I know that several clients calculate #WoT scores and use them to filter content. #wikifreedia, for example, allows me to turn on a WoT filter which currently reduces the number of entries from 503 to 180, with 323 not visible bc they’ve been filtered out.

Is anyone using the WoT scores to stratify content? For example, wikifreedia shows me 11 articles on “nostr” by 11 different authors. Are there plans nostr:npub18a5dah5p2jwvppz04ljj3u2hfdg7p908fy05dz7edz0cwaumhqwsqjpzjs to arrange those in order from the highest to the lowest WoT score? Seems like that’s the next step.

Is anyone else already doing anything like this? (For any content, not just wiki.)

Reply to this note

Please Login to reply.

Discussion

Dammit I meant to tag nostr:npub1l2vyh47mk2p0qlsku7hg0vn29faehy9hy34ygaclpn66ukqp3afqutajft — for some reason damus doesn’t show me your profile when I try to select it for a note.

just saw this

entries are ranked by wot within each topic

I see the rankings now, although I hadn’t previously noticed them when I made the post. I must’ve looked right past them!

Looks like we’re calculating the WoT score using similar methods. Our rankings are almost exactly the same order! I’d say this is the most useful method available right now, although it’s still a bit of a popularity contest. What if the best author is someone who’s recognized as high quality, but has a low follower count? Our WoT score method does not know how to boost a person like that!

So my short term plan is going to be to experiment with some alternate scoring systems that will boost author scores based on contextual data. There are various ways to do that, and the challenge is making it sophisticated enough to add value but not too complicated for devs to implement or users to use.

There is a PR for a nip that is better suited for trust rankings on specific topics than using follower lists to calculate wot scores:

https://github.com/nostr-protocol/nips/pull/1220

I also like this proposal by nostr:npub1elta7cneng3w8p9y4dw633qzdjr4kyvaparuyuttyrx6e8xp7xnq32cume for contextual trust attestations:

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

I’m wondering whether anyone has implemented step 3 of this roadmap:

https://habla.news/a/naddr1qvzqqqr4gupzpef89h53f0fsza2ugwdc3e54nfpun5nxfqclpy79r6w8nxsk5yp0qqxnzdesxqerwvesxvmn2dpj5hl046

Surely someone has, and I just haven’t seen it yet. Doing so should motivate users to make use of Lez’s NIP-77 as well as arthurfranca’s Trust Rank NIP which you linked.

Coracle sorts search results by wot

Is that what the From network option does? How do you combine the WoT score and the age of the post?