Seeing everything built on nostr is bittersweet: social, communities, survey forms, wikis, journal apps, etc. Finally a protocol that is simple and open enough to see this kind of community input and software growth. But then the community has no urgency to create a more eclectic experience outside of circle jerking Bitcoin. It's very 1 note. Community specific relays and apps that utilize selecting them during user onboarding is one possible way to correct the course.
Discussion
I'm not a big fan of community-specific relays, or really anything-specific relays because I think, for Nostr to work well and to be both easy to use and censorship resistant, relays need to be rather interchangeable in practice. I'm also not sure why we would need community-specific apps.
We do have Communities on Nostr (not NIP-01, so not mandatory) which work similarly as to Reddit and we have hashtags.
I started a community in the current implementation (librearts) so I'm familiar, but people are working on community specific relays. These would let relay owners moderate their own relays more completely. The makers of nostr all have talked about smaller vs bigger relays being the focus. They don't want a system with just a couple relays that everyone's on because it hurts the decentralization and sustainability, but what that means in a global view is that you're not sure who's going to see what you post similar to the issue with mastodon, and that problem extends to the current implementation of communities. For global view this is less of a big deal since you are generally just posting very general things out to anyone following, but for communities, you want to rely in the fact that the whole community can see it including strangers who don't grow you. Community relays fix this issue. People subscribe to relays/communities that they want that can then be smaller relays (or groups of relays for redundancy), but can also be sure that everyone in that community will see the post. And the reason we would want an app is to create the best first experience where you are seeing things you like, and also it would be needed because it's a totally different app concept to the Twitter-replacement app that we're on now. This would be an app more akin to Reddit or Lemmy but with proper architecture where each community can govern itself, be a paid relay, or complete with another community. And an app would just help people sign up and pick topics/communities they like, and then they would immediate have a main feed of strangers posting about things they actually are interested in. The communities could assist choose to be secret and not report to any people running a directory/app. It solves the problem of too big relays, helps the problem of discovering things you like from people you don't even know to look up, and makes them onboarding process easy to get an app tailored to your interests.
The point of using multiple relays is censorship resistance. For censorship resistance to be a thing, the same post need to be on multiple relays and any two people who want to follow each other need to use multiple relays.
> These would let relay owners moderate their own relays more completely.
Why would you want any moderation whatsoever at the relay level?
> be a paid relay
That'd lead to even less adoption. People don't generally pay to be on social media. Not even very very little. Mainstream platforms handle this by having a business model where the non-paying users are what paying users are actually paying for.
> And an app would just help people sign up and pick topics/communities they like, and then they would immediate have a main feed of strangers posting about things they actually are interested in.
But what's of this that can't be achieved with either Nostr Communities or hashtags?
> The communities could assist choose to be secret and not report to any people running a directory/app.
To me this seems to be the only usecase of this.
Why are big relays a problem?