State Channel

Basics

Allows users to transact unlimited times off-chain with two on-chain transactions.
From The Daily Gwei #11 (19-6-2020):

"The concept of State Channels has been [around] for a while (it’s the technology that the Lightning Network on Bitcoin uses) and is currently on Ethereum in the form of the Raiden Network, Celer Network and more."

"A State Channel is in essence a two-way discussion channel between users, or between a user and a service (a machine). Messages take the form of transactions, such as “I want to buy a beer for 3$” or “I want to rent this TV channel for one hour for 5$”. Participants sign each message of the discussion, making the series of transactions impossible to refute later on."
From this blog by PlotX (25-1-2021):

"When it comes to scaling solutions, Channels are one of the first widely discussed solutions. They allow participants to exchange transactions off-chain x number of times, while only submitting two transactions to the base layer. The most popular types of channels are state channels and their subtypes — payment channels.

Although channels have the potential to easily process thousands of TPS, some of their downsides are that while using channels, participants have to lock-up their funds in a multisig contract, which practically means no support for open participation. Moreover, this scaling solution is application-specific and cannot be used to scale general-purpose smart contracts."

From this article (16-12-2019) for an in-depth look into channels:

"State channels are a very broad and simple way to think about blockchain interactions which could occur on the blockchain, but instead get conducted off of the blockchain, without significantly increasing the risk of any participant. The most well known example of this strategy is the idea of payment channels in Bitcoin, which allow for instant fee-less payments to be sent directly between two parties. Here is a list of some of the State Channel Projects.

State channels are the general form of payment channels, applying the same idea to any kind of state-altering operation normally performed on a blockchain.

Moving these interactions off of the chain without requiring any additional trust can lead to significant improvements in cost and speed. State channels will be a critical part of scaling blockchain technologies to support higher levels of use.

The basic components of a state channel are very simple:

A 2 way state channel

Part of the blockchain state is locked via multisignature or some sort of smart contract, so that a specific set of participants must completely agree with each other to update it.
Participants update the state amongst themselves by constructing and signing transactions that could be submitted to the blockchain, but instead are merely held onto for now. Each new update “trumps” previous updates.
Finally, participants submit the state back to the blockchain, which closes the state channel and unlocks the state again (usually in a different configuration than it started with).

That’s it! If the “state” being updated between participants was a digital currency balance, then we would have a payment channel. Steps 1 and 3, which open and close the channel, involve blockchain operations. But in step 2 an unlimited number of updates can be rapidly made without the need to involve the blockchain at all — and this is where the power of state channels comes into play, because only steps 1 and 3 need to be published to the network, pay fees, or wait for confirmations. In fact, with careful planning and design, state channels can remain open almost indefinitely, and be used as part of larger hub and spoke systems to power an entire economy or ecosystem.

Despite my simple description here, state/payment channels have generally been perceived as quite complicated. There are several reasons for this, and one of them is that there are some important subtleties hidden in my phrasing of the three steps. Let’s take a closer look at what these simple phrases imply, starting with:

"could be submitted to the blockchain"

In order for state channels to work, participants have to be assured that they could publish the current state of the channel to the blockchain at any time. This results in some important limitations, such as the fact that someone has to stay online to protect each individual party’s interests until the channel is closed.

Imagine that when we initiated a payment channel I started with 100 bitcoins and you started with 10. If we first sign an update that transfers 10 of those bitcoins to me, and then later sign an update that transfers 50 back to you, the later update is obviously more beneficial to you than the earlier one is. If you were to unexpectedly lose internet access, and I were to pretend the second update never happened, I might be able to publish the first update to the blockchain and effectively steal 50 bitcoins from you! What you need is somebody to stay online with a copy of that later transaction so that they can “trump” the earlier one and make sure your bitcoins are protected. It doesn’t have to be you — you could send a copy to many random servers who agree via smart contract to publish it only if needed (for a small fee of course). But however you do it, you need to be assured that the latest signed update to the state is available to trump all others. Which leads us to our next subtle phrase:

"Each new update “trumps” previous updates"

