It explained all the traditional approaches, which are all able to help discoverability and shareability of data between servers, and then says "the solution is relays" and then describes something that doesn't seem to be relaying anything. It sounds like a single dumb, untrusted message store on a single server that doesn't relay anything anywhere. It even specifically says "Relays don’t talk to each other, and users only need to join a small number of relays to gain autonomy—at least two, and certainly less than a dozen".

Not sure where the less than a dozen relay bit comes from. Are they expecting clients to do all the relaying between the relays? If so, wouldn't you get every relay getting pummeled by a load of clients simultaneously, all trying to push the same message. It sounds like the complete opposite of what you actually want. The article seems to just stop short at exactly the point when it should say how what they're proposing actually works.

Nostr is so simple because it handwaves away the fact that everybody seems to use the same small set of relays and there's nothing stopping them from censoring the network. I'm also not aware of any incentives for the relay operators either.

Should likely be called a "database server" since it's main purpose is to host user data and perform queries over it. A relay is something connecting two devices and makes a best effort to get out of their way.

Nevertheless: NOSTR is the most exciting social network that I've seen in the past 20 years. The concept of owning the keys without a blockchain associated enables not just decentralization, it also permits a complete offline functioning to login, view private messages and so much more that isn't possible from any other popular social network predecessor.

> N^2 scaling: if every fed has to talk to every other fed to exchange messages, the number of connections will scale exponentially

No. That's quadratic growth, which is a fairly mild form of polynomial growth, which is much much much slower than exponential growth.

   k   k^2   2^k
   1     1     1
  10   100  1024
 100   1e4  1e30

1. Content discovery

2. Spam

3. Content moderation

I can see relays offering unique solutions to each one. But now they are more than just dumb servers.

You get to the point where you might as well just write posts locally then submit them to X, Facebook, etc. You get the same result. And if you include a cryptographic signature with each post, you can prove you are the same person across the different platforms.

Boom. Same as Nostr, but with existing platforms

Maybe like... the author thought a nostr is similar to, I dunno, a pack or tribe or something?

Or if you really care about the crypto piece, then freenet.

All the "downsides" of a superpeer (as the article says - "centralisation with extra steps") but without the benefit of dynamic peering thereby resulting in incomplete routing.

i.e. by its nature Nostr results in a fragmented network, which ends up looking very much like the federated network, albeit more interconnected.

Thats not necessarily a bad thing, but its a bit of a confused article, IMHO.

This is NNTP.

"Nostr doesn't subscribe to political ideals of "free speech" — it simply recognizes that different people have different morals and preferences and each server, being privately owned, can follow their own criteria for rejecting content as they please and users are free to choose what to read and from where."

Their statement underlines the fact that nostr is a stream of dirty sewage and they want users to submit their valuable user-created content into this sewage. Then they turn around and say that the sewage is not a problem because you can filter it and even use it as drinking water later on!

I don't see how a person with real-life social rank and social capital will sign up to something like this, or be willing to maintain a technical interface to the "stream of different morals".

You'd need to put immense trust into the "filtering" process so that you are not involuntarily exposed to rubbish. And on the other hand your valuable user-generated content could be showing up in another context with your name attached, directly next to some extremely degenerate trash created by "people with different morals" as nostr calls it. Advertisers have big problems when their brands are advertised next to problematic topics, it is the same with people.

How can you rationalize this as a good value proposition? People want to impress an audience with their user-generated content. And you only want to impress someone you look up to.

If I could sign up to a social network of people who can put a nail into the wall, take a daily shower, brush their teeth, and live in a democratic country I would immediately do so. If I want to get exposed to "different morals" I just open any of the other existing social networks. Until then I'm stuck here :P

A step in the right direction for sure! But I don't feel like Nostr is the final target that nature is shooting for here.

Same thing over and over again.

- how well does such an ecosystem resist enshittification? Given some of the other comments, Nostr itself would not. However, is that true for every relay networks?

- does the Willow protocol have the same basic constraints? I know willow works with user-owned keys, but can it also organize as something similar to relays?

- local-first apps organized this way would be an interesting ecosystem

- how well would this work with keyhive? (Local first access control)

On the other hand, what're the economic incentive to run relays? If there are economies of scale, we swiftly go back to the oligarchic model.

Activists, in this case, are people with a social mission that they deem it's more important than any other considerations: they think ideology K is dangerous and they are trying to prevent as many as possible recipients to be exposed to it. They will report you on Threads or Facebook to ban you, if you speak in favor of K. They will send e-mails to your employer. They will even send bomb threats to venues where you gather to celebrate K. If they are moderators, they will not only ban you if mention K in a positive light, but they will try to avoid other people from hearing K-speech as well. If they run a Mastodon instance, for example, they will have a ban list of other instances that are K-friendly, and they will make sure that, if you are using their instance, you can't see any posts about K. If you're curious about K, now you have to do the inconvenient dance of switching between two instances that in theory should be federated, but in practice are two different networks that don't speak with each other. This is good for activists, but bad for you, if you don't want to take sides on a culture war you don't really care about.

A relay-based architecture makes the work of activists a bit less relevant: they can still run their instance and ban every mention of K, of course, but now you can subscribe to their instance AND another instance that doesn't ban people who speak fondly of K, and they can't limit or control that in any way. In theory (and everything is a bit theoretical at the moment), relays that heavily censor certain topics are less preferable to a generic public than relays that don't do that, so activist moderators will pay their effort to shape discourse with less participation from users. Of course, if relays ban something universally considered bad, such as spam, they will have more success than if they ban some heavily divisive point of view that 50% of the public shares. In theory, these controversial actors can even advertise friendly relays without you knowing, and your client can decide to follow them transparently (the intent is "I want content from this user", the behaviour is "follow relays they advertise behind the scenes"). Of course they have to do that before they're banned, but the point is that, for every activist relay that tries to remove K from public discourse, there will always be one or more generalist or counter-activist relay that welcomes K, and you can choose to follow both at the same time, with the same client and the same identity, and nobody can do a damn thing about it.

Sounds like REST. The original REST, not the botched CRUD that companies pushed for.

https://roy.gbiv.com/pubs/dissertation/fielding_dissertation...

> The combination of layered system and uniform interface constraints induces architectural properties similar to those of the uniform pipe-and-filter style.

See also Figure 5-8.

The dissertation is all about deriving that network style.

FUD. I and many others on HN run our own email servers with essentially no delivery problems.

And what they’re about to become is going to be something more like political yard signs.