Here's the idea:

We have a solid existing mechanism for digital signatures: I can publish a public key, and if I sign my communications with it (such as this post, for instance), you could be sure that only the owner of that private key could have written it.

I could go around collecting public keys of other people online, and I could associate some "trust" value with each one. I could publish my list (and sign it), and others could perhaps say "I trust you, so I'll trust the people you trust a bit as well".

In this way, a web of trust grows, and we could use this information to filter the crap we encounter. If someone "backstabs" and starts spewing adverts from their previously-trusted persona, communities could adjust their trust values to punish that behavior.

Basically, it's the same thing we do in daily life, with our interpersonal relationships, just made explicit and cryptographically secure.

----

I realize there's PGP, and it's "web of trust" concept, and it's close but not the same. As I understand it, that's based on "key signing parties" where you verify that someone is who they say they are. It has a notion of physical identity. What I'm describing doesn't care about that; you could have 100 anonymous online personas, and what matters is only how they behave, not what physical person they're tied to. Also, you'd use associated trust values to rank info you see, rather than having a yes/no verification. Also also, PGP was a UX disaster, so I'd want something much easier-to-use.

I realize that it could be abused (eg: witch hunts), but I don't see it as being worse than what can happen in real life.

I'm also aware of things like China's "Social Credit System"[1]. But that is a centralized system. What I'm describing would be controlled by each individual.

[1] https://en.wikipedia.org/wiki/Social_Credit_System