As a byproduct of participating in the SPP3 process (so far), I thought of a way to improve signalling processes, which could theoretically improve consensus-finding and by extension decision-making speed for ENS (or any DAO).
We created the equivalent, using ENS-names and their text records for declaring subreddits, and instantiating admins. Voting weight is driven by token holdings or voting power.
I seeded example converstations, that would make sense to me to try out:
We’ve talked about it a lot amongst many delegates and you can even see an early lite paper we published at the d/acc residency; Research Retreat Proceedings
The challenge here comes down to the “Signalling mechanism”
Under simulations we found this does little to flip outcomes given current distributions.
Would be interested to see how you approached this.
I like this problem space, sounds like we’re both circling the same thing! Interesting, that you stumbled across this during your SPP2 experience, and mine came from my SPP3 experience.
There is something here.
From the Signals abstract…
Signals introduces a board-based on-chain primitive that transforms preference intensity into structured, time-weighted support. Participants lock ERC-20 tokens for configurable durations, and each lock mints a transferable ERC-721 position that can be redeemed once an initiative succeeds or expires.
…there is nothing done onchain with holders.vote except ENS records for permissions and ERC-20/721 standards. That’s the beauty of it. It’s not onchain, nothing is binding, it’s nowhere near as heavy as a formal proposal. It’s signal, from some fraction of the DAO. Then, because the conversations are threaded in a way any reddit user would be familiar with, you get this interesting acceleration of the highest signal percolating to the top.
Look at Katherine’s current tempcheck. There are a couple of reactions to a few of the people chosen for the roles the proposal defines. But is this question a material factor to the DAO? Is everybody reading that entire thread? Are people really going to chime in, on every person? Or more likely, skimming, and not able to send appropriate signal efficiently enough to influence a better outcome. Or, did Katherine get the selected people perfect [and therefore those replies can be ignored]?
Doing something like this…
…can allow people to discuss, and weigh in, quickly. Upvoting/downvoting the content of each reply.
If the conversation is important, based on votes, then it gets more visibility. In a normal forum, everybody basically has to skim/read everybody’s random thoughts, even if they don’t matter at all. Like, for instance, my reply to Katherine’s post.
Their are other interesting properties I created with holders.vote, such that any new emerging faction can form on the same platform, using the same token as another faction. And people can follow both, using the same units, but different permissions - set by the leaders of those factions.
From a mechanism perspective, I’m in the camp that we need systems and incentives to drive participation and collaboration. Without them we get what we see today: mostly unstructured, demotivated participants. Lightly echoing @AvsA’s recent point on distribution.
Influence without commitment, in my world, adds noise, and the example you posted somewhat proves the point. I assume it was set up as a demo, but it’s one account with five posts. To the non-diligent reader that’s just more statements to digest at low cost, adding to the information entropy rather than cutting through it.
An adjacent direction worth looking at in this space is @gregskril’s private delegate statements (zk.ens.gregskril.com): Semaphore pools where delegates prove voting power and speak anonymously.
I’m in the camp that we need systems and incentives to drive participation and collaboration.
What if it feels like you need incentives, only because the current systems we use, are competing with more engaging demands for our time? Part of the role of a delegate, is to partially to sift through noise. The harder that job is, the more of an economic cost is associated with participating. If you reduce that cost, I believe participation would climb. Holders.vote does that.
Yes that specific post was set up as a demo, all by me. Its to illustrate. Point me to another platform that could allow any user, with or without VP, to set up a way to capture signal as rich, and as fast, as what one user could do above, with 1 post and 6 comments. Assuming for a second it had critical mass. Katherine could have made that post, just the same as an anon. And if it got upvoted, then it means it should be discussed, and then signal captured.
Other systems don’t solve the new-member curation problem as elegantly as reddit. As soon as medium of collaboration has enough people’s attention, you get spam and noise. So you end up with gates, checks, and moderation. But then it becomes “how do you moderate this?”. Forums, like the one were on, I would argue aren’t as engaging as a reddit-style discussion board. They don’t bubble high-signal up as effectively as reddit, with threaded conversations. Reddit mod techniques have solved this in a balanced way. Ive re-made those techniques in a web 3 native way. Good mods will grow communities, and bad ones will get replaced by competing factions.
I don’t think anyone is saying regular forums are better than a reddit-style system. The way you’ve graphically presented the information is pretty much exactly what we’d expect to see with a Signals implementation. The difference is that Signals allows people to lock up real capital as conviction behind their votes/comments, instead of no-cost actions (which are susceptible to Reddit-style astroturfing and/or low-conviction opinions being amplified).
I would expect a lock up, or even any prompt for a signature at all, to add friction to the UX, such that participation would take more effort. No?
Would it make sense to integrate holders.vote, with the Signals protocol? Such that, it’s an optional to do an onchain txn, to “boost” something.
I think that’d be a smart, and valuable, addition. Is the protocol ready, to build against something like that? I’ll take a deeper look a the repos. Want to collab?
I’m sure a combination of the two is probably ideal. Someone who simply wants to put forward a good idea can do so, and someone who doesn’t have anything to say but wants to lend more credibility to an idea than a single upvote could “boost” it with a lockup.
This piece also covers “endorsements” which could add a new dimension to validating high-conviction opinions: Signals Protocol | Lighthouse Labs
I’d be keen to hear your product ideas! I’ll send you a DM.