david on Nostr: I think and discuss lots of things about which I know very little. Otherwise I’d ...
I think and discuss lots of things about which I know very little. Otherwise I’d remain completely ignorant about everything forever.
If I render an opinion or judgment about something, I do my best to indicate my level of confidence. Have I spent a lifetime studying the thing? 2 minutes skimming some random document? Somewhere in between?
I know a little, but not a lot about pubky. I’d love to learn more and be pleasantly surprised about what you’re building. I’m already happy to see you’re incorporating neo4j. Which means you will have easy access to neo4j graph data science library, including centrality algorithms like PageRank. Using neo4j, you can today calculate personalized PageRank and GrapeRank web of trust scores on close to 200k nostr pubkeys, and you can export those scores in a format that is easily accessible by any nostr client. The PageRank part uses neo4j GDS and takes roughly 15 seconds.
So here’s what I want to know: How long before I can run personalized centrality algorithms on pubky? Using real data with lots of real users?
If I render an opinion or judgment about something, I do my best to indicate my level of confidence. Have I spent a lifetime studying the thing? 2 minutes skimming some random document? Somewhere in between?
I know a little, but not a lot about pubky. I’d love to learn more and be pleasantly surprised about what you’re building. I’m already happy to see you’re incorporating neo4j. Which means you will have easy access to neo4j graph data science library, including centrality algorithms like PageRank. Using neo4j, you can today calculate personalized PageRank and GrapeRank web of trust scores on close to 200k nostr pubkeys, and you can export those scores in a format that is easily accessible by any nostr client. The PageRank part uses neo4j GDS and takes roughly 15 seconds.
So here’s what I want to know: How long before I can run personalized centrality algorithms on pubky? Using real data with lots of real users?