What exactly is the Lightning Network and what is it like now?

In February 2015, Joseph Poon and Thaddeus Dryja published a draft of what they called the "Lightning Network". At the time it was just an incomplete proposal and had no code, but it caused quite a bit of excitement in the Bitcoin community: no wonder, since the idea allowed for near-instant Bitcoin payments between any two parties! This article summarizes the ideas in that draft and the recent developments of the Lightning Network since it was the first step in the development of trustless off-chain Bitcoin transactions.

What is the Lightning Network: A Cache for Bitcoin

One possible thing is for two parties to set up a chain of transactions, where only the last transaction in the chain actually enters the Bitcoin blockchain. This is the idea behind simple payment channels, which already exist; you can keep sending someone replaceable transactions, each with a slightly larger amount than the last, and once a certain condition is met, the channel is terminated and only the last payment is broadcast to the entire network.

It turns out that with only a few, almost non-controversial Bitcoin upgrades, people can create more general payment channels that allow bidirectional payments, and also allow "conditional payments". Conditional payments allow you to build a payment network. In fact, you can say "if Bob pays Carol, I will pay Bob" in a secure and trustless way. After something happens, your wallet automatically broadcasts this conditional payment transaction to the Bitcoin network and waits.

The Lightning Network paper proposes a mechanism for creating channels and payment networks. But similar ideas have been floating around for a while, and this paper brings them together in a powerful focus. Recently in Zurich, Christian Decker and Roger Wattenhofe proposed a different mechanism for doing the same thing, so this is a hot spot for Bitcoin innovation right now.

How is it going?

I started working on the Lightning Network after Blockstream hired me (an experienced Linux kernel and open source developer) in May 2015. I had submitted a series of technical explanations of the initial draft of the Lightning Network paper and had several exploratory conversations with Joseph Poon about possible implementations of the Lightning Network.

We created a public mailing list and in June I started submitting prototype code to Github. These implement the generic channels and conditional payments mentioned above, or more technically, Hash Time Locked Contracts Time Locked Contracts. The approach used is the approach described in the recently published draft paper (with some minor changes). The code runs on Blockstream's Elements Alpha sidechain, which already has the necessary features needed to support Lightning in Bitcoin. We are now completing the first draft of the inter-node network protocol, which includes a full state machine, and two different applications at different stages of development that have been announced on the mailing list. We are also trying and selecting solutions to problems such as how intermediary nodes are discovered and loaded, as well as other technical issues.

Some people ask, why do we have a developer working on Lightning Network? Blockstream is committed to improving Bitcoin so that it can support more applications and use cases. With our background in engineering and protocol development, improving Bitcoin's usability like Lightning Network is our core interest. As a company focused on open source development, it is obvious that this new technology field of Lightning Network is also open source and license-free, just like Bitcoin itself.

Some people also expressed concerns about the Lightning Network from the perspective of Bitcoin's scalability: Does Lightning help Bitcoin's scalability? In the long run, it may be, when small payments become the norm in the Lightning Network, and only anchored transactions and final transactions enter the block. But if the Lightning Network is widely successful, it will inevitably put considerable pressure on the blockchain as a large number of users flock to Bitcoin. Maybe small payments for artists will become Bitcoin's killer application? Maybe instant payments will trigger Bitcoin's second wave of innovation? But no one knows yet.

The underlying engineering work to turn Lightning Network into a real, usable product is being completed, and it will be a playground for developers in the near future. This is exciting work, and we welcome others to join in.

What steps are required before 'actual testing'?

In addition to writing Lightning code and defining the protocol, there are a few features that need to be adopted by Bitcoin in advance:

1. OP_CHECKLOCKTIMEVERIFY (BIP 65)

2.OP_CHECKSEQUENCEVERIFY (BIP 112, which needs BIP 68)

3. Malleability fixes (BIP 62 Bitcoin Improvement Proposal 62)

What will the Lightning Network look like?

One of the most fascinating conversations I had with Joseph Poon, the original author of the Lightning Network paper, was when we imagined what the Lightning Network would look like once it was implemented. We imagined that your phone would have a real-time connection to the Lightning Network by establishing channels with 5 random nodes, and then you could use it to make small payments of varying amounts to anyone on the network.

Even more interesting is that your mobile app can also provide liquidity to the network by moving Bitcoin back to itself. For example, if the flash sale of Satoshi's autobiography triggered a large number of transactions in one direction directed to the seller, the overloaded transaction channel might provide tips to people who want to move funds in the opposite direction, so those people can collect more tips. If your five random channels happen to do this, your mobile app can send some coins to itself and get a small profit.

Will this actually happen? I don’t know! But it just goes to show how different our world would be if a usable, fast, peer-to-peer micropayment network existed.