when did nostr start using the world’s easiest to manipulate algorithm to mass label everyone based on if influencers follow them or mass-report them due to a grudge
Login to reply
Replies (102)
🤔
Hmmmmm.
Hmm indeed. What you saying semi
Did it? I don't think I have seen that yet. Also "nostr" can't do such a thing. Only somebody using nostr, so what? Good for them.
WoT is broken has been always will be
we have spambots like @tzongocu that have high WoT scores, and AI slop makes people easy to manipulate
report-based systems can and are being exploited because people can’t agree to disagree
So we give up? Cause you tell us to?
You talking about the new amethyst thing?
no, I am providing my opinion here.
but it has been repeatedly shown time and time again that WoT does not work, and if it was such a good system, it would have already seen major research done into it for the decades that the concept has existed
not specifically. WoT is flawed and I have been calling it out for a long while.
That npub is 0 for me 🤓 Blocked.
the problem is anyone can turn up such a AIslop account, make it like posts/follow en masse, get followers and then reuse it for spam. rinse and repeat
WoT has been something humans have used since we decided to start hunting in tribes.
Just updated Amethyst from zapstore. What feature are you talking about?
GrapeRank is t just based off of likes and follows though. And it's configurable for each user. It's spam resistant because of its configurable nature.
I don't get it either
an algorithm which uses inherently flawed primitives is flawed no matter how much :
- mutes/reports: inherently personal, and can be used by attackers
- follows: content interest != reputabulity
- likes/engagement: even worse than follows, someone could spam and you only see 1 good post you like
the current algorithms are too shallow and assume a relatively static environment where everything has already been bootstrapped
What if I don't trust anyone?
If a nostr influencer follows in the woods, does anyone hear it?
loneliness and poverty usually
WoT score on Coracle -3 😑, that's the simplest metric
I never saw WoT as "easiest to manipulate," millions of bots would seem to have no effect on it from an internal perspective. Not perfect of course, but what's the better option in your opinion?
(I'm assuming WoT here btw or did i miss something?)
yes you are, the problem is humans are easy to manipulate with automated accounts, where the reputation can be built up with AI slop and burnt up for scams and impersonation
The topic is *algos* for WoT.
you have a miserable life
WoT is inherently flawed for social media, and the context of application matters.
people can be uniquely recognized and you can’t really create 2nd identities, but on nostr “trust” does not exist and has a lower bar, fake content can be automated, and identities are free
“spam filtering algorithms” are a superset of WoT
Do you need help, guys? 🖖
Flawed concept or execution? Given it's configurable and optional, and to me the idea seems sound, I don't see the problem many seem to be expressing today.
Is the issue that some large client rolls out a default setting that many users are too lazy to configure or disable?
concept
Oh, interesting. Think I've seen you mention the gaming aspect is too easy, but that seems to be more a flaw with easily-fooled users. I'd expect one could configure to only score based on a selection of the people they follow, who one believes do a good job of filtering their follows. That's basically what I attempt to do today when I stumble across a new npub; look at shared follows, scan for the ones I know apply a rather strict criteria. Works aight for me, mixed in with a little timeline review. I do really hate the scammy types of accounts out there, but that's only gonna get worse without some type of WoT I'd imagine.
Guessing opinionated relays is your fix? Less reliance on follow behavior and more on whatever algo you use or intend to deploy on Nostrland.
So you haven't read GrapeRank. You can just say that.
Tbh, I'm on the fence when it comes to algorithms and WoT.
Bout even pros & cons.
Fair. We'll break it down on Monday.
You can still use a graph offline.
https://iris.to/note1tlzzsszhk5xaghfdeuvlewajrz0hvxnsgdtln7uv7jff6zhtpdxqt4ndzx
Why did @npub142gy...xrj0 give @primal influencers the power to censor what amethyst users can see? Lol
View quoted note →
Millions of bots lol
You only need 5 bots to to manipulate this so called web of trust.
It would be extremely easy to exploit if other clients adopt it.
One person could easily shadowban anyone they want from new users with a handful of reply bots
Like grunk doesn't like your cave paintings therefore all of grunks friends being banned from seeing your cave?
Except, grunk had to use physical violence to enforce that, while on nostr some Devs are intentionally trying to create a censorship apparatus
Amethyst can bring in censorship.
But anyone who pushes other clients to follow them, they've lost my respect and my trust.
Spam filtering without shadowbanninng non spammers seems like a much better approach to me.
It makes it easier for me to trust your sense of judgement if you question everyone elses.
It's about client side censorship happening on the amethyst app.
Which mobile apps are not doing this? I hear #amethyst is, this true? If is it is shit
I feel you bro
Betraying that trust came at a high price. With WoT algos, they can just get a new npub and start over.
