Subnostr

I agree w/ 1) and 2). However, nothing with unpaid endorsements makes them different - LinkedIn is a place where we can exchange (meaningless?) unpaid endorsements as long as we don't compete w/ each other too much.

That's why PageRank worked on the web - it is very robust to this kind of abuse. If you gave away as much 'weight/sats/etc' as you gained, then you're where you started. Only an asymmetry in the final weight distribution will show who is actually more popular, not a plain count of local events/links, which are easy to fake.

David King 2y ago

also, don’t forget that PageRank has “trusted setup“. you can’t provide relative ranking of everything compared to everything else without telling the system which is doing the calculation what good looks like. everything is relative. stanford.edu was a trusted domain from which the Google crawl began. could probably borrow the concept of ”trusted people” from which the graph of trust is computed based on sats weighted follow edges

Reply to this note

Please Login to reply.

Discussion

brugeman 2y ago

It's important to be able to filter the abuse. If zap-endorsements are easy to fake then I don't think they'd work for the seed at scale.

With nip05 that's more or less trivial - you see a bunch of spammers all from the same domain, and you blacklist it. From experience, even some root zones will have to be blacklisted. How things evolve around zaps is unclear for me now, so we'll see.