There are no out-of-the-box implementations except in JS.

How can anyone even figure it by themselves if there is not a high-level explanation of what is that they should be figuring out? I wanted to ask nostr:npub1xpuz4qerklyck9evtg40wgrthq5rce2mumwuuygnxcg6q02lz9ms275ams to implement it on Yana, but I didn't even know how.

It's weird because he has an infinitely huge feature list and is implementing everything he can see, but perhaps he didn't realize the outbox model was a possibility?

Reply to this note

Please Login to reply.

Discussion

My point is that we still don't know exactly what's the right approach and we shouldn't make a detailed guideline of how to implement it because we want different ideas

perhaps what would be cool is, for the people that have implemented it (Gossip, Coracle, NDK, Snort), to do a panel or something and discuss lessons learned or something like that because there's a whole lot more than just looking for NIP-65 events when it comes to implementing outbox model (i.e. answering the question "what relays should I read/write for X event/pubkey/tag")

wrt a high level overview; I think Mike's original blog post is a really good read https://mikedilger.com/gossip-model/ and I would plug purplepag.es' website for an extra paragraph on it https://purplepag.es/what

I was thinking more about a generic guideline mentioning existing implementations and approaches and suggesting some future ideas, but I think I should have been clearer on that.

A panel on Nostrasia about this would be awesome, nostr:npub16vrkgd28wq6n0h77lqgu8h4fdu0eapxgyj0zqq6ngfvjf2vs3nuq5mp2va!

I agree that would help.

Could that be a NIP in itself, or rather additional guidelines to NIP-65?

I am aware of the gossip model nostr:npub180cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsyjh6w6.

Actually when I first started using and researching nostr what I immediately imagined how it would work was what is described in this gossip/out-of-the-box.

I was actually surprised that most clients don't do it this was.

It is what it feels natural and more decentralized way of reading content from people you follow.

I realize it is much more demanding to implement in a way it doesn't explode in terms of data usage, memory consumption and overall client stability.

But I'm determined in giving it a try in nostr:npub1gp4xzpmluelsakjtayc4wtzj97fhj5kakqjvsannu00xkdlf4x8s0xdqyq. I hope others will also.