It’s not that people don’t understand the buttons. The issue is that voting is meant to do more than record whether someone likes or dislikes a post. It’s supposed to help surface useful contributions and moderate the discussion. I’d expect users to respect that purpose, but the system also makes it easy to use votes as simple approval and disapproval, and that’s the behavior it often rewards.

Thanks, that helps clarify the differences between Lemmy and PieFed. My concern isn’t mainly who can see vote records. It’s that votes can influence which posts and comments people see.

My main issue is that people often use votes as likes and dislikes, even though they’re also meant to help moderate content by handing some of that responsibility to the user base. In practice, that can be abused. I’m not exactly sure how Lemmy calculates scores or how much they affect visibility, but if downvotes were purely cosmetic, I wouldn’t have the same concern. Mass downvotes can bury good posts, while mediocre posts can get boosted simply because they’re popular.

Users can choose how votes factor into their feeds, and disabling downvotes locally won’t stop other instances from using their own votes. I’d still want reactions to be cosmetic, or at least never used by the protocol to rank or suppress content.

The topic and flair idea overlaps with what PieFed already does. The part I’m still trying to work out is whether spaces should be entirely local to each instance or whether instances could optionally share directory entries. I don’t want the largest instance to control the organization, and smaller instances should be free to organize things differently. A shared directory, if there is one, would need to be optional and not authoritative. I haven’t figured out the governance for that yet.

Yes, PieFed already organizes communities into topics, so that part of the idea isn’t new. I don’t mean nodes when I say spaces. Nodes host the service and communities; spaces are a separate directory layer that associates with communities. The distinction I’m exploring is a shared, federated directory that isn’t owned by a node or by any one community. I’d need to explain more clearly how that differs from PieFed’s topics.

Votes and karma aren’t the same thing, agreed. My concern is that Lemmy’s votes still affect what people see. When posts and comments are ranked by votes, visibility is shaped by the user majority, which doesn’t necessarily reflect relevance or quality. Voting can help surface useful content, but it isn’t a substitute for moderation.

I’m not proposing a restricted set of communities. Anyone could create a community. The difference is that communities could be associated with a shared, global directory of spaces, and no one would own that directory. Spaces wouldn’t be communities themselves; they’d help organize and connect communities around subjects. For example, a History space could list several independently run history communities.

I’m not saying votes are useless. The concern is that a karma system can’t reliably distinguish relevance or quality from popularity, disagreement, or a coordinated downvote. Votes may work well for you, but they can still bury a worthwhile post or boost a bad one. I also agree that without karma, the system would be meaningfully different from Reddit.

I’m not proposing a wiki where everyone edits the same pages. Threads would remain discussions, and spaces and subjects would be reusable tags. Moderators would handle rule enforcement, not own or control the subject matter.

I think good moderation could make the difference. If moderators are chosen for fair, merit-based judgment and their decisions are transparent and appealable, paying them could be worth the investment. The forum might not work for everyone, but it could flourish for people who want that kind of discussion.

I’m sorry, I thought that this was a general community.