I think this is a great idea. And I have a solution to the bootstrap problem.
The solution is: interpretation of proxy data “as if” it were NIP-77 formatted data. For example, if you “like” a wikifreedia article filed under Category X, my software interprets that “as if” you issued a NIP-77 attestation, to that author, in the context of X. But since like != trust (well, maybe it does and maybe it doesn’t, I have no way of knowing), I “interpret” a really low confidence, like maybe 1% or 5%. That way, we start out with a sea of NIP-77-formatted trust data. And you’ll know that if you REALLY trust someone in some context, like if I REALLY trust LynAlden (npub1a2c…w83a) to write wikifreedia articles on economics, and if I want my app to give an added boost to her other content and I’m OK with saying this publicly, then I’m gonna have to issue an actual NIP-77 attestation and set the confidence to something more meaningful than 1% or 5%. In this way, the pool of high quality contextual trust attestations can accumulate gradually.
straycat / david
npub1u5…rldq3
2024-05-19 16:30:33
in reply to nevent1q…axm0
Author Public Key
npub1u5njm6g5h5cpw4wy8xugu62e5s7f6fnysv0sj0z3a8rengt2zqhsxrldq3Published at
2024-05-19 16:30:33Event JSON
{
"id": "f5fbe0d8843d419b21f494ba4c6a354254b976b29ac37772bbb808d591978b21",
"pubkey": "e5272de914bd301755c439b88e6959a43c9d2664831f093c51e9c799a16a102f",
"created_at": 1716136233,
"kind": 1,
"tags": [
[
"e",
"2496aa39c85f8ae37b5eb3ba1999274cad4f0501cd8933218887962cb0f519ae",
"wss://purplepag.es/",
"root"
],
[
"e",
"ae8e42f31d6fdf3f3f82a1d23dd57f69a6635a419d8946ffebbb024cf7dd84ff",
"",
"reply"
],
[
"p",
"cfd7df62799a22e384a4ab5da8c4026c875b119d0f47c2716b20cdac9cc1f1a6"
],
[
"p",
"eab0e756d32b80bcd464f3d844b8040303075a13eabc3599a762c9ac7ab91f4f"
]
],
"content": "I think this is a great idea. And I have a solution to the bootstrap problem.\n\nThe solution is: interpretation of proxy data “as if” it were NIP-77 formatted data. For example, if you “like” a wikifreedia article filed under Category X, my software interprets that “as if” you issued a NIP-77 attestation, to that author, in the context of X. But since like != trust (well, maybe it does and maybe it doesn’t, I have no way of knowing), I “interpret” a really low confidence, like maybe 1% or 5%. That way, we start out with a sea of NIP-77-formatted trust data. And you’ll know that if you REALLY trust someone in some context, like if I REALLY trust nostr:npub1a2cww4kn9wqte4ry70vyfwqyqvpswksna27rtxd8vty6c74era8sdcw83a to write wikifreedia articles on economics, and if I want my app to give an added boost to her other content and I’m OK with saying this publicly, then I’m gonna have to issue an actual NIP-77 attestation and set the confidence to something more meaningful than 1% or 5%. In this way, the pool of high quality contextual trust attestations can accumulate gradually.",
"sig": "68b63446ac743ba26254ac9c721d3d8b014ce7743da1ec1c68c1ea399e80cbfa57bc525866be3e2505ab07d80d2b6271e834d27e738e2c79bb2085f73f865dbc"
}