But that's what happened though. Outbox users found it and interacted with it spreading it around via the old broadcast method.
Right - but it’s back to the same Fiatjaf experiment of months ago. If the users didn’t rebroadcast it to large relays your reach would’ve been drastically impacted. Scaling via outbox requires the reach to happen without the rebroadcasting (so that all relays don’t end up storing all the same notes) so this feels like celebrating a fake win.
I understand and agree with you. I also know that only a few clients support outbox at this time. We need more of this.
Is there documentation for clients to support the outbox model?
Kind of. There are specs and there are blog posts and there are also existing implementations to review - but there isn’t a one size fits all solution, its more of a concept. NIP-65: https://github.com/nostr-protocol/nips/blob/master/65.md Mike Dillger’s post: https://mikedilger.com/gossip-model/ Hodlbod’s post: https://njump.me/naddr1qvzqqqr4gupzp978pfzrv6n9xhq5tvenl9e74pklmskh4xw6vxxyp3j8qkke3cezqq2nskt2w9vx6dznfdvj64rpw4mk5nmxf3v9xsd0gdy
Thank you very much for this
very few clients have so far implemented OUTBOX right?
That’s correct