That is the lower end. The higher end just ends up reinforcing the hyperconcentrated feeds everyone already has, that have them constantly moaning that Nostr is dying because they see such a small subset of notes. So that a few people taking a break causes their feed to collapse.
Less of the algorithms please.
More of the PoW to tackle spam and bots.
I follow who I follow.
View quoted note →
There is a reason any form of PageRank and similar algorithms have only been used as a factor, not the only way
WoT has a serious cold start problem while at the same time relying on a metric that can be manipulated (humans deciding to follow anyone)
The "filter spam" description is not clear. @Vitor Pamplona?
if you post the same note 5 times you will get automatically blocked for amethyst users who have that option on (a majority)
sounds reasonable
Epstein didn't kill himself.
There's a million reasons to repeat exact messages 5 times even if you exclude "gm" and "gn".
If it blocks only the spammy message that's fine. As a heuristic for spam accounts it looks trivial to get around, leaving only a trap for honest users.
Feel free to disable the filter. But it still works quite well. I just checked it again a couple months ago.
We exclude less than 50 char posts.
And no, the only reason to post the same thing multiple times (including tags) is to spam everyone.
I like that control but it should be opt in and well explained. Maybe a nudge to activate it: we think this is spam. Disable spam based on this heuristic?
So is the client putting accounts on its spammer list based on that rule or are the spam posts what's filtered? If a spammer spawns a new author for every message, does he get through the filter? Did all the "gm" folks get banned?
without a time limit it changes from reasonable to stupid
Yeah lots of things pass the filter. The feature is just for the most dumb spammers, which are particularly common in public chats, particularly when involving crypto scams.
Anything can be used as a censorship apparatus.
You post gifs.
I hate gifs.
I had gifs to my apps filters, hiding them all.
Users decide they don't like that feature.
Users use another Nostr client.
Nostr working as intended.
Are you referring to wot? Or to muting and using friends’ mutes as a mute?
Just getting a new npub is easy, rebuilding the reputation is hard part.
But they end up on my hand curated list of blocked users? So now, a false positive would be indistinguishable from hand curated?
Also on public chats, kids can get excited and spam to the amusement of their friends. Do they get permanently blocked for all Amethyst users or only their friends that would care to unblock them?
No, it's a separate list. You can see it on your security filters. People have told me a lot of hipotheticals about the filter. Human behavior doesn't do that, even with excited kids.
But now with WoT, we will start having scores per event, instead of just per key, and that would hopefully make this filter obsolete once for all.
Not with automation. We keep seeing them do this.
Show me one that's new and has a high GrapeRank score in my WoT.
I don't see them, anymore. I only use good relays.
I don't have a high score in your GrapeRank, for what it is worth.
last week?
I guess I do too because I don't see them either.
And this makes me curious as to what you don't use or do to have a lower score. Is it zaps? I am seeing people I don't follow in the 90's and upon inspection it looks like some of them at least zap a lot.
I'm one of Nostr's most-prolofic and generous zappers, so nope.
I get blocked and reported a lot, and have effectively stopped gaining followers for a year, or so.
I don't fit in. I'm the spam.
That's wild to me. Why do you get blocked and reported? 😡 Fucking people.
where can I see his graperank?
Jealousy, probably.
Just use better relays that aren't run by assholes.
Haters gonna hate, F em
You are not the only one.
Yeah, there's a couple of us. And there are people who attack me personally, and when I defend myself, they mute me. And then their alts also all mute me. 😂
There was a whole reporting-and-muting swarm that hit me, about a year ago, after I complained about scat-eating porn, snuff porn, rape porn, and personalized porn (using my pics) getting dropped in my replies and sent to my DMs. Completely filled my notifications, for days, which is when I switched to my own theforest relay. Turns out, nobody will pay to send me that stuff. LOL
It was automated, but people I was following were following some of them, and they all followed each other, so I suddenly dropped off the feeds. I think WoT would actually help prevent that, as it detects sudden clusters of new npubs appearing, but I don't think it necessarily is accurate when rating one particular npub.
Yeah, they like to report my selfies as porn, for the lulz. I got blocked and reported as a crypto-scammer, as well, due to a Damus bug, where someone mentioning me in a DM caused someone else to get a DM that seemed to come from me. Took a while for them to fix it, but my DMs were full of people asking what I contacted them about. Took me forever, to figure out what was going on.
A lot of people on here like their women bland and docile, on principle. I do not deliver the requested experience. 😂
I think that's the point, Derek. I think people are exactly concerned about tribalism.
WoT is technically cool. Every nerd (including me) loves a graph database. But it has a lot of sharp corners socially.
I love the reputation scoring for relays, but I can manage my personal social interactions without scores and badges.
