Most new kinds going forward will not have a NIP in the NIP repo.
People need to adjust to the new reality that they can't manage 5 gazillion NIPs and many future NIPs will be for use cases that the repo owners can't understand or evaluate.
Let it go. Move on.
An open protocol doesn't need a Central Committee to run it.
Yep. There’s NIPs for core functionality, then there’s everything else
Yeah, things like npubs/nsec, how to define a relay, how to structure a note, how to delete a note, etc. made sense in a centralized place.
Trying to explain to some Twitter-clone devs why I need some particular NIP for transmitting EU export papers, to cover them with transport insurance for dangerous goods in flight...
Not gonna happen.
I suspect we'll soon reach a point where more NIP development is happening outside of the repo than inside of it.
Which is proof that the protocol works.
I invision sort of NIP-clusters, where people within a certain topic area or branch can geek-out together, without involving people who have no invested interests in their work.
You don't need to ask for permission.
Just build.
So, if you develop a Nostr search, browsing or querying tool and you limit it by kind, you will soon be filtering out most of the notes.
Which is fine, if that is your intention, but it means there's now a market for a more kind-agnostic tool.
And that will be the biggest, most-dynamic market.
Still Nostr might want to learn from XMPP here which tried pretty much the same and practically failed: Too many relevant stuff in extensions, a core protocol so limited it's hardly usable for any serious communication, and users in the end unable to handle this mess because in a combination of clients and servers implementing random extensions in random quality, the best to do is hope that your messages are seen at all and there's little to nothing to be taken as granted in there.
That's why we still need a central storage for NIPs. But that storage shouldn't be a repo.
Repos are a bottleneck.
