Why We Built bigNOSTRals: The Case for Protocol-Level Analytics

Nostr ("Notes and Other Stuff Transmitted by Relays") is the simplest open protocol for a censorship-resistant global social network. It does not rely on a trusted central server, uses cryptographic keys and signatures for tamper-proofing, and is not P2P-based — so it actually works.

But the protocol has a blind spot: there is no easy way to see what is happening across the entire network. Relay operators know their own traffic, and clients know what they have fetched. No one has a global view. That is where bigNOSTRals comes in.

The Problem: Flying Blind

If you are a researcher, a platform builder, or just a curious mind, you want to ask questions like:

• What is the actual size of the active user base?
• How is bot content distributed and engaged with?
• Which content verticals are growing fastest?
• What are the network topology patterns between relays?

Without an open analytics pipeline, every team has to build their own ad-hoc solution from scratch — if they even try. Most just give up.

The Solution: Open Data, Open Code

bigNOSTRals is built on three layers:

1. Big Ears — a lightweight client that listens to relay events and writes them to BigQuery.
2. dbt Models — an open-source data product with visible metrics and transformations.
3. Public Datasets — BigQuery datasets anyone can query, fork, or build on.

All open source. All permissionless. Just like the protocol it serves.

Why This Matters

Analytics is not surveillance when it is transparent, opt-out, and built by the community. By making the data pipeline itself open source, we let anyone audit how metrics are defined — and correct them when they are wrong.

That is the Nostr way. Protocol over platform. Code over hype.