At Open Notes, we're building a system for community-driven constructive moderation and annotation that can be added to anything. Under the hood, we're using the open-source Twitter/X Community Notes algorithm (though that doesn't really kick in until you've got some scale). We're interested in providing everyone with tools for managing discourse that go beyond traditional moderation. Discord is the demo/reference integration, but we want it go anywhere and everywhere. Part of our thesis is that we want to get to where people are already talking rather than drag them to a clean and empty new room where we ask them to continue the conversation.
It's interesting that Pol.is was just recently on HN (https://news.ycombinator.com/item?id=46992815) because we're obviously inspired by them as well as the whole canon of social choice theory--we're just going at it from a different angle. It's long been true that if you wanted to trap me/yourself in a conversation, you could just bring up the Condorcet criterion (amongst others), so I'm finally turning an obsession into an actual product.
We want to enable people to make decisions about conversations as close to the conversation as possible while minimizing impact on live threads. Later, this nicely extends into all sorts of group decisionmaking. As our conversations are increasingly awash in AI of all sorts (as moderators, participants, analysts, etc.), things that help manage the discourse to fit the needs of individual communities need to be scalable but without drowning human choice in an ocean of automation.
Also, we're open-source: https://github.com/opennotes-ai/opennotes
Would love to hear people's thoughts and reactions. This has so much surface area ("all online discourse"), it's hard to formulate specific questions so instead we built a thing and now we'd love to see if it works for folks.