But what’s your opinion on it?
It seems users will always choose the largest relays because they want reach and content.
Nostr appears to be centralizing because the incentives are pushing it there.
Am i missing something that will push it to decentralize?
Oh, one more
3) isn’t it dangerous for free speech to force all the content through just a few relays?
There seems to be two issues:
1) Does the architecture cause it to centralize over time by creating an incentive to connect to the largest relays?
2) Who will run these larger relays? Will they continue to be subsidized by wealthy bitcoiners?
Good post. 🙏
Your analysis though is missing a dive into the centralization gravity of the tech architecture itself.
I understand your point that there will always be a psychopath willing to subsidize freedom. Maybe that’s true and continues. It’s been true so far.
But psychopaths are not enough to keep nostr decentralized.
Relays are NOT equivalent. A relay with 100k users is not the same as a relay with two users.
Relays equal reach. Everyone wants the greatest audience for their posts and to never miss a post from someone they follow. This creates a natural incentive to connect to the largest relays.
The problem with the architecture is that nostr clients limit the number of relays you can have. For example, nostr:npub18m76awca3y37hkvuneavuw6pjj4525fw90necxmadrvjg0sdy6qsngq955 recommends ten or less.
This means that when you’re limited to the number of relays, you want to connect to the largest (greatest cache of content and most connected users). Nobody wants to connect to the smallest.
This creates a centralizing gravity where relays get larger and more concentrated. Eventually we could end up with 10 massive relays. There would be no need for the 11th so few people would connect to it.
As you know from your analysis on shitcoin blockchains, they have a natural centralizing gravity. They need to be faster and cheaper than their competitors. This forces them to go POS and run fewer, larger nodes to increase throughput.
Nostr’s architecture has this same centralizing gravity.
This needs to be discussed, verified, and if confirmed… then fixed.
Ultimately, we’re not on team nostr v1. We’re all on team freedom and privacy.
nostr:npub1excellx58e497gan6fcsdnseujkjm7ym5yp3m4rp0ud4j8ss39js2pn72a nostr:npub1qny3tkh0acurzla8x3zy4nhrjz5zd8l9sy9jys09umwng00manysew95gx
nostr:npub1xtscya34g58tk0z605fvr788k263gsu6cy9x0mhnm87echrgufzsevkk5s nostr:npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z
It’s not good. We want decentralization so that nostr is censorship resistant. It’s better to have many smaller nodes.
The issue is that the clients don’t work well with a large number of small nodes. For example, Damus recommends 10 or fewer.
If you can only connect to a limited number of nodes, then you will choose the largest ones so that you can maximize your reach.
If everyone chooses the largest nodes, then we have increasing centralization… which isn’t good.
Lyn, just curious, do you run a relay? If so, what would be the monthly cost that you would be personally willing to bear for hosting fees?
Why do you say that the relays are well incentivized? Aren’t all the current major relays subsidized? It seems that nostr works because Jack and others are essentially bearing the cost. There are artificial subsidies here.
We also need to acknowledge that the client architecture is pushing the community to have fewer very large nodes. Not a good trend.
The first step to fix a problem seems to be to acknowledge it. Then people can brainstorm solutions. Not sure motivational “nothing to see here” posts are helpful, even if well intentioned.
Such an incredible video. He lied to Congress — which is a crime. Ask yourself… will there be any accountability?
Is the rule of law dead? I think it is. And, if it *is* dead, what does that mean for the future of the country?
This obviously sounds like crazy hyperbole, but it’s possible that we could be watching the last “real” election in the 🇺🇸. After this point, half the country may believe that there is no point to voting anymore.
IMHO, to step back from the brink, there should be a truth and reconciliation process (with accountability) immediately.
Not gonna happen. 🙅♂️ nostr:note155ft6rtryta7t6ymvu6wh0r95042wlm3s3dfx77trphlgrp8sq2sznfsyt
Study should have also said that people who don’t own bitcoin are more anxious and less hopeful for the future.
Zuck admits official govt pressure. This should allow court cases to proceed. Courts previously ruled that the complainants couldn’t prove govt pressure.
Remember, the govt lied, saying the hunter laptop was fake, when they knew it wasn’t. Proof that govt agencies actively interfered with the election. Conspiracy theorists right again. 😎 nostr:note15x8gefuf9hcf94yq30wwd9a6w0va0g6kzjcwrurh5jnp500nsuesdj4tje
The world continues to decentralize. Money, communications, sanctions power…
https://mishtalk.com/economics/china-gains-secret-access-to-nvdia-microchips-by-renting-computers/
Still believe that we’re following the thesis from the Sovereign Individual book and this eventually results in the breakup of the nations as we know them today.
