[eykanal]

This exactly. Worth mentioning that "censoring" can occur in any of a number of ways; blocking select traffic, slowing select traffic, "forgetting" specific nodes, redirecting other nodes at will, performing MITM attacks (if the protocol isn't secure), etc etc.

Also, beyond just no positive incentives, there are nontrivial negatives... they're hubs for an entire network, which can be a lot of traffic and bandwidth if peers are sharing anything other than text. That's a potentially significant cost for literally just being a dumb router. The idea of charging for this doesn't make sense... you don't choose a router, it's automatic based on location, so there's no incentive for quality. That ends up being a race to the bottom, which there's no room for arbitrage; prices are driven down to near-zero profit.

Abuse-wise, the model is fundamentally flawed. Economically, the idea kinda works so long as hub traffic is low enough to be swallowed in background noise for whoever manages the hub. Beyond that the model breaks pretty quickly.

[shark_laser]

Read up on the outbox model and zaps. Also check out Bitchat for a real world example of Nostr being effectively used without even requiring Internet connectivity.

You cannot censor Nostr.

Also, check out how zaps work, and relay authentication. You can charge for relays if you want.

[acdha]

Can you summarize how those prevent the listed problems? Tossing around absolutes like “you cannot censor Nostr” sounds like a religious assertion rather than technical analysis.

[shark_laser]

I have posted very similar replies to other messages in this thread and don't want to repeat myself too much at the risk of being considered spam.

But... Outbox model prevents censorship because you push your (cryptographically signed and so impossible to impersonate) messages to multiple relays. To your own preferred relays, as well as to the preferred relays of others who are involved in the conversation, as well as to a couple of global relays for easy discoverability.

These global relays are useful, but are interchangeable and totally replaceable. As soon as you've connected with someone you can retrieve their updates, because you know their preferred relays, and can query them directly.

In this way Nostr has the benefits of centralised networks for discoverability, federated networks for communities, and private individual web site for p2p and archival purposes. As well as making it impossible to censor.

And if you take down THE ENTIRE INTERNET in order to censor Nostr? Well, Bitchat is Nostr via Bluetooth Mesh Networks. Do a quick search and find out where and when it has been used (Nepal, Indonesia, and elsewhere)

And as for zaps fixing the economic problem, I'm not sure what else to say other than you can give and receive value directly using the Lightning Network. It is seamless in most Nostr clients, and built into the Nostr protocol. If you don't believe in Value For Value (v4v) then you can just charge a fee, and the economics problem is solved.

[shakna]

Encrypted but not always leak proof. [0] There have been viable attacks for relays to perform.

[0] https://eprint.iacr.org/2025/1459

[nunobrito]

NOSTR is a protocol that doesn't detail all implementation details so it wouldn't be fair to point HTML as culprit for flaws of web browsers.

That is a good paper, the leaks are mentioned the app Damus (notes browser) which wasn't really much worried about verifying the authenticity of the notes. The details: https://crypto-sec-n.github.io/

These are apps developed on free time and made available for free so these issues are bound to exist and be repaired.

[pjc50]

> You cannot censor Nostr.

A government could make it illegal to run or connect to nodes. It could DPI traffic in and out of the country, and block known nostr relays. Or it could just mandate that smartphone manufacturers block it, which would take out a large fraction of potential users.

(How does nostr avoid hosting known CSAM? Because that is the one thing that law enforcement will definitely come after)

[supermatt]

> You cannot censor Nostr.

Sure you can. A relay operator absolutely can censor what goes through their relay. More to the point, you cant even prove that such censorship has occurred.

Nostr is censorship resistant in that you can publish to multiple relays, but that is far from censorship-proof.

[MangoToupe]

Could this be run by, say, a public library or are there concerns about liability?

It also seems like this is sort of reinventing email.

[nunobrito]

The concept of public library are the "super-relays", which are always available and basically accept any note you send their way.

It is "kind of" like reinventing email with PGP. Main difference is that you can choose to send the message in plain text with a cryptographic signature that proves it was sent from you or full encrypted like PGP.

There is still (in my opinion) a disadvantage when compared to PGP: key rotation. Once you create a key pair in NOSTR it is your identity forever, whereas in PGP you have mechanisms to declare a key obsolete and generate a new one.

In overall PGP failed over the last 30 years, sharing public keys with other people was always the biggest difficulty for real adoption. With NOSTR this process is kind of solved but we are yet to see about adoption.

[attila-lendvai]

signing and encryption are separate operations also in PGP.

and yes, one of the hardest parts of this domain is the implementation of the web of trust (key management).