It's good for things like markets (seller score) but little badges and social status points in the public square are maybe not ideal.
WoT (or reputation score, really) is useful in a lot of special domains, but people ratings and transitive mutes are just kind of gross.
Yeah, I'm over 2 years, now. It's been sorta crazy. 😂
@Laser and some of his little friends muted me, recently, after I said it's unchristian to demand the death penalty for someone being homosexual.
People think I'm getting muted because I did something terrible. No, they do something terrible and I say, Hey, that's terrible! and then they mute me. 🤣
The same thing happened to me since the "great" programmer The Daniel did that crap with the block/unfollow list, and put me there, even though I never did anything to him. Probably some post offended him and... And there was a period about two years ago when everyone could share lists of blocked people.
You know what I thought about wot yesterday...
The herd effect works perfectly here too.
Here I had the claim to find smarter people, who think for themselves, but I was wrong. There are only a few and they suffer what you suffered, or other things.
There's no point in getting angry anymore. I'm speechless at this moment.
Yeah, I'm not very biddable myself
how do you answer idiots? iLL gO 1st "i dO/\/'t" luvYA Stella/ #loveIZlogical
My dear I came to conclusion long ago that muted people on Nostr are those you should listen too... We are very low on Bullshit tolerance and we dont suck those nymphs stories... they dont like that we have a strong opinionated 'voice' they cant tame. Everything always comes to the light of a day at least in this existence. They should all buckle up... 💀✨️🤣🥳⭐️
What is wild to you? You really don't know what is happening here?
Just asking...
The most friction I've experienced on Nostr is after exposing BS. They'll repost my notes and add lies in an attempt to smear me, then claim to mute me. Or, in one case, vauguely threaten to stalk me. I've never check who has muted me, because I really don't care.
Point out lies, or false dichotomies, if you want to lose favor on Nostr. Cheer when the fiat price of bitcoin increases to get zaps. That's our community algo.
Are you sure? Blocked Users is long. Spammers is empty.
Yep, then you don't have anyone filtered. The other list is all manual that you added.
So it's not a bug? I'm too fast blocking for your algo to get triggered? Ok. Hope you're right.
Nobody likes their hypocrisy being pointed out. It takes real balls to admit when you were wrong, hadn't thought things through or jumped to conclusions through prejudice.
I didn't realize being muted by someone on nostr has such an effect. I figure it would just be something thaglt one user's client does.
I mute a ton of people.
I mostly mute / block pedo spam shit. I otherwise enjoy reading other peoples takes.
I didn't know you could see who has muted you.
Virtually everything you do on Nostr is public.
Want to have scary fun? Login to a nostr client using someone else's npub. Yes, their npub.
You often can, yes. That's how the data gets included in the algo.
i think people are afraid to use mute, when in reality, if we're supposed to be in charge of our attention and our feeds, then muting is just an extension of that. muting allows both parties to win and be in control over their experiences.
My WoT doesn't respond to bots at all. Because i don't follow bots. What measurement or action is responding to this bot activity and why in this scenario?
What do you mean "yes you are"? I haven't made a personal statement above. But again I was asking for the mechanism by which WoT is easy to manipulate. Are you saying that bots pretending to be fun or useful in a certain context to get people to follow them and then "turning on them" is the mechanism to propagate fake mutes/bans and that these naturally get spread through the follower network too?
I'm asking because i wasn't aware that some of these actions would be automatically adopted if that's what is being discussed. In other words, I was aware that people couldn't get "into" my bubble without some connection to the WoT rather than the notion that if a few accounts *inside* of my WoT would make an unintentional ban/mute for someone that I may want to follow. I actually wouldn't like that behavior generally. People mute for simple opinion, a response, or just being disagreeable. That s a negative thing to "crowdsource" so cheaply, imo.
\/\/OT succor _ 
idk 🤔;.;😉 @Ryan
idk 🤔;.;😉 @Ryan
“yes you are (assuming correctly)”
Yes, people can and have made bots that pretend to be useful that they then abuse to make spambots bypass WoT.
There alsl are WoT algos that use personal mutes and reports (which can be abused; for example Calle trying to silence Cashu critics) to automatically hide content
@calle NA calle*/*ya strEEt is naturalpATH*****
Gotcha I didn't realize mutes were transferred WoT style and generally think the muting/blocking aspect should be personal, not transmitted unless explicitly agreed. In other words I think it should be "loose automatic white list, and manual blacklist" in implementation. Obviously likely scaling considerations there but i think that's a better default risk trade off in my mind at least. Could be missing something though, haven't thought about this deeply with hard specifics.
off-topic but can you point me to some of the cashu critiques?
So you are a new user or just lack reading comprehension?