Laeserin on Nostr: That's what we currently have. Recommended follows and trending are algorithms based ...
That's what we currently have. Recommended follows and trending are algorithms based upon being an influencer, being a gazillionaire, and/or having friends in high places, as those are the things that primarily determine what is high-quality content, currently.
I am saying that we should have an algorithm (including ones like manual curation, training your own positive and negative filters, and creating topical communities), that actually takes the quality and relevance of THE CONTENT ITSELF into account, when recommending content.
DVMs are the first attempt at this, but they seem to depend a lot on ranking by WoT, so the results are just a subset of trending.
I am saying that we should have an algorithm (including ones like manual curation, training your own positive and negative filters, and creating topical communities), that actually takes the quality and relevance of THE CONTENT ITSELF into account, when recommending content.
DVMs are the first attempt at this, but they seem to depend a lot on ranking by WoT, so the results are just a subset of trending.