Nothing about the Nostr protocol struck me as particularly interesting. Spam control, moderation and anonymity are not really dealt with. Why the hype?
>Nostr is a protocol, designed for simplicity, that aims to create a censorship-resistant global social network.
So they basically want to create twitter without Musk (or anyone in charge for that matter). Nothing wrong with that goal, it's just highly unlikely to succeed given the fundamental shortcomings of this approach.
For a number of years I was a forum moderator for a game forum called Uru Obsession. Eventually that wound down and the forum closed, meaning all those discussions have also been lost (unless someone backed them up - by the time they closed I had moved on, so I dunno), and that community as far as I know has mostly dissolved since its closure.
A gossip protocol means there is no host - clients communicate directly with each other and also store the conversation locally so someone else deciding to stop paying hosting costs becomes a non-issue.
I've seen a lot of people leave Facebook over the last couple of years. And I mean a lot. Oh, they might still log in every month or two to see what's up with friend and family, but daily use? Nope.
I think people just need to be lead to greener pastures. Right now the alpha geeks aren't cooler & better, don't have great & obvious advantages for being out on the frontier trying cool shit. The Tim O'Reilly "Follow The Alpha Geeks" advice is rarely wrong, in my view, for the alpha geeks mostly want to expand capabilities & power & enable, in ways most consumer efforts are too bounded & limited to go for, but we keep forgetting this wisdom's words anyways.
Once the alpha geeks are unqualifiedly better than the mundane normy-nets, the tables will start to turn. I think the geeks are doing the good work, are putting in the right effort.
Dogfood your way to success. Do what empassions & excites you. Don't worry about l-users. Focus on being really good & powerful. You'll be out competed if you do what sigmoid10 says & compete to be the lowest common denominator of social networking, and your product will suck as bad as everything else we have.
Truly good works market themselves. Places where genuine authentic people (and creative fun bots) mix & share themselves in are what we are searching for, is the authenticity that the engagement-loop corporate networks break & burry. There's different races here. I do think the broader we are searching for better more open pattern en mass to replace the walled garden networks (a challenge many distributeers reject), but the path to victory is assymetric competition, is tapping into different sources of value & raising it up in different ways.
Do you believe in humanity? Or do you think synthetic gloss shit forever & ever will always win?
The protocol is surprisingly simple to read [1], many relays and clients exist already.
I exchanged messages with a friend of mine who was using a very different client and it just worked!
Personally I like the fact that you can 'like' posts by sending a couple of sats via Lightning. I think it is a great motivator to write thoughtful, quality content.
Currently nostr is radical, weird and unpolished. The Amethyst client is slow at times. But the pace of development is incredible.
[1] https://github.com/nostr-protocol/nips/blob/master/01.md
Adding because Nostr seems to be marketed as something more decentralized than that, and I'd like to get a deeper understanding.
With mastodon your client only talks to one "instance" to see all the "instances" that "instance" is federated with.
With Nostr your client connects to every "relay" you want to see content from, there isn't really any communication between "relays".
assuming it's the same as Scuttlebutt (also gossip protocol) a relay is literally just a relay. There's no "home server". You don't have an account on anyone's machine. You just shove data out to "people" who are listening to you. In scuttlebutt the relays are configured so that anyone can ask it to "follow" them and then they send their data to it. Anyone who listens to that relay can get any data the relay has.
in scuttlebutt the problem was that I never felt like i could trust a relay to exist for more than 6 months, so i just followed every one i could. No-one wants to set up a relay, and relays have to have a static domain / ip so that you know where to look for them. It's not like tor where you can just leave your computer on and that's good enough. I expect the same problems here.
This makes it way easier for people to write clients or bots or whatever, but it also tosses out many of the guarantees people who used the original scuttlebot took for granted.
When you follow someone, or someone follows you, the follower's client will get a list of relays the person they want to follow is posting to. The follower can connect to any of those relays and get any new content.
On many relays it's possible to get a firehose feed of everything posted to that relay. On free public relays this firehose feed may contain lots of spam accounts.
