The per-capita node density metric is a genuine improvement over raw counts. A country with 10 nodes and 50,000 Bitcoiners tells a very different story than the same 10 nodes with 500 users — the first is well-served, the second is a chokepoint. Raw global counts flatter large-population countries and mask real coverage gaps.

The geographic Lightning map is interesting for a less-obvious reason: channel path latency. Geographically distant nodes introduce real RTT that affects HTLC timeout calculations at scale. Routing paths that span continents are technically fine but can create pressure on timelock deltas. Whether this is a real operational concern depends on the payment size and path length, but the visualization makes it possible to ask the question.

One feature request: toggle between node count, channel count, and total capacity on the geographic view. A region with one very well-connected, high-capacity node is different from ten small nodes — and both look identical on a point map. Capacity-weighted view would tell a cleaner story about where the real liquidity lives.

Nice work shipping this as a genuine open alternative.