Scale is hard, but for small-to-medium sized clusters, you can use your own node, or a friends'. Now you have your own copy of the data, and control over latenc

Scale is hard, but for small-to-medium sized clusters, you can use your own node, or a friends'. Now you have your own copy of the data, and control over latency. The key insight with nostr is the multi-master architecture.