Replies (56)
Nice feature!
Wow this is huge!
Come to our WoT hackathon :)
Isn't this the same as having social credit scores? Seems like it could definitely lead to that at the very least...
Fantastic work Cody! I'm doing to get this deployed on my own instance ASAP.
Isnt this the same as social credit scores? Def seems like a precursor..
Isnt this the same as social credit scores? Def seems like a precursor..
Isnt this the same as social credit scores? Def seems like a precursor..
I've made that connection too, but not quite. Its relative to who you know, who they know, if they've made their profile identifiable, and have a vanity address linked. The only "social credit" factor is if the given calculation weighs against user reports.
Either way, WoT scoring exists in lieu of being a truly open protocol that people use as a platform for speech & is wholly voluntary, but highly encouraged to use to filter out the noise and spam programmatically.
No system is perfect and there will always be people who are considered a false positive.
Reputation filter: censorship by default, following people by you own individual choice is more discerning. Global noise can be beautiful, but not 100% of the time. If you never see ugly, then beauty doesn't exist.
I just built a simple pagerank service using AI
The scores are currently based on the network of relationships, your reputation increases when followed by high-reputation users. Since there are a lot of spam accounts in our network, most normal users end up in the top 10%.
I think this is quite different from a social credit system.
it's soft defense for a soft attack method. i'm sure there will be ways found to game current schemes and they will get more advanced and circumvent these attacks.
The main purpose of this filter is to weed out spam accounts. Since there are a lot of spam users on the Nostr network, normal users usually end up in the top 10%. But this isn’t fixed, so the choice is left to the user. To avoid filtering out too many legitimate users, I’ve set the default threshold at 40%.
This is bound to be a long-term battle.
If you used AI, then it's the AI that built it, not you, isn't it?
It doesn’t seem that important, but yes, most of the code was generated by AI. I just handled the review, fixed AI’s mistakes, and made some small optimizations.
it's necessary security, but i think the current designs will limit legit new people from appearing, i think money is a primary simple thing, LN+nostr... NWC. the idea of not zapping to use stuff will be normalized.
i mean, denormielized
We are building a spec that anyone can build these services in such a way that clients can just allow users to pick one. Interoperability between clients and service providers is the goal. Your perspective as a client dev would be greatly appreciated.
I initially thought about the new user issue as well. But I realized that new users are usually brought in by an existing user, once that user follows them, their reputation is higher than that of spam accounts.
that's onboarding flow issue that needs to be dealt with also then
Sounds great! What do you need me to do?
Fill the form to get the invites to the call:
Web of Trust Hackathon
Join NosFabrica in the first ever #wotathon to develop the Web of Trust ecosystem for Nostr, build open source libraries, relays and applications, ...
We usually meet every Thursday but this week is on Friday, the 2nd.
I would love to get you presenting a bit of this approach, focusing on what you need as a client and probably how simple solutions can work. The group is mostly service providers that have a hard time figuring out what clients need (especially which formats are better/easier)
I'm interested to understand how user's reputation is derived. I don't want to eat your time, so if you have explained it before or have a link that would be very much appreciated.
I’m not really into meetings, mostly because my English isn’t great haha. But if your protocol is out on GitHub or somewhere, I’d be happy to check it out and throw in a few suggestions.
You don’t need to do any configuration. You might want to double-check that your code is up to date, and try refreshing the page a few times since it could be cached.
Don't worry, half of us aren't great english speakers either. :)
We have a few NIPs and drafts of different ways to do the same thing. We are currently trying to figure out which one is better for which type of service.
Pensei a mesma coisa quando vi o recurso.
It’s basically a simple combination of the PageRank and TrustRank algorithms, both of which have plenty of resources online.
In short, if you’re followed by high-reputation users, your reputation goes up too. The more trusted users vouch for you, the more trustworthy you become.
I’ll take a look at those NIPs 👍
Thanks, I forgot to merge from master into my own branch before (that's why I deleted the original post, apologies) 💀
Got it working now on desktop, but I don't seem to get the view on mobile. I have flushed the cache on cloudflare, but will test again later on mobile, this is great, thank you!
