this repo has no description
🔐 Expanded juror weighting section with trust graph rationale, sybil resistance, vouching mechanics and anti-spam penalties
+1
-1
+1
-1
Deep Funding.md
+1
-1
Deep Funding.md
···
127
127
- If there are reviewers/validators/jurors, need to be public so they have some sort of reputation
128
128
- Reputation system for Jurors
129
129
- E.g: whose score is closer to the final one. This biases towards averages
130
-
- Use graph algorithms ([MaxFlow](https://maxflow.one/how-it-works)) to weight jurors
130
+
- Use graph algorithms ([MaxFlow](https://maxflow.one/how-it-works)) to weight jurors. This trust layer makes all human inputs (juror ratings, edge proposals, curation) sybil-resistant and accountable. People don't get "one vote per account": they get influence proportional to how much the network trusts them, based on who publicly vouches for whom. Vouches are binary and simple, but they're recursively weighted by the voucher's own trust and penalized if someone vouches for too many others, which makes spam and fake networks costly.
131
131
- Account for jurors' biases with Hierarchical Bradley Terry or similar
132
132
- Allow anyone to be a juror, select jurors based on their skills
133
133
- Stake-based flow: