I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.

Replies (84)

Are you sure that you've only seen 5 spam posts in the lfays3 days? Amethyst is telling me that you are using public relays like damus.io, nos.lol, and nostr.mom. I'm often seeing more than 5 spam replies in these relays alone. Nostr "celebrity" posts are getting particularly hard spammed right now.
I'm still suffering from the reply guys.... I also don't really understand how to manage my Nostr account... The fact that I made onto this network borders on a miracle...
Vitor Pamplona's avatar Vitor Pamplona
I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.
View quoted note →
Sim, eu removi os relays infectados e deu certo.
Vitor Pamplona's avatar Vitor Pamplona
I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.
View quoted note →
I am still seeing the reply guy… On every message. I’m tempted to by an android just for amethyst …
Got it. So this means that you are writing to but not reading from public relays? Not reading from any of the big public relays certainly works for me, but it also means blocking a lot of legit notes right? As in, I assume that most folks are only posting to big public relays.
Ah, this is brilliant. I didn't know about the "re-broadcast on reply" aspect of Nostr. One last question if you don't mind. Is there a way to stop downloading notes from a specific relay but still keep it in the general list? I want to write to public relays and still read from a few of them without adding a lot of relays as inbox. The ⬇️ button appears to be all of nothing at the moment (i.e., if I disable it I won't get posts from any of my General relays)
After pruning back my relay list, Damus is the only one left letting through the spam.
This made it much, much better. Thank you! Hopefully I'm not giving up on a lot of legit content, e.g., some welcome posts have disappeared from the timeline.
I've just disabled downloading from a few public relays, and the spam is basically gone. If you're using Amethyst and struggling with spam, the ⬇️ (down arrow button) below each relay is clickable. image Thanks, @Vitor Pamplona, for helping me out. #Amethyst
Vitor Pamplona's avatar Vitor Pamplona
I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.
View quoted note →
Got it! Thank you. So when detecting a relay sending me the spam, should I delete it also from general relays? I am getting spam from relays that are neither in my inbox nor outbox
Default avatar
Angela 1 year ago
Is the spam a fediverse thing again? I swear from inception the concept of that sounded like the laziest puppet master nonsense ever.
A request: If I am reporting a note for Spam/ Harmful content, I don't want a reaction attached to it ...which I think Amethyst adds ⚠️ to the note I don't want to be seen reacting to such post. Just want the report 1984 event only to be sent to relays . Please help πŸ™
Because 1. It counts as engaging with the post which is not the intent ( I almost never give a negative reaction. If I don't like something I just ignore it ) 2. There are clients that show the List of Likes/Reactions and these posts pop-up if Mute Lists are not used & pollute that page. And like you yourself mentioned, what counts as a positive & negative reaction itself is a flexible contextual thing...eg.I use 😱 emoji, mostly as a faux-shock & never negatively. So I am not sure it's useful to add reaction of any sort, especially if I can't undo it. Would appreciate if you could reconsider.
Default avatar
Angela 1 year ago
Good that it’s probably not. πŸ•ŠοΈ
It doesn't count as encouraging. Reactions can be positive or negative. There are plenty of reactions ( πŸ‘ŽπŸ˜‘πŸ€¬πŸ˜°πŸ˜“πŸ‘‡β¬‡οΈ) being used daily to demonstrate discontent with posts and people definitely don't want to see them in their profiles. Whoever is encouraging everything you react to needs to fix their client to filter the negative stuff out.
But that's the point no? In the maybe 1-2 contexts I used 🀬 emoji, it was sarcastic and not a negative reaction at all. People use emojis differently and it varies with individuals... It's far more difficult for clients to judge and selectively filter, unless they customize to users the emojis they want to filter our (This doesn't even count custom emojis -- how to judge that?) And Amethyst can atleast be transparent that an ⚠️ reaction is also being sent... Then people can take a call if they want that. I had stopped reporting stuff being I don't want to react whatsover to csam or illegal stuff notes & I don't want to take that risk of being seen as engaging with those.
