The fact that people like the fake Gaza guy bot or some others I won't name are still showing up basically everywhere is proof that no lesson was learned from all these years of spam, claims that "WoT fixes this", excessive muting and that any basement kid that wanted to actually break the Nostr experience for everybody would be able to do it easily. But I still have hope the tide will turn.
fiatjaf's avatar fiatjaf
Let me say it clearly: The only way to fix reply spam is for people to use inbox relays that filter spam. Clients then should read replies by default only from people's advertised relays. Finally, if you see reply spam in someone else's thread it should be clear whose fault that is, so that person will be ashamed and fix it.
View quoted note →

Replies (16)

I'm going to create my own client where you need to pay 21 sats or so to follow/friend request a npub before you can interact, react etc with an npub because the devs are doing a terrible job with their clients. Clients are full of spam and they aren't doing anything about it.
Spam exists on internet, that's a fact. It was existing and it surely will. We can disable it by using KYC and every account needs gov ID but that's not solution we need.
There must be something I don’t understand. Maybe this is an outbox model thing which I don’t use. My client pulls all notes it finds period. So if the bot replies to something, and that note lands on Primal or any other big relay, my client will pull it. How can I filter at a relay level?
You don't need "ID," robots don't look human yet. All it takes is people meeting face-to-face to verify each other. I've said this a million times, I don't understand how nobody is hearing me.
The question is: your client pulls all notes _from where_? Also, does it download all the contents of the Primal relay or only a subset?
Hmm… these are good questions. not 100% sure how Damus does it. It just pulls from those right? And probably not ALL content. Just what it needs I’m assuming. Not sure if other clients do this better generally speaking. image
Some bot farms zap, sometimes only themselves, sometimes established accounts, too. It's too cheap to be a filter.
21_21_21's avatar
21_21_21 yesterday
Web of trust would work if more clients implemented it, but afaict very few actually do. Same with PoW. Many clients let you add PoW to a note but don't automatically filter out non-PoW notes nor do they compute a "lifetime PoW" score for other users which would be much more useful than just the PoW from a single note.
21_21_21's avatar
21_21_21 yesterday
Also I just read your IPFS criticisms after discovering you are one of the few devs on here who sees the potential of drivechains. re: IPFS limitations, look at freenet for solving the content addressable vs location addressable conundrum. Ian Clarke is doing some very compelling work there.
Sunday Midnight's avatar
Sunday Midnight 18 hours ago
Moving from a Web of Trust to a Web of Distrust. We don’t need to know who to trust. We need to know who NOT to trust, and let that signal propagate.