ord 1.0 tracking issue

[casey]

I think we should have a plan for getting ord out of perpetual beta and making a 1.0 release. For me, I think this is less about some new insane feature, and more about ord reaching a level of feature completeness, stability, and code quality such that it no longer makes sense to call it beta software.

Post 1.0, I think that index incompatibilities should require only a minor version bump, as we have been doing. I think users would just get confused if ord is at version 74.1.3 because we've done 74 breaking changes to the index.

What should be required for 1.0?

• Use proper error types #3192 Not having proper error types makes it too easy to not think about which errors are being produced where, and I the quality of our error messages definitely suffers because we don't think about whether or not we are surfacing enough context for the user to understand the error. I consider this a general code quality and usability issue, so I think it's important for 1.0
• Migrate Wallet to BDK #4173 Wrapping the Bitcoin Core wallet very much feels like a legacy decision at this point, and one which we pay a cost for in terms of velocity and functionality.
• Attributes #4222 Having a good, end-to-end story for legacy collections getting on chain is important, and feels like it should be part of 1.0 for me. Also, I think that structured metadata is a basic NFT standard feature, and inscriptions should have it.

[raphjaph]

What about proper type seperation between the ord and ordinals crate?

Such that ordinals is a pure library with nice types and helper functions and ord is a pure binary crate.

[casey]

What about proper type seperation between the ord and ordinals crate?

I think that would be nice, but not necessarily 1.0 material. For 1.0, I'm thinking about things which are missing basic features and technical debt. We would certainly need to do that for the 1.0 of the ordinals crate itself, but not the ord crate.