Superchain in deprecated Armadillo bytes

Stateless clients, sidechains and Sharding
1

If FACON is going to be deployed and deprecate 12 bytes used currently by Armadillo, shouldn’t Rootstock use these fields for the Super parents and uncles instead of doing them inside RSK headers?

struct SuperHeader {
    /// Endorsement of the most recent super parent
    parent: SuperPointer,
    /// Endorsement of the most recent super uncle
    uncle: SuperPointer,
    /// Claim about the RBTC balance in sats
    //// at the end of execution of this RSK block.
    rbtc_balance: Amount
}

/// A pointer to the number of BTC blocks **back** until the 
/// the coinbase transaction with the [SuperHeader] pointed at.
struct SuperPointer(u16); 

struct Amount(u64);

Benefits:

  1. You can detect any PoW Fraud claims on Bitcoin itself, if 1% of miners claim that an RSK block is invalid, they only need to keep mining, and that fork becomes evident on Bitcoin itself, which might be incorporated in BitVM bridges, where you demand that there is no fork longer than X% of the one you are trying to pegout from.
  2. You don’t need to download as much data (or process as much data in Risc Zero) since you don’t need to download the RSK block, which would double the data you need to download for Bitcoin headers and coinbase + its proof.
  3. You don’t need to talk to Rootstock peers or depend on their numbers to see if the network is healthy, you can even ask mempool[dot]space to add some UI helpers for rootstock state, without doing any backend work.
  4. If you accept the Ephemeral Sidechain idea, you can use the rbtc_balance to calculate how many endorsements needed for a Superblock to be considered checkpointed, and then you only download RSK blocks from there, which has bounded growth, while Superchains have unbounded growth.