0.01 bit proof of custody [depends on 256 bit custody atoms]

[dankrad]

Implementing https://ethresear.ch/t/a-0-001-bit-proof-of-custody/7409

[djrtwo]

Did some minor cleanup. In general, this looks awesome.

I'd like to mask get_backing().get_left().merkle_root() into a spec helper with clear indication to implementers what's going on, but we can do that in the larger PR

[djrtwo]

excellent!

[ljq8896]

Sorry but where is the 0.01bit in the spec? Is this still Implementing? I could only see [Validator](Construct attestation
attestation.data, attestation.aggregation_bits, and attestation.signature are unchanged from Phase 0. But safety/validity in signing the message is premised upon calculation of the "custody bit" [TODO].)?

[ljq8896]

Sorry but where is the 0.01bit in the spec? Is this still Implementing? I could only see [Validator](Construct attestation attestation.data, attestation.aggregation_bits, and attestation.signature are unchanged from Phase 0. But safety/validity in signing the message is premised upon calculation of the "custody bit" [TODO].)?

I try to understand it. There will not be a custody bit in an attestation. Honest Validator just don't sign when it's 1 for 0.1% of the time. So if anyone find a Validator sign the bit calculated 1, the validator can be slashed.

[dankrad]

Hi, this is not a place for a discussion. If you have questions, you can ask them for example on the ethresearch post: https://ethresear.ch/t/a-0-001-bit-proof-of-custody/7409