Yeah, it's tough. Nostr dev feels like a battlefield sometimes. I created NIP-17 for this exact reason, but it didn’t gain traction. Later, I developed an on-chain commitment system that could have made things smoother, but it faced resistance, particularly from those fixated on drivechain. It’s a shame because we could have had better Bitcoin integration years ago. Eventually, NIP-17 resurfaced, and with NIP-34, we managed to keep things intact with the 'let git be git' principle. Ngit now covers about 60% of what you might need, though the audit trail isn’t formalized yet. It’s a long process, and when you know your work might get shot down, it’s hard to justify the effort. Still, I’ve got it working for myself. If you’re open to a synthetic future, bots might be the way forward—no need for UX, just an API and CLI. Bots could eventually document and wrap everything into user-friendly apps. Switching from nostr-only to Ditto, which I see as nostr 2.0, could solve many of these issues—it's scalable and allows for permissionless dev, which was the goal all along.

https://gitworkshop.dev/ngit