To make this part of the state channel work, the locking and unlocking mechanisms have to be properly designed so that old state updates submitted to the blockchain have a chance to be corrected by the newer state updates which replaced them. The simplest way is to have any unlocking attempt start a timer, during which any newer update can replace the old update (restarting the timer as well). When the timer completes, the channel is closed and the state adjusted to reflect the last update received. The length of the timer would be chosen for each state channel, balancing the inconvenience of a long channel closing time with the increased safety it would provide against internet connection or blockchain problems. Alternatively, you could structure the channel with a financial penalty so that anyone publishing an inaccurate update to the blockchain will lose more than they could gain by pretending later transactions didn’t happen.

But the mechanism ends up not mattering very much, because (going back to the previous point) the game theory of this situation puts a twist on things. As long as this mechanism is theoretically sound, it will probably never have to be used. Actually going through the timer/penalty process may introduce extra fees, delays, or other inconveniences; given that forcing someone into the mechanism can’t give you any advantage anyways, parties to a state channel will probably just close the channel out by mutually agreeing on a final channel state. This final close-out operation needs to be fundamentally different from the normal “intermediate” updates (since it will bypass the “trumping” mechanism above), so participants will only sign a final close-out transaction once for each portion of the state locked within a particular channel.

The details of these “subtleties” aren’t especially important. What it all ultimately breaks down to is that participants open the channel by setting up a “judge” smart contract, sign promises to each other which the judge can enforce and adjudicate if necessary, and then close the channel by agreeing amongst themselves so that the judge’s adjudication isn’t needed. As long as the “judge” mechanism can be assumed to be reliable, these promises can be counted as instant transfers, with the judge only appealed to in exceptional circumstances, such as when one party disappears.

Of course, these details are only part of the reason people think that state/payment channels are complicated. A much bigger one is that Bitcoin payment channels are complicated. Building a “judge” mechanism in Bitcoin with even reasonably useful properties is surprisingly intricate. But once you have a clear concept of state channels in general, you can see that this only comes from trying to implement the idea in a constrained context. Basic smart contract features like a timer mechanism and allowing two different paths to be taken depending on the signed message submitted are just plain harder to do in Bitcoin. Some of these features are being gradually added or built. By seeing that payment channels are only a special subcase of the broader “state channel” idea, we realise that this is a much broader technique, and that state channels can apply to any smart contract which deals with frequent updates between a defined set of participants. You can anticipate seeing this approach in many (if not most) distributed applications going forward."

Pros and cons using it for scaling

State Channel pros

State channels have strong privacy properties: This is because everything is happening “inside” a channel between participants, rather than broadcast publicly and recorded on-chain. Only the opening and closing transactions must be public. Whereas in sidechains every transaction is published on the sidechain which is received by every participant on the sidechain irrespective of the fact that you are not interacting with all of the participants on the sidechain.
State channels have instant finality, meaning that as soon as both parties sign a state update, it can be considered final. Both parties have a very high guarantee that, if necessary, they can “enforce” that state on-chain. But as discussed above that the state channel closing could take variable time considering the security level of the transaction. Whereas in sidechains you have a blockchain on the other side. So the finality depends on the mining power of the sidechain.

State Channel cons

State Channels need 100% availability of all the participants involved: As we have discussed above that if anyone the participants goes unavailable, then this could prove costly to him. The participants can use someone to represent him if he goes unavailable, but the possibility of the representative getting attacked or bribed makes it a problem for state channel. Whereas in sidechains you don’t have to be available all the time you are on the sidechain.
State channels are best used for applications with a defined set of participants: This is because the Judge contract (the contract used to lock the state) must always know the participants/entities (i.e. addresses) that are part of a given channel. We can add and remove people, but it requires a change to the contract each time. Whereas in sidechains there is no such limitation on the movement of the participants.
State Channels are particularly useful where participants are going to be exchanging many state updates over a long period of time: This is because there is an initial cost to creating a channel in deploying the Judge contract. But once it is deployed, the cost per state update inside that channel is extremely low."