That sounds reasonable. I suppose at the extremes it needs people of high reputation to venture deep into the noise to rescue those who sit with a Purgatory of ignorance.
How does an account get reputation? I like the approach but it typically leads to cold start problem
A new user only needs to be followed by a regular account to boost their reputation into the top 60%, since spam accounts rarely get any followers, following them would expose the spam account owner.
Exactly. If a newcomer starts completely disconnected from everyone in the network, it’s indeed hard to enter the trust graph. But if they come to Nostr through someone they know, they can gain some initial reputation from that recommender, enough to help them stand out from the spam accounts.
Do we already have a feature to allow invite someone to Nostr using Jumble, so they automatically create an account following the user who invited him?
It’s a good idea, but it wouldn’t help with the initial reputation issue. A new user would still need the inviter to follow them in order to gain any reputation.
How is user reputation calculated? I understood that as people with a higher reputation follow others with a lower reputation, their reputation increases, but how did these more reputable users get a higher reputation at the beginning?
Yes for sure, if its not the government deciding what's socially acceptable or not..
Apologies for the multiple replies, not sure what happened there..
It would be interesting if the notes of users who are still low in reputation, but who activate PoW, had more relevance in spite of everything.
It would also be nice if the filter were modular. I, for example, would like to use "off" in responses, "WoT" in feed and "40%" in notifications.
I must confess that I don't understand this feature well.
Everything within the Universe and even it's border with Eternity are governed by boundaries. These boundaries are the descriptions of complete absence of Zero upto the Unity of One.
Bitcoin is a mirror of the Universe and it's boundary structures.
Percentages are expressions of complete Absence 0% upto complete Unity of 100%
As such, it's very important to ensure that people can cross the boundary of 0%, or just above and avoid being ignored and or invisible. This is just like the first time someone joins the Bitcoin network, they cross the boundary of Zero Bitcoin into Being (Being in custody of Bitcoin). Like an animal being born, without fear and or favour and or permission. Bitcoin allows all equal access, without any qualification/caveat.
Humans are not good at feeling and knowing the extremes. The extremes at Zero and or the extremes of the exponential. That's where most get lost as the mind doesn't easily go there.
Something is either full open or it is not, this is Zero Boundary Logic and it takes some thought to see it, and more importantly recognise it, in all it's forms.
There is no easy answer, as if there was, then it's also easy to game.
I'm not criticising, I'm observing and setting out my observations. That’s the impossible that Bitcoin got right.
bullshit.
Firstly, we are talking about a consensus system and not bitcoin.
secondly, bitcoin proof of work, proof of work and goes beyond this nonsense of 0 and 100.
I understand why you call bullshit, I don't blame you.
It doesn't matter what words or even words in any language. Boundaries are devoid of our mathematics. However, spoken language does follow the boundaries of the Universe.
I explain this here and there's also a link to my paper and other papers can be found there too.
See the medium article below and my other articles. I've been studying Bitcoin since 2013 and my writing reflects my journey of discovery.
I don't have time to explain it, as first I'd have to fully understand what makes up your mind.
https://medium.com/@j0e810ggs/a-mirror-of-the-universes-boundary-structure-captured-in-language-3ba5f31202bdEveryone starts with the same score. The system then recalculates reputations iteratively, and after enough rounds the scores converge and stabilize.
Yeah, I’ve been debating whether it makes sense to split it into multiple, separate filters.
Good suggestion 👍
You don’t really need to understand the details, just know that it helps filter out spam. If you’re still seeing spam, you can raise the threshold a bit.
Ok, so the UI could be a lot simpler. Something like: Seeing spams? raise the threshold a bit. Just this.
Hello
I'M Muslim and I invite people to learn about Islam.
Please watch my blog 👇explaining with pictures and videos about Islam
https://jesusen1.blogspot.com/2017/12/what-is-islam.html
I wish you a happy life....Thank you
in 2025:
A reputation filter that hides notes from the bottom X% of users by reputation.
It works across feeds, notifications, and replies, and is especially effective at reducing spam when browsing large public relays. #devstr
