I was summarizing my last #soveng demo day, and instead of pushing everything to GitHub, I thought about making my first-ever Nostr long-form post. So here it is: about geohash filtering in Nostr relays.
In the age of lavish LLMs, my writing style tends to go to the other extreme, making everything as compact as I can. I hope this form of text compression is lossless, but if not, you can always use LLMs to reconstruct the original π
https://primal.net/a/naddr1qvzqqqr4gupzq935hpa4ln755mz07tedu969pnxwgmu6hc9s9fccwmzedmqkt0ldqyfhwumn8ghj7ur4wfcxcetsv9njuetn9uq32amnwvaz7tmjv4kxz7fwv3sk6atn9e5k7tcqyenk2mmgv9eksttswfjkv6tc94nxjmr5v4ez6en0wgkkummnw3ez6un9d3shjuc4duwz4
Login to reply
Replies (2)
> Current Nostr relays canβt efficiently query hierarchical or spatial data like geohashes. This limits map applications that need to find events by region and zoom depth. The proposed solution adds geohash prefix filtering using a caret (^) operator. The prefix concept could also extend to other structured tags, such as dates, to support broader hierarchical queries.
π