I believe we will eventually have the tools to use methods exactly as you describe over nostr, as well as other methods along similar lines. But a lot of infrastructure needs to be built first. For starters: I will want my Grapevine (my WoT) to tell me that npub_ew, which claims to be Ed Witten, is in fact the real Ed Witten and not some bot impersonator. Next, I will want my Grapevine to give me a list of npubs who are trusted in any given context, with the context in this case being “to review papers in theoretical physics” or something along those lines. But then the question arises: who curates the lists of contexts? Who decides that the category: ToE should a subcategory of physics? The answer needs to be: my Grapevine. Because I sure as hell don’t want the World Wide Web Consortium or any other centralized body to be in charge of defining contexts for me. This is why I am building Brainstorm, a personized WoT nostr relay and knowledge graph curated with the assistance of your Grapevine. Coming along slowly but surely! 😊

Replies (1)

Great to hear you are working on this problem! I think these problem will get a lot easier with just more data. You can check @Pip the WoT guy's it works amazingly well for finding real people who have a decent amount of connections. With more data, algorithms will be able to figure thinks out pretty reliably just doing some kind of "unsupervised learning".