Coming up with a protocol is not that hard. The technical side is fun to work with, but I guess technologists don't realize that the human side is way harder to do.
After I muted a small number of people and used some paid relays I don’t see spam or bad content.
Coolest feature is that you can switch clients while keeping the same relays and all your data stays with you.
Plus a bunch of clients have this auto translate feature so people are talking to each other regardless of the language they speak. I started following some people in other languages - very unique social media experience.
Congrats on your first comment by the way.
It's a pretty nice way to keep your feed quiet (not spammy), but also allow your notes to make it out to the broader public nostr sphere.
Why would you need to deal with anonymity?
Your identity on the network is a public/private key pair. You don't have to associate your real name with your posts, if you don't want to. And you can have multiple key pair identities if you want.
Posts are plain text by default, but encrypted private messages can also be exchanged.
You can choose what relays to post to, and obfuscate your ip address with a vpn or tor if you desire.
> can the relay log everything
Yes, subject to the above mitigations.
I'm ready to join a social media website that requires a social security number, photo ID, and interview process.
I've had enough of propagandists and marketers.
Nostr is all based on anonymous cryptographic identifiers, so it seems like you have some special definition of anonymity that you are looking for, as it seems nothing if not anonymous. Having a stable identifier allows relays to know who to send versus who not to send, and allows connecting data together. Users are free to sock puppet up to their hearts content, if they wish to further diffuse traffic.
The appeal? The appeal here is that this is an incredibly malleable & comprehensible low level tool for messaging. Competitors like AtProto or ActivityPub involve complex protocols to exchange/syndicate data around, as much as the payload of the messages themselves. They are high level visions for what a network is. By compare, Nostr's low level approach is organic & searching not a refined final product, but a thriving ecosystem of expanding ideas.
Nostr has extreme elegance as a protocol by being focused primarily on messages themselves, which start as very simple & understandable self signing devices. The transport & exchange of messages is almost incidental, and indeed, Nostr over shoebox or carrier pigeon is possible. This allows a lot more flexibility with how the network can form distributed connections, allows great offline capabilities, allows creative relays & creative/selective distribution mechanisms to form.
Nostr is an excellent base layer. The base specs are quite short & direct. It's a protocol one can happily implement in a weekend.
Nostr has incredibly wild applications, because it is a simple extensible base. There's a wide variety of interesting capabilities that have already need accepted as Nostr Implementation Possibilities, NIPS, that grow & build on one another. Nostr base protocol is just a start, just the seed of an idea, one that's meant to be iterated on & expanded, and it's so easy & direct to do so. This is the biggest advantage by far; I cannot stress this enough. Not trying to do absolutely everything & making a modular simple protocol to start building & iterating from is all the wins, is the Bazaar to the ambitious Cathedrals. https://github.com/nostr-protocol/nips
Nostr is by far the most malleable, most open set of possibilities, the most grow able, of the social networks we have. Everything else seems to have been designed to arrive somewhat fully formed, ready to go, but Nostr's strength is that it doesn't purport to know every use case & to have a total picture of what it is. It's a much simpler idea, with much more focus on finding out the uses.
How would the relay evaluate what is spam and what is not?
Event IDs are a SHA-256 over the event's payload and metadata, so the idea is that you put some extra metadata saying "I'm doing a bunch of extra work to generate an ID with N leading zeros in the ID", and then a nonce value. You generate the ID, and check to see if it has the number of leading zeroes you wanted. If it doesn't, you increment the nonce and try again. If it does, you're all good- you sign the event and send it along.
Because the ID must be a SHA256 of the rest of the event, you have a fairly good indication of how much work the client would have had to do to generate that nonce. The more zeroes in the ID, the more effort they would have had to expend.
So, as a relay operator, you can define a policy that you won't relay events that don't meet this proof-of-work requirement, and boom, no more spamming.
Of course, there are other ways to handle the spam problem, such as requiring authentication mechanisms or external attestation of messages. But there are multiple tools in the toolbox here.
Jack Dorsey is a co-founder, that's it. I feel like "nostr.com" has been showing up on the HN front page, with no context, at least once every week.