Statechains are an unique second layer protocol initially developed by Ruben Somsen in 2018, relying on the eltoo (or LN Symmetry) proposal. In 2021 a variation of the unique proposal, Mercury, was constructed by CommerceBlock. In 2024, an extra iteration of the unique Mercury scheme was constructed, Mercury Layer.
The Statechain protocol is a little more sophisticated to debate in comparison with different techniques resembling Ark or Lightning due to the vary of variations which might be potential between the unique proposed design, the 2 which have been truly carried out, and different potential designs which have been loosely proposed.
Like Ark, Statechains rely upon a centralized coordinating server to be able to perform. In contrast to Ark, they’ve a barely totally different belief mannequin than a vUTXO in an Ark batch. They rely upon the coordinating server to delete beforehand generated shares of a non-public key to be able to stay trustless, however so long as the server follows the outlined protocol and does so, they supply a powerful safety assure.
The overall thought of a Statechain is to have the ability to switch possession of a whole UTXO between totally different customers off-chain, facilitated by the coordinator. There isn’t a requirement for receiving liquidity like Lightning, or the coordinator server to supply any liquidity like Ark.
To start, we are going to have a look at the unique protocol proposed by Ruben Somsen.
The Unique Statechain
Statechains are successfully a pre-signed transaction permitting the present proprietor of the Statechain to unilaterally withdraw on-chain each time they need, and a historical past signed messages cryptographically proving that previous homeowners and the receivers they despatched the Statechain to accepted these transfers.
The unique design was constructed on eltoo utilizing ANYPREVOUT, however the present plans on easy methods to allow the identical performance make use of CHECKTEMPLATEVERIFY and CHECKSIGFROMSTACK (a excessive stage clarification of that is on the finish of the CHECKSIGFROMSTACK article). The essential thought is a script enabling a pre-signed transaction to spend any UTXO that has that script and locks the suitable quantity of bitcoin, relatively than being tied to spending a single particular UTXO.
Within the protocol, a person wishing to deposit their cash to a Statechain approaches a coordinator server and goes via a deposit protocol. The depositing person, Bob, generates a key that can be uniquely owned by him, but additionally a second “transitory” key that can ultimately be shared (extra on this quickly). They then craft a deposit transaction locking their coin to a multisig requiring the coordinator’s key and the transitory key to signal.
Utilizing this multisig, Bob and the coordinator signal a transaction that spends that coin and creates a UTXO that may both be spent by some other transaction signed by the transitory key and the coordinator’s key utilizing LN Symmetry, or Bob’s distinctive key after a timelock. Bob can now fund the multisig with the suitable quantity, and the Statechain has been created.
To switch a Statechain to Charlie, Bob should undergo a multistep course of. First, Bob indicators a message along with his distinctive non-public key that attests to the actual fact he’s going to switch the Statechain to Charlie. Charlie should additionally signal a message testifying to the truth that he has obtained the Statechain from Bob. Lastly, the coordinator server should signal a brand new transaction permitting Charlie to unilaterally declare the Statechain on-chain earlier than Bob sends Charlie a duplicate of the transitory key.
All of that is made atomic utilizing adapter signatures. These are signatures which might be modified in such a means utilizing a random piece of knowledge that renders them invalid, however could be made legitimate once more as soon as the holder of the signature receives that piece of knowledge. All the messages, and the brand new pre-signed transaction are signed with adapter signatures, and atomically made legitimate on the similar time via the discharge of the adapter knowledge.
Holders of a Statechain should belief that the coordinator server by no means conspires with a earlier proprietor to signal a direct closure of the Statechain and steal funds from the present proprietor, however the chain of pre-signed messages can show {that a} coordinator has participated in theft in the event that they had been to take action. If a previous proprietor makes an attempt to make use of their pre-signed transaction to steal the funds, the timelock on the spend path utilizing solely their key permits the present proprietor to submit their pre-signed transaction and appropriately declare the funds on chain.
Mercury and Mercury Layer
The unique Statechain structure requires a softfork to be able to perform. CommerceBlock designed their variant of Statechains to perform with out a softfork, however so as to take action tradeoffs had been made by way of performance.