When you report, or you mute those notes or authors, you are engaging with them: you are creating Report Events or adding their keys or IDs to mute lists or block lists. It's the same thing as a reaction. Same thing happens on clients that mark event's as read, etc. Those are all interactions. There is no way to only interact with positive content on Nostr. It's the wrong way to look at it. You are always interacting. That interaction can be positive or negative. But they will always be there... As they should. Clients don't need to parse reactions into positive or negative if they don't want to. The only thing they absolutely CANNOT assume is that all reactions should be interpreted as the author supporting them. We will make it more transparent that it's being sent. Right now only people using Amber can see the reaction event being signed together with the report and the mute list change.
It's absolutely not the same as a reaction. Reactions are contextual and can be used differently by people. I am not saying it's endorsing the content but it's definitely engaging. A report is clearly a report / Mute/Block- there's no room for misinterpretation! and moreover, I can remove a Block/Mute if I so wish. How can I remove a reaction that I was on my behalf that I was not aware of?
And thank you for considering to make it more transparent πŸ‘That helps !
They're not not binary so yes, the client cannot make assumptions nor to read them nor imply arbitrary meaning automatically and auto send their arbitrary meaning saying that's what the user wanted when it clearly wasn't. Reports should be performed privately between the user and the client and publicly only by the client, so your users are not associated with unwanted content directly.
Reports are also contextual. There are plenty of people sending Nudity reports for things other people don't think it is Nudity at all. Reports are absolutely the same thing as a bad reaction. They just use a different event numbers. But if you use them without interpreting what the user was trying to say, you are going to have a bad time in both cases. You can just send a delete for a reaction. I have not tested it recently, but if you see the post again (might need to unblock the user) you should see your warning label there. You can click on it to delete it just like you do with any other reaction on Amethyst.
Reports ARE associating the content with the user of the app. There is no privacy on it. Users are creating an event and putting the reported id on it. Reports ARE NEVER PRIVATE. The name "report" should already tell you that. Mutes are private on Amethyst, but other clients make those public as well. They are all associating the user with the bad content. There is no way to do any of this without that association.
Intentionality cannot be assumed as well. Lots of people have been replying, reposting and even zapping spammers and impersonators lately. This notion of "users should avoid associating themselves with bad content" doen't work because it is inevitable. It's better if we normalize that association and then filter it down by what it actually meant at the time. On top of that, Clients do a lot in the background. We are generally signing tons of events left and right to just operate. Yeah, we can ask permissions every single time, but that will get annoying really quickly. We already see that with Amber that people check the first times and then just approve everything the app needs to do to operate.
I found primal and the relay with the little rocket were the worst. If you look at the contact information and click on the administrators of the relay it seems they are not very active. They must of both gone painting and ended their careers as sysadmins. Primal doesn't delete drafts because I added them and my drafts keep coming back. It's not a good relay
The main point here is privacy - what he is saying is correct in that some of us do not want to build public data graph. Say I see "Nudity" content - I may want to report it to ensure that those that don't want "Nudity" content can interpret that and hide post. But I definitely don't want to have it public... people will be like - look at Rockstar spending his day browsing nudes on Nostr (would be a cool meme though). The solution here is to delegate reports... like I wouldn't mind sending my Nudity reports to people I trust (like you Vitor / Amethyst), but I don't want to be involved further in acting on those reports. These should aggregated and then used depending on people's preferences.
Agree, any action over reports or reactions must account for those cultural differences. They are like relay hints. Relays and other clients can use them in any way they see fit as long as they know this variance is there.
Agree about Primal. So many ReplyGuys come through that relay. I recently deleted relay.primal.net and that BS ReplyGuy spam immediately dropped.
How does the bot associate the WoT of the sender to the report it files? A report from you should be more valid to your followers than a report from a spammer that has not follower. If that information is not in the Report, I am not sure what's the point of the system because anyone could just report anything they don't like and just hide behind the bot. Meaning, people get to report and take posts down without being accountable to the public.
