Moderation design
I'm not a huge fan that individual users own substrs. You could do it like Mirage does it and just use hashtags instead of a whole thing to own and moderate. I feel like part of the...
I get that concern, and honestly I think it depends on what layer you consider “centralized.”
The actual posts, replies, reposts, zaps, etc. inside Substr are still regular Nostr events that live on relays and work outside the app. We intentionally avoided building some closed backend feed where content only exists inside Substr.
The “substr” itself is more like a structured community layer on top — mainly for organization, discovery, moderation, rules, layout, filtering, and context. Similar to how a subreddit gives context to posts, except the underlying content is still Nostr-native.
A big part of the design is actually user-created communities. The whole idea behind Substr is helping people find like-minded groups more easily, or allowing creators/projects/shows/interests to have their own dedicated place instead of everything fighting for attention in one endless global feed.
A good example:
- a podcast could have its own substr where episodes, clips, discussions, livestreams, and community posts live together
- an open-source project could have a focused place for updates, feedback, and support
- a niche interest group could have a calmer feed without unrelated noise constantly mixing in
Once you move toward actual community spaces, some form of moderation or curation naturally becomes part of the equation. Otherwise every space eventually turns into “just another hashtag stream” with spam, collisions, and no real continuity or identity around the group.
That said, I’m definitely not claiming the current model is the final answer. I’m very open to experimenting with different moderation structures over time:
- shared moderators - already working
- community voting - already working
- mirror views
- relay-based filtering
- optional open/unmoderated modes
- more collaborative ownership models
I just think there’s a balance between pure protocol idealism and building something that feels usable and understandable for real communities and creators.
So for me it’s less about re-centralizing Nostr, and more about experimenting with a community layer on top of open protocols while keeping the underlying content portable and interoperable.
The actual posts, replies, reposts, zaps, etc. inside Substr are still regular Nostr events that live on relays and work outside the app. We intentionally avoided building some closed backend feed where content only exists inside Substr.
The “substr” itself is more like a structured community layer on top — mainly for organization, discovery, moderation, rules, layout, filtering, and context. Similar to how a subreddit gives context to posts, except the underlying content is still Nostr-native.
A big part of the design is actually user-created communities. The whole idea behind Substr is helping people find like-minded groups more easily, or allowing creators/projects/shows/interests to have their own dedicated place instead of everything fighting for attention in one endless global feed.
A good example:
- a podcast could have its own substr where episodes, clips, discussions, livestreams, and community posts live together
- an open-source project could have a focused place for updates, feedback, and support
- a niche interest group could have a calmer feed without unrelated noise constantly mixing in
Once you move toward actual community spaces, some form of moderation or curation naturally becomes part of the equation. Otherwise every space eventually turns into “just another hashtag stream” with spam, collisions, and no real continuity or identity around the group.
That said, I’m definitely not claiming the current model is the final answer. I’m very open to experimenting with different moderation structures over time:
- shared moderators - already working
- community voting - already working
- mirror views
- relay-based filtering
- optional open/unmoderated modes
- more collaborative ownership models
I just think there’s a balance between pure protocol idealism and building something that feels usable and understandable for real communities and creators.
So for me it’s less about re-centralizing Nostr, and more about experimenting with a community layer on top of open protocols while keeping the underlying content portable and interoperable.