Nostr Web Client

nostr:nprofile1qyw8wumn8ghj7mn0wd68ytngw4eh5mmwv4nhjtnhdaexcep0qy88wumn8ghj7mn0wvhxcmmv9uq36amnwvaz7tmwdaehgu3wvf5hgcm0d9hx2u3wwdhkx6tpdshszymhwden5te0wp6hyurvv4cxzeewv4ej7qg4waehxw309aex2mrp0yhxgctdw4eju6t09uq3wamnwvaz7tmjv4kxz7fwdehhxarj9e3xzmny9uq32amnwvaz7tmjv4kxz7fwdehhxarj9e3xwtcpzamhxue69uhhyetvv9ujuurjd9kkzmpwdejhgtcqyr8a0hmz0xdz9cuy5j44m2xyqfkgwkc3n5850sn3dvsvmtyuc8c6v76umwt just read your NIP 77, a couple thoughts.

- On first read, the format is excellent for what it is. I especially like that you included `confidence`.

- I would be hesitant to ask relays to support this. It could be done using special-purpose relays, but DVMs might be a better option.

- I love the ability to request provenance of a trust recommendation by including 30077s in the request. However, this would seem to lock any other metric for trust out of recommendations, making it impossible to make inferences based on kind 7s, 3s, etc., which I think is important to support as there will be orders of magnitude more data from those sources than from explicit attestations. Am I wrong about that?

https://github.com/lez/nips/blob/master/77.md

Reply to this note

Please Login to reply.

Discussion

Lez 1y ago

Thanks for reading! I made a PoC implementation in nostr-rs-relay, it's not complex at all.

Are you suggesting DVMs because of complexity or performance overhead of the querying? I feel because this is essentially data retreival only, there is no computation involved, that DVMs are overkill for this task. However, DVM's could be used for processing the results of this query, run algorithms on top, possibly including kinds 7s and 3s and so on.

I might be too optimistic on this feature, and expect large webs of trusts emerging. But if somehow they would grow big, then kind:7 events, created only by a 30077-trusted group of people would be an excellent input for algorithms to decide on content quality.

Overall, during discussions on this topic I felt there is a need to move away from follow lists because they are too broad already.

hodlbod 1y ago

I see, I was misreading the spec and was thinking it was more complex than it is. Even so, both for complexity and performance reasons it's best to keep relays very simple. A DVM would also allow you to introduce the new format with universal support, rather than only getting support from relay operators that opt-in.

And yeah, I see how other algorithms could actually compose with this. So other than the REQ update, I think this is quite good.

franzap 1y ago

Agree with the REQ update, DVMs are better!