The essential thought is identical as the unique design, all customers maintain a pre-signed transaction that permits them to assert their funds unilaterally, and the coordinator server nonetheless performs a task in facilitating off-chain transfers that requires them to be trusted to behave actually. The 2 main variations are how these transactions are signed, and the construction of the pre-signed transaction customers are given.
The place the signing is worried, there is no such thing as a longer a transitory non-public key that’s handed from person to person. As a substitute of this, a multiparty-computation protocol (MPC) is used in order that the unique proprietor and the coordinator server are in a position to collaboratively generate partial items of a non-public key with out both of them ever possessing the total key. This secret’s used to signal the pre-signed transactions. The MPC protocol permits the present proprietor and coordinator to have interaction in a second protocol with a 3rd occasion, the receiver of a switch, to regenerate totally different items that add as much as the identical non-public key. In each the Mercury and Mercury Layer protocol, after finishing a switch an trustworthy coordinator server deletes the important thing materials similar to the earlier proprietor. So long as that is executed, it’s not potential for the coordinator to signal a transaction with a earlier proprietor, as the brand new piece of key materials they’ve is just not appropriate with the piece any earlier proprietor would possibly nonetheless have. That is truly a stronger assure, so long as the coordinator is trustworthy, than the unique proposal.
The pre-signed transaction construction for Mercury and Mercury Layer can’t use LN Symmetry, as this isn’t potential with out a softfork. In lieu of this, CommerceBlock opted to make use of decrementing timelocks. The unique proprietor’s pre-signed transaction is timelocked utilizing nLocktime to a time far out sooner or later from the purpose of the Statechain’s creation. As every subsequent person receives the Statechain throughout a switch, the nLocktime worth of their transaction is a few pre-determined size of time shorter than the earlier proprietor. This ensures {that a} earlier proprietor is incapable of even making an attempt to submit their transaction on-chain earlier than the present proprietor can, but it surely additionally signifies that ultimately in some unspecified time in the future the present proprietor should shut their Statechain on-chain earlier than earlier homeowners’ transactions begin turning into legitimate.
The foremost distinction between Mercury and Mercury Layer is how these transactions are signed. Within the case of Mercury, the coordinator server merely sees the transaction proposed, verifies it, after which indicators it. Mercury Layer makes use of a blind-signing protocol, which means that they don’t truly see any particulars of the transaction they’re signing. This necessitates the server monitoring Statechains utilizing anonymized information on the server, and a particular authorization key of the present proprietor in order that they are often certain they’re solely signing legitimate transfers.
Synergy With Different Layers
Statechains can synergize with different Layer 2s which might be based mostly on pre-signed transactions. As an illustration, a part of the unique proposal steered a mix of Statechains and Lightning Channels. As a result of each are merely pre-signed transactions, it’s potential to really nest a Lightning channel on high of a Statechain. This merely requires the present proprietor’s unilateral exit key to be a multisig, and the creation of the pre-signed transactions spending that output right into a Lightning channel. This permits Lightning channels to be opened and closed totally off-chain.
In a similar way, it’s potential to nest a Statechain on high of a vUTXO in an Ark batch. This merely requires the pre-signed transactions mandatory for a Statechain to be constructed, spending the vUTXO output.
Wrapping Up
Statechains usually are not totally trustless, however they’re a really belief minimized scheme that may be very liquidity environment friendly and permits freely transferring UTXOs off-chain between any customers keen to simply accept the belief mannequin of Statechains.
Whereas the unique proposal has but to be constructed, the 2 implementations designed by CommerceBlock have been utterly carried out. Each failed to attain something greater than marginal use in the actual world. Whether or not this is because of customers being unwilling to simply accept the belief mannequin concerned, or just a failure in advertising or consciousness is one thing that can not be absolutely ascertained.
Regardless, on condition that there are two full implementations and designs for a extra versatile variation ought to LN Symmetry ever grow to be potential on Bitcoin, this an possibility that can at all times be right here. The great factor about open supply software program is that it’ll at all times be there no matter whether or not individuals use it now, ought to they select to sooner or later.
