Group Thoughts (going beyond yes or no proposals)


Hey everyone!

There was some interest in this topic in the telegram, so I’m writing up a forum topic for it.

I’ve written a blog post called Group Thoughts on this topic. It covers an idea called a social network stack that might be used to develop and execute specific proposals in a decentralized way by taking advantage of the upvote-style systems that have already proven to work well in online communities, rather than depending on any form of strict consensus or all-or-nothing yes-no proposals. Hopefully the blog can make the idea clear!

Last year, I also made a very poorly coded web app version of an example first layer: Delfy. It’s a suggestion social network – each participant can spend a few minutes putting in their best 3 ideas by making new ideas, editing others’, or voting for others and the system will sort the best ones to the top in a simple but pretty powerful way. A management team could use this to get a better sense of what their community needs, since it saves time distilling community comments and makes participation very simple for community members. It’s a way for ideas to realistically filter bottom-up instead of top-down in a large community.

I’d love to see others’ ideas, and I hope this sparks some discussion!


Very nice! Coincidentally, recently in writing documentation for Arc.js I have described the concept of “proposals” as “ideas”.


Important might be to compare betting (staking) on the outcome of the vote on a proposal (as is currently what stakers do with DAOstack’s GenesisProcotol voting machine) versus betting on the outcome of the implementation of the idea being proposed.


That does seem super important. As it is now, we’re incentivizing people to stake on proposals that will pass, not proposals that will have the best outcomes (unless I’m mistaken? Please correct me if so!). That might lead to worse proposals passing (people choose the ice cream, even when they know they should eat the broccoli?).

But incentivizing best outcomes is super, super hard, because it means measuring outcomes and establishing causality. Will the Genesis DAO have agreed upon goals that will be measurable? I don’t see a way to create them in the structure now (though something like Delfy could do it :stuck_out_tongue:). If we did have those goals, we’d then need to plan measuring goal metrics into each proposal–e.g. how many new users did this ad generate if new users is a main goal?

In the long run, incentivizing staking on proposals that will pass and be most successful–instead of just most likely to pass–seems both better and possible. But it’ll take time to set that up, so I think it makes sense to start with the simpler stake-on-passing and see what happens – maybe it’ll end up aligning really well with highly successful proposals even at fairly large scales. Once the genesis DAO is actually running, I think we’ll all have a lot more to say on this stuff :slight_smile: