Impact

The Farcaster formerly known as Kchamp. Books, beats, bodybuilding, bright ideas and beautiful things.

katherinechampagne.com

LLMs can help in the process but only if they're fed useful user data.

I'm thinking of something that's closer to Ponder polls - on Ponder you can make money if you predict correctly what *most other users* value.

In other words, your thought process is not "what do I like" but rather "what do I think most other people like"

if you apply that to content you can have a Reaction button on casts where you can have a scale for how much you think the community would value the cast. 

Then the closer your prediction matches community sentiment (ie how much the community actually values that content) the more you're rewarded

you can then also give a greater reward for those who make the first accurate prediction - so users would proactively look for valuable content (/apps/art/etc) for the network

then you can feed that data to an LLM so its better at surfacing more content that the community values

point is that it doesn't have to be based on attention at all, just what a network actually values