Yes, I’m hoping that ecash will make these types of systems possible. In addition to paid hashtags, ecash could potentially help get rid of free riders.
For example, if you run a desktop, you accumulate sats as others download from you. When you’re using your phone, you’re paying sats to others. Videos especially are pricey. We forget that many nostr relays are paying for this and essentially making a donation to the network.
Another way to reduce spam is to rely on the trust network. For example, I follow (and trust) you. You follow others, and so on. I essentially trust you and everyone, directly or indirectly, trusted by you.
Just like the six degrees of Kevin Bacon, this might be almost everyone. I would be willing to get content and hashtags from this expanded network.
Spammers would have to be followed by someone for their content to be visible to the larger trust network. Since most people don’t follow spammers, this should keep much of it out. (And of course, you should be able to block individual accounts too in case you don’t like anyone).
The downside to this is, of course, bootstrapping for new users. Newbies who don’t know anyone are locked out until they get someone to follow them.
Many people are saying “put it in the client”. Of course, this would be ideal.
A few challenges:
1) Right now, each client can only connect to a limited number of relays. For example, Damus recommends ~10. This forces nostr clients to connect to a small number of large (high bandwidth) relays to reach all other users. Basically, if you can only connect to 10, then you want to connect to the same 10 as everyone else so that you can reach everyone.
2) If relays were installed in phones, they would quickly go online and offline. Each client would then be forced to connect to a massive number of low bandwidth relays. This is the opposite of what happens today.
3) Mobile devices have limited bandwidth and aren’t designed to be low latency servers. Desktops can handle more load, they generally have higher bandwidth limits, and can be left running for longer. People need to be motivated to run desktops to support the network.
To move the relays into the clients (which should be done to make the network decentralized) the core underlying protocol would need to be completely redesigned.
I’ve been brainstorming and experimenting with such a thing but it’s not ready for presentation. I think I’ve loosely “solved” the broadcasting piece (hint: bees and flowers). But I still have two massive missing pieces.
First - replies (replying to someone else’s message so that 3rd parties can see it)
Second - data efficient discovery (eg. how to allow hashtags in a way that doesn’t drown your device in spam).
Both of the missing pieces are essentially data discoverability problems so perhaps they will eventually have a common solution 🤞
The challenge is minimizing latency, bandwidth, and data replication. Also, no matter what, moving the relays into the clients will likely degrade the UX. Centralized systems are faster, more responsive, and therefore generally have better UX. nostr:note1l7yww5ltcd5an05vcevht5ltqx4z5rnrus29j28zddwgeskns7dsrrqtw0
Very strange. Did the French give him citizenship for defying Russia?
They might not know who their customers are. But they can take the service offline or they can spy on the users if they want to or are compelled. You won’t know because you can’t verify. This is why I’m saying that the service isn’t decentralized.
Better than nothing, sure. But why do you think the hosting company will not take the relay offline if demanded?
Why do you think they will not deliver the IP addresses of the poster’s npubs, if demanded?
Centralized hosting providers are also one throat to choke.
This is a fantastic use of his skills. Hope he spends more time on nostr and less on GSR.
Relays can track IPs. This does not mean they are not decentralized. Use a VPN and/or Orbot/Tor. I agree with nostr:npub1wmr34t36fy03m8hvgl96zl3znndyzyaqhwmwdtshwmtkg03fetaqhjg240, we need to advertise the onion and dns addresses of relays and have and normalize relay setups that use onion and Tor by default.
Distributed isn’t decentralized. The only reason this is debated is because nostr hasn’t faced any real attacks yet.
You seem to be saying that the solution for average folks is to understand onion routing and use VPNs. That seems unlikely and impractical. For average folks, if it’s not safe by default then it’s not safe at all.
If it’s not “private by default” then can it ever be safe for regular users?
Or does a newbie introduction to nostr now require a lesson in VPNs and onion routing?
Were you confused when you saw Russia criticize and quickly attempt consular support for Durov given their history?
Me too.
The rumor I’m hearing is that many in the Russian military (potentially Iran too) were using Telegram. I find this hard to believe given Telegram’s insecurity, but who knows. Apparently Western countries want access to these military messages. 🤷♂️