Dirac Delta on Nostr: Imagine I wanted to say I trusted the opinion of pubkeys I verified because I know ...

Imagine I wanted to say I trusted the opinion of pubkeys I verified because I know them in person. Now let's say I travel to some city in Brazil and want to know what the "trusted" opinions are of the nostruber drivers, and nostrbnb lodgings. I would have to rely on the review of the review of the review... to the nth degree, and is such a system really effective? Even if relying on "vouching" works, it clearly works less well the farther apart nodes are, and in many cases, services will be inherently geographical.

If it can be shown the Sybil resistance deteriorates rapidly in a web-of-reviews, determining a better Sybil resistant reputation scheme may be one of the most important open problems in Nostr. I put forth the concept of "proof of transaction" in which interacting parties pay a fee to a relay in order to prove that a transaction was in fact between disparate parties. Another concept is that clients could use different heuristics based on time.

My concept is still rather hand-wavy and may prove infeasible. However, if it does work, then the reputational aspects of a pubkey are arbitrarily portable rather than conditionally portable based on the global network topology. This would be huge.