maybe theres no point to protocol dev anymore, we should all back to making proprietary backends so we can deliver real value instead of handicapping ourselves with the hard work of interoperability. I can see primal taking over marketshare and us protocol devs going into obscurity.
Discussion
The HORNETS Relay is now equipped with immudb — a tamper-resistant database supporting:
-key-value store
-JSON document store
-SQLite store
All 3 are tamper-resistant. Shout out to Adam our main backend dev for finding immudb. We’ve spent the last month or so migrating over to it, hence the delay for launching.
Originally we used graviton for our tamper-resistant db, but graviton only supported key-value store. Tamper-resistant dbs are extremely rare.
The beauty of tamper-resistant dbs is it becomes hard for an attacker to secretly delete content from the db of a relay. An attacker would have to delete the relay’s entire db. Relay operators would obviously notice this, creating intense tamper-evident proof.
The document store enables querying. Haven’t used it for plain text querying yet — mainly using for the git notes currently. I was going to make it a surprise, but you asked!
We’ve been building this relay to be a sort-of universal backend for any nostr app. Obviously not all apps will follow.
I think it’s counter-productive for everyone to make their own implementation. That’s what Martti thought about frontend work too and merged with us per the same point. Same applies to the backend. Would love to team-up with Damus especially.
How is this not just another un-spec’d relay?
Documentation will be out soon
Like I said it’s not even ready yet
It’s all in one… that’s the idea. Blossom, Merkle trees, negentropy, all 3 DB types, relay dashboard for managing it, bitcoin payments for being a paid relay based on GB allocation per user.
It literally lets you toggle which NIPs you want to enable from the dashboard. Modular as it gets. Picking up where strfry left off.
I think there’s another problem too, with the NIPs repo declining in quality (anything can be a spec as long as you market it well enough to unsuspecting people)
Actual improvements to protocol get ignored
Also, are you going to be implementing the login with nostr proposal I made?
