our weekly damus meetings are actually fun, today we were discussing all the possible ways to improve discoverability and community building. I have to cut us off or we would go on for hours.

one thing I keep thinking about: with all the discovery tools we're adding (dave, network visualizer, etc): it seems what we're really doing is building a search engine for nostr. One aspect of the search engine is an AI that has a birds eye view of whats going on at any point in time, and can suggest content you might have missed that you find interesting.

There is lots of interesting content on here. If we continue to build the best tools for surfacing this content, that will go a long way to improve the value of the network as a whole.

Reply to this note

Please Login to reply.

Discussion

Are you thinking about decentralization of the search engine in some way? I know that decentralization and search are almost contradictory to each other...

the way i am thinking about it is using nip50 search to pull notes with terms matching the content, then some local algorithm is used to find the most relevant notes.

We’re very far from an advanced search engine like google, its more like a search engine for the notes in your neighborhood

Isn't there a problem with NIP-50 support? It is a while since I checked, but a lot of relays didn't implemented it. But you are absolutely right anything will be a huge improvement. If I want to find one of my older notes I just do grep over jsonl file with my old notes... So not the best user experience. 🤣

yeah even if the relay lacks it, the minute the notes hit your notedeck/nostrdb node they are nip50 searchable.

Not all relays need to have search capabilities, but its helpful so you don’t have to pull as much data when looking for specific things

Real ID is a trojan horse. I believe no one in power anymore. The whole system is rigged an must go