I'm getting tired of women going around half naked. The only ones I like to look at are wearing an ankle length dress and waist length hair.
Haha πŸ˜… I love Amethyst πŸ’œ And Vitor does make changes based on user feedback -I remember he even did a translation UX overhaul after some users complained. So hoping for the best 🀞
Yes but what is the solution for relays... it is making the assumption that any relay is currently set up to prevent this spam. It is the same problem whether it's on a client or relay, and shifting responsibility to the relay gains you IP rate limiting at best, but this guy is throttling his own script to not get rate limited. Whatever the solution is is the same for clients and relays alike.
Well, in one case, people will report that as Nudity and relays of that culture will accept the report and delete the event. Same for followers in the same culture. While in the west, the post will stay. So, processing reports and reactions is always culture dependent. As long as everybody acknologdges that, it should be fine. Reporting is free speech. Reporting a reporter for misreporting something is also free speech. And people can make decisions based on those records.
It doesn't, that's a trade-off. I've considered allowing users to pick public/private reports, because public reports of spam would be far more useful/less dangerous than public reports of e.g. csam or harassment.
Clients have an easier time because we only have one user. We can delete anything that specific user doesn't like. So, we don't need much to make this work. Relays on the other hand, have the full contact list of everyone and can build much smarter filters based on WoT. I don't think there is any way around it. It's WoT or cease to exist because no one will use you. And when I mean WoT, I mean pagerank-like graph analysis of pubkeys, not the shallow follows of follows stuff.
He posted it on here:
Vitor Pamplona's avatar Vitor Pamplona
I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.
View quoted note →
I think this is the main point... you've already implemented private / public reporting in a way that you see the most relevant. And the way I understand your position is: reporting has to be 100% public, otherwise you don't consider it a valid report. So if someone wants to report "nudity" you would insist they do it publicly... there is no support for say reporting it privately or through intermediary. In any case - this has been a great conversation and thanks to @hodlbod for his input as well. I'll keep looking into how to properly categorize notes and then also give that technology to others, think it's very important for scaling Nostr in the future.
It's more than that - in some cultures people will openly bring up problems while in others it is considered rude / egoistical. I think discussing it from perspective of privacy consciousness rather than "how much female skin is showing" would make the discussion easier to follow and understand.
Private by default with option to make it public seems is a great idea. I think it's all about providing visibility setting. In between - where you report it to trusted delegated party is what would be default for me.
Make sense. I just don't see a private framework for reports. Because if you can take a post down while hiding behind privacy, then everybody will just take everything down. To me there are two options: Users can either participate in the reporting system or not. If they do, they MUST be accountable to their reports. If they are reporting out of a whim, their society should be aware of what they are doing and react accordingly.
The participation of users is impacted by how public the system is. And there is obviously a range between forcing 100% public and completely private. By finding balance you would improve the system in absolute terms. I understood at certain point of conversation that for you it is design decision... so we'll likely finish this conversation with "agree to disagree". But I see we're all better off and with more info on how to go forward... and I would like to contribute some code to this in near future. Also really glad @hodlbod joined, learned some cool new things.
Adam Snyder's avatar
Adam Snyder 1 year ago
Added the bot name to the hidden words list and my feed appears normal again. My question is this though, does all that spam clog the relays? I'm not actually "blocking" it, only hiding it. I'm OK with that, I'm just wondering if there's a computational burden on the relays, clients etc when literally every note is duplicated multiple times in its comments, and now comments are being duplicated as well.
Vitor Pamplona's avatar Vitor Pamplona
I have only seen 5 spam posts in 3 days. If you are still seeing spam: 1. You are not paying attention to your relay list. You should remove/disable relays that are sending them. Amethyst shows relays the post came from below the author picture in Complete UI mode on settings. Move to safer relays. 2. Use hidden words in the Security Filters to hide common names/phrases out. 3. Always report impersonators and csam content. This will alert your followers of potential bot traps and let your relay operators know you didn't liked their posts.
View quoted note →
↑