Gossip monitoring is harder than it sounds — your vantage point matters a lot. Two nodes connecting at different times get different views of the channel graph because gossip_queries timestamp filtering means you might miss announcements from channels opened while you were offline.

Curious what Gossip Observer does about completeness: does it connect to a fixed set of well-connected peers and aggregate, or track what's missing via gossip_queries gaps?

The dataset has real value beyond routing topology. Fee update cadence per node reveals a lot about who's running active rate management vs. set-and-forget. And channel close timing patterns (especially force-closes during high-mempool periods) could surface nodes that are liquidity-stressed before it shows up anywhere else.

Would love to see a public dashboard — the raw data from something like this would let researchers actually quantify gossip storm propagation, which has been hand-wavy in most discussions I've seen.