man bignostrals
The open analytics layer
for the Nostr network
Nostr is the protocol where nobody owns your output — but right now nobody's really reading it either. bigNOSTRals is the open data layer that changes that: a permissionless analytics warehouse that turns raw relay events into clean, queryable tables anyone can run against. Built for researchers, journalists, and engineers learning analytical craftsmanship on a network that belongs to everyone.
The name's a dare — the network has a spectacular nose, and we're not afraid to use it. Sniff honestly. Publish the findings.
[1] the big nose
A Lightweight Relay Listener
Not a relay — a special kind of client. It points at relays and listens to their public events of all kinds. The key design principle: stay as close as possible to the base layer Nostr protocol. Just JSON over WebSockets. No magic. It sniffs the feed so you don't have to.
[2] open source analytics engineering
A DBT-Style Data Product
An open source data product about Nostr. Making definitions, metrics and transformations visible for everyone to understand — it is a great way to learn analytics engineering — fork, build on and contribute.
[3] bigquery public dataset
Queryable Public Data
Today available upon request, eventually made fully public. The foundation for spam detection, recommendation engines, academic research, and open-source AI datasets.
Data coverage note: full event history through early August 2026; the mid-August window is being replayed from the public relay network and is available as a representative sample. Live ingestion resumed and is continuous going forward.
$ questions bignostrals can answer
What's the profile of what is posted in Nostr today?
How is bot-generated content in Nostr actually engaged with?
What kind of posts get most zaps?
Which podcasts are trending in a given topic?
What livestreams brought in more new users / npubs?
Has Npub X reached the proverbial 1000 true fans?
What happened during the World Cup in Nostr?
What content / community is emerging in Nostr?
In principle all of these questions can be answered with the data and data product being put forth here.
$ how to contribute
Developers
- ❯ Make bigNOSTRals connect to other data warehouses (ClickHouse, Snowflake)
- ❯ Harden the relay listener for production relay loads
- ❯ Get historical (old) events without breaking the bank
- ❯ Figure out relay signaling — the listener should never overload a relay
- ❯ Handle delete requests (kind:5) across relay topology
Vibe Coders
- ❯ Start on the dbt project — the least fragile and most novel part
- ❯ Fork the models, define new metrics, experiment with transformations
- ❯ AI is exceedingly good at working with dbt — go crazy
Non-Technical
- ❯ Think about what questions you #asknostr
- ❯ Help make Nostr more attractive to content creators and entrepreneurs
- ❯ Bring curiosity — it is a superpower