We need like a following.space for hidden words, mute lists and relay lists. In that way, anyone can share settings that set up clients to combat spam in a decentralized fashion.
Discussion
Nice
Mute words are at best a band aid I think. Spam filtering has to be at the relay level.
that's what I think too
but what relays? wouldn't that exclude new users? what if relays decide that some controversial opinion is spam and ban it?
I think very cheap paid relays are an option. The admission is like a CAPTCHA that adds considerable friction for spammers.
If we had easy to host paid relay software we could create a market to drive prices down and also lead to censorship resistance.
Or require an actual CAPTCHA before posting to a relay. Maybe there's some way of providing proof of CAPTCHA too so you only have to solve it once and you can replay it on many relays.
there's no such a way and captchas can be bypassed with AI
Are you sure? I think you can construct them in such a way that they're at least extremely hard to solve by automated means.
I am intrigued by a pay per note posted model, like what keychat.io is trying with their stamps model for messaging. Ecash makes it feasible.
That's a great idea, I've tried to mute the words I found on the spam posts but hasn't been enough
Absolutely. I recall there was a discussion a few weeks back of widening 'following starter packs' to include these extras.
Very important for onboarding.
so every move by one spammer has to be countered with a move from every nostr user? How can that scale?
In a centralized system, the central authority can take action against spammers that benefits all users. However, they can also take actions that hurt all users.
In a decentralized system, users are protected from a centralized authority acting against them, but it means they also must take on the responsibility of acting for their own benefit, since no cental authority exists who can do it for them.
nice excuse but doesn't scale and if that's your solution nostr is dead
the actual nostr solution must be to use relays as filters somehow
Which requires users to take respinsibility by choosing or running their own quality relays.
What I said is not an excuse, it's a simple reality. And maybe that means Nostr will die because people cannot handle the responsibility of freedom.
It is. The whole idea is to connect to or operate relays that align with your preferences. The things that make nostr censorship resistant also enable tight censorship on a small scale with the option to opt out. It's about keeping the exits open. That point is lost when relays are hidden and keys are obscured away, though. Trying to dumb it down rather than leaning into unique possibilities, I believe, is hurting more than helping. People are dealing with these things they need but don't understand and are told they don't need to think about them until they do, which is usually when something bad is happening. They could instead be exploring fun options and learning how to manage those bad situations in the process. There's just very few ways to interact on the protocol like that right now, because who wants to build half a bridge and hope someone else will build the other half? That would be terribly frustrating.
Here's how the Little Rascals dealt with this problem.
Vitor, for some reason and without doing anything (changing relays, blocking profiles or filtering words), some people who use the same relays as the bot's note on Amethyst are not receiving messages from them and others are.
Can you tell me what is allowing this filter?
Well, they don't reply to all messages. Relays are also fighting this quite hard. Reports are working and if the client loads those reports (and the filter is active on Amethyst), it will hide those posts.
If anyone should opt to create something like that, I hope that they include a big disclaimer at the top of the page, that results will vary by client. As to not set false expectations.
Just wanted to toss that thought out there. 😊
