Content pfp
Content
@
https://ethereum.org
0 reply
0 recast
0 reaction

. pfp
.
@chaskin.eth
CT outrage farming is slowing Ethereum’s progress. We need better ways to discuss important issues The same open-source tech behind community notes can help reveal shared understanding instead of fueling division I think the Farcaster protocol is part of the solution, but we need to build a new way on top of it to surface real consensus Here’s my take on how we can make it happen. If you want to help, reach out https://paragraph.xyz/@chaskin/the-time-to-build-a-better-social-network-for-ethereum-is-now
7 replies
5 recasts
28 reactions

dionysuz pfp
dionysuz
@dionysuz.eth
great post, this highlights things that have been on my mind lately. i am a dev and willing to help bring this to life a few things i still want to help figure out, curious if you have thoughts: - how can we also make sure we are including diverse opinions that aren’t so biased (agreed on dunking is bad, but bag bias is also real) - i’ve been thinking of a way to still pull in twitter data read only, perhaps piping through an LLM, just to stay up to date while not engaging with the platform. possibly putting an LLM between - the cost issue, twitter being “free”, might be harder to get more people to join if casts need to be paid for i have more thoughts i’m sure, will need to think about it. but im happy to help. thanks for the great article, this is an insanely important issue and why im also more so on farcaster now!
1 reply
0 recast
1 reaction

. pfp
.
@chaskin.eth
To your 3 points - Good point. We could also explore weighting mechanisms to balance representation across different stakeholder groups (devs, users, researchers, etc.). Open to ideas on making this more robust - Yeah, this could be useful, an LLM filter could help strip out the rage bait while still surfacing real discussions worth tracking. Curious if anyone has experimented with this already - Definitely a consideration. One approach is using Farcaster’s social graph but not requiring paid casts, just using it as an input while running the actual discussion layer separately. But we don’t even need Farcaster for this to work, the Polis algorithm can function independently
0 reply
0 recast
0 reaction