What about building a Censorship-resistant global proving network using Nostr ?

I built a small Proof of Concept using Nostr for the coordination layer and STWO for the generation of validity proofs.

Typical flow:

1. User submits a proving request to the network using a Nostr client.

2. A prover agent picks the request and generate a STARK proof using STWO prover.

3. The proof is published to the Nostr network.

4. The user can verify the proof using the STWO verifier agent.

The POC for now is only able to generate a Circle STARK proof for a Fibonacci computation, using STWO and a custom AIR.

Open questions:

- Fee mechanism (could leverage NIP-57 with Lightning Zaps).

- Spam protection (could be cool to use NIP-13 Proof of Work and try to reuse the grinding part of the STARK protocol,

Weikeng Chen

has started to draft idea on how to potentially do this).

- Standardisation and serialisation of proofs.

- Generalisation to arbitrary programs using Cairo.

- Batching & aggregation of proofs.

I am big fan of Nostr and it's potential. You can imagine a lof of different applications and use cases that can leverage the simplicity and censorship resistance of Nostr. I am pretty sure that ZK will play a big role in this, both for scaling of computation and also for privacy.

https://github.com/AbdelStark/askeladd

Reply to this note

Please Login to reply.

Discussion

This feels like a DVM.

Question though, can we do STARKs without any dependency on a blockchain?

What is a DVM?

Yes, 100%, you can prove off chain compute.

I do think is Data Vending Machine.

thank you

Yes indeed. I discovered NIP-90 thanks to some folks mentioning DVM. You are absolutely right.

Yes STARKs can be used completely offchain.

DVM themselves could run verifiable computation. We just need to add to the basic spec 🚀

Yeah that’s a wonderful idea! Let’s work on it together?

For sure! Let's get some proof of computation out :)

💪

Wanna put a simple fib as a DVM so that clients can request the computation, get the result and somehow verify it?

Interesting! Do you have any specific real-life use cases in mind for this system?

No specific use case in mind for now but I feel that the demand for decentralized proving will constantly increase in the future.

AI might be a big vertical needing this.