": "Q60",
      "osm_node": 123456
    }
  ]
}

Wikidata Integration

• wikidata_user: Your Wikidata Q entity (e.g., Q185493 for Ramanujan)
• wikidata: Location Q entity (e.g., Q84 for London)
• Links to semantic web of knowledge

OSM Integration

• osm_user: Your OpenStreetMap username
• osm_node: Specific OSM node ID for location

How It Works

1. Download torrent metadata (437 KB, not the full 85 GB)
2. Map locations to pieces using golden ratio distribution:

   lat_norm = (lat + 90.0) / 180.0
   lon_norm = (lon + 180.0) / 360.0
   combined = (lat_norm * φ + lon_norm) % 1.0
   piece_idx = (combined * num_pieces) as usize

3. Shard by Monster prime (mod 71) for distributed storage
4. Selective download using aria2c/transmission

Monster Group Sharding

Pieces are sharded using the largest Monster prime (71):

• Enables distributed storage across 71 shards
• Each shard can be independently fetched/cached
• Follows Monster Group structure (15 primes)

Add Your Locations

Edit src/main.rs:

const LOCATIONS: &[Location] = &[
    Location { name: "YourCity", lat: 40.7128, lon: -74.0060 },
    // Add more...
];

Requirements

• Nix with flakes enabled
• Or: Rust toolchain with OpenSSL

Build Without Nix

cargo build --release
./target/release/osm-planet-torrent

Selective Download

Once you have the index, use aria2c to fetch specific pieces:

# Download only piece 16945 (Cambridge)
aria2c --select-file=16945 osm-planet.torrent

Integration

Part of CICADIA-71 bootstrap system:

• 10-step journey: Monster Group → Ramanujan → Temple
• Connects abstract math to physical locations
• Proves system can ground itself in reality

Files

• src/main.rs - Location indexer
• src/shard.rs - Monster prime sharding
• flake.nix - Pure Nix build
• Cargo.toml - Rust dependencies (rustls, no native-tls)

License

MIT

Related

• CICADIA-71 Bootstrap
• OpenStreetMap
• Monster Group

---

"From 85 GB to 20 MB - fetch only what you need!" 🌍🕉️