I’m now thinking about how to use signature adapters (and PTLC) on the second side to make the protocol even more private:
- Basic signature adapters do not protect from the situation where one party can take money from both outputs (with the knowledge of t). Hence, you need to add a multisignature to the adapters. In the TAS case, the escrow key is formed in such a way that only Bob can take it after Alice publishes the secret.
- This should work seamlessly with both Schnorr signatures and classic ECDSA. Atomic swaps like here require musig
- Just today, we made such a swap between Bitcoin and Ethereum:
Parameters used for the swap k: b550385a62c0eba5a837ee11962e380818a50611caa1113e33a411528ebb193d; K: 03d39da41952d4ae038a49b693b313ec956ad80d9fc940dc5afedf86351e9fa930
Transactions:
- Alice locks BTC: 850e9258bf8b3bb280d32a647198d8024aece543dc283f7bfa526f4c0ceb1ab8
- Bob locks ETH: 723919c0e8ec57d38792ec29b2cb82ee885b9fbbc886b34ff40fb5d3f7cc9b43
- Alice withdraws ETH from the contract: 47546191a7c99ec4a7ddc243d6ea75d345ab3aff0762e09dd2f537731bd484f3
- Bob spends BTC: 859dbfaa901d7106aecc8cb29966ede0c9d7a17c2cae31f4d420c1d770e9706d