#13 - October 19th, 2020
⚡Christian Decker, LN Markets API, Reading, Podcast and much more!
Welcome to the 13th issue of this newsletter which covers market trends, Lightning Network developments and updates about LN Markets.
If you enjoy this content, feel free to spread the word!
🤓 Lightning Research
Christian Decker, Bitcoin enthusiast since 2009 and researcher at Blockstream, made publicly available a dataset of historical gossip messages that enables any researcher to replay the evolution of the Lightning Network over time, to derive statistics, and overall to better understand the network and its future.
In order to demystify gossip communication, the role it plays on the Lightning Network, and how the data he collects can be used for Lightning researchers, we asked Christian a few questions. A warm thanks to him for his precious explanations 🙏
If you find these datasets useful, or use them for your research, feel free to open a PR on his repo to add your publication and ping Christian or #lnresearch on Twitter to add it!
What is gossip communication and how does it work in the Lightning Network?
Christian DECKER: When performing a payment in the Lightning Network, the sender constructs the route that the payment should take through the network. This so called source-routing, is necessary because we use onion routing to hide information from intermediate nodes forwarding the payment.
This is in stark contract with how information is routed on the public network, where each device or router only has to know how to get closer to the destination, and not the full path.
In order to compute a route then an endpoint needs to have a pretty good idea on how the network is structured overall. This is similar to a roadmap on which a driver can plan the streets he is going to drive down in order to get from point A to point B. But how do we get a roadmap for the Lightning Network? This is were the gossip protocol comes into play: the nodes in the network exchange information about the channels that exist in the network, their parameters such as fees, and the nodes themselves. This information is broadcast to all nodes in the network, and so each node can incrementally build up a local view of what the network looks like.
Which type of gossip data do you collect in the Lightning Network?
Christian DECKER: We have two types of objects we share information about in the network:
Nodes: a node may want to advertise to the network at large that it supports some features, such as keysend or MPP support, and we added a couple of vanity fields such as an alias and colors, used mainly for visualizations of the network. Nodes signalling support for features allows us to incrementally roll out new features without requiring all nodes to update in lock-step.
Channels: we can only route a payment directly from node A to node B if there is a channel, so both endpoints announce that there is a channel once it has been confirmed, at which point others may use the channel to route an indirect payment through it. Channels have metadata attached, such as maximum size of payment that it supports, and fees required to route a payment.
For this we have 3 types of messages:
node_announcements: each node may announce its presence on the
network and update parameters at will.channel_announcements: once a channel has reached 6 confirmations the endpoints may create an announcement that tells the rest of the network about the channel's existence. Both endpoints need to sign this message, so if either party doesn't want to announce the channel it'll remain private. A channel can only be announced once.
channel_updates: an endpoint may update the parameters such as fees for one direction of the channel. These parameters may be updated anytime.
All of that information is public, and required for the network as a whole to work. However nodes may opt out of sharing their channels and their node information, that just means that they won't be used to route payments for others, but they can still send and receive without any problems.
The datasets consist of the 3 types of messages mentioned above, collected from a number of nodes I operate, in an aggregated and deduplicated form.
I should note that the datasets reflect only the public network, and not the parts of the network that chose to remain private, which according to my previous research (see video below) consists of about 40% of the network as a whole :-)
Any ideas on how researchers could use this dataset?
Christian DECKER: The dataset can be used to reconstruct the network at any point in time since we started collecting, basically allowing to replay the evolution of the network over time, or an analysis of the structure and behavior of the network.
A couple of publications that already made use of the data did just that, look into the structure of the network, in the form of centrality measures and resilience against node failures, or look into the network population by identifying the implementation used or the (rough) geographic region where the node is running.
But this is just the beginning, and I think people will find the datasets very interesting. I myself am curious what others will uncover, and am looking forward to the new discoveries.
Ultimately this is about informing our decisions going forward: by understanding the current network, its performance metrics, and its limitations we can steer the evolution of the protocol and the network into a better future.
Overall, what's your view on the future of the Lightning Network?
Christian DECKER: I'm very optimistic about the project as a whole: it's a fascinating project, that may finally deliver what everybody was hoping Bitcoin could be back in 2009 when I first got started with Bitcoin. I feel the same incredible enthusiasm in the community, and the drive to dive deeper and explore as I did back then :-)
⚒️ Building
The much requested LN Markets API is now open!
For authentication, you can generate a JSON Web Token (JWT) in your profile section for various scopes (user, positions, deposit, withdraw):
Here is the full doc of the API, have fun Lightning trading bots 🤖!
Our next realase coming soon includes the leaderboard showing the most profitable Lightning traders! What would you want us to build next? Please let us know your thoughts on Twitter and Telegram.
🔮 Reading
Extremely bullish read on Lightning by Max Webster from Version One!
A few points we particularly liked:
His concise definition of the Lightning Network as “a network of bi-directional payment channels that allows for instant settlement between parties and routing of payments over a TOR-like p2p network settled in units of bitcoin”
The value proposition of the Lightning that no other blockchain or smart contracts platform can match: instant settlement in bitcoin with nearly 0 fees and movement of bitcoin (including in DeFi-style smart contracts) without any counterparty risk
The major hurdles yet to overcome: gaining developer mind-share/network effects, gaining meaningful liquidity, overcoming design challenges that limit its scalability across chains, reducing capital intensity, improving UI/UX
His views on the ultimate use cases for Lightning: payments & finance (with gaming, exchanges, P2P derivatives, etc) and Web 3 (with chat, tokens on Lightning, NFTs, etc.)
🎙️ Podcast
For those who share our passion of everything that is at the intersection between Lightning & Finance, we recommand the What Bitcoin Did podcast with Paolo Ardoino, Bitfinex CTO (disclaimer: Bitfinex is an investor in LN Markets).
A few interesting takeaways on Paolo and Bitfinex view on Lightning, starting at minute 40:
Paolo believes in a layered approach to blockchain building, with a Bitcoin layer 1 which should remain as simple and efficient as possible and multiple layers that could add more complexity on top of it
Bitfinex is thinking into enabling Lightning deposits for refilling collateral in derivatives trading positions
Why Bitfinex is funding projects like RGB and OmniLayer: to enable people to move Bitcoin and Tether at the same speed, almost instantly, on the Lightning Network
To summarize, in Paolo’s words:
“It’s important to us to support Bitcoin and to show respect to the coin that made all of this possible.”
⚡ Bonus
Another great tribute to the orange coin “that made all of this possible” by Coinbase, which is searching for at least two Bitcoin development grant recipients:
👋 instant Bitcoin hedging and thank you @BTCCuracao for this very cool video!
Please feel free to reach out to us on Twitter and Telegram.
Thank you all for the support and let’s keep building a BILLION sats company together!