We interviewed the person who controls 40% of Lightning Network's capacity. Here's what they had to say

Since launching on the Bitcoin mainnet in early 2018, the Lightning Network has experienced significant adoption and growth. The first Lightning Network implementation developed by Lightning Labs was launched in beta in March 2018. The other two popular implementations were developed by ACINQ and Blockstream which were launched in late March and late June, respectively. 

Since its launch, the Lightning Network has actively been stressed tested in the public. In less than two years, the layer two scaling solution for Bitcoin has grown to nearly 10,000 public nodes with over 36,000 active channels with a total capacity of almost $7 million at the time of this writing.

Behind the Lightning Network’s growth is LNBIG, a mysterious entity that, at times, made up over 40% of the Lightning Network’s capacity. Recently, The Block reached out to the anonymous individual behind the project to see if they would answer some burning questions our team and members of the community have.

Our conversation covered a wide range of topics, from the person’s background, their reason for launching so many Lightning nodes, their vision for the future of Bitcoin and the Lightning Network, and how they feel about the growth of Ethereum and DeFi.

This interview has been edited for clarity

The Block: Do you want to share something about yourself?

I have been familiar with bitcoin since the beginning of 2014. I am a programmer, and I was initially skeptical about this system. When I first read the Bitcoin whitepaper, I understood the point. Then I tried the “Bitcoin Core” (“Bitcoin Qt” at the time) node, and I waited a long time for it to sync, which did not impress me. Only after I tried Andreas Schildbach’s Bitcoin Wallet and made several payments between it and Bitcoin Core and the bitcoin transactions go through immediately — only then was I really impressed by this technology.

Then I spent days reading the Bitcoin Wiki, delving into the protocol, doing mental experiments on how they could break the system. It was then that I realized that bitcoin is a great way to preserve value. It was at that time that I began to actively transfer all my savings and salary into bitcoin. The price fell from $600–$700 to $230 back then but I did not have a bit of doubt and continued putting all my savings into bitcoin. I heard a lot from my friends about the problems with banks that close accounts for offshore companies, about the unattainable conditions that they put before their clients, and that only strengthened my confidence in Bitcoin.

The Block: How much of LN’s capacity do LNBIG’s nodes currently control?

The sum of all local balances in channels in LNBIG network is 331.5 BTC. The total LN capacity is 832.2 BTC. But, I have 1,800 channels between my nodes and each channel with a capacity of 0.16. So LNBIG’s capacity is about 40% of the total capacity. About 288 BTC is allocated to channels (0.16*1800) among themselves. If you take this into account, then the channels between other nodes have about 48 BTC. It’s 5.7% only.

The Block: According to 1ML, you currently have close to 60% of capacity. Why is there such a large discrepancy?

The Block: What is the most interesting thing you learned by running 40% of LN’s capacity?

I began to get a better understanding of how channels work. I think that I already know roughly how many transactions are currently going through LN. I tried my hand at programming in Node.JS and Vue to build my site, lnbig.com.

The Block: What was the reason that you launched so many LN nodes? You said that you don’t run the nodes for financial reasons but rather as a way to support the network. Is that still the case?

At first, I had no idea I would be running so many nodes. I experimented in March-April 2018 with the testnet. I also wanted to understand the technical side. At first, I tried c-lightning, but I had problems compiling it. Then Eclair fell away because not all OS’s have the correct version of Java, and installing it from the source is a painful process. Then I tried LND and, surprisingly, it turned out to be easy to compile and run.

Having played with LND, I temporarily abandoned this experiment until the fall. In fall, services began to appear that accept Lightning and I wanted to pay.

It is worth noting that I realized very early on that it was very important to support merchants with [bitcoin] purchases. Initially, I switched all my earnings to bitcoin. But I paid with bitcoin wherever possible. It is very important to give bitcoin feedback to merchants, as a client, letting them know that you exist, that they knowingly connected a payment method with bitcoin. Otherwise, they may refuse this method of accepting payments which would make it worse for everyone who works in the bitcoin sphere. So, I wanted to pay with bitcoin, I had bitcoin. I found the Eclair Wallet. Connected to the ACINQ node. But many payments did not go through because there were no routes. Everything worked badly, and I also had to pay commissions to a third party.

I manage the servers at my company, and at that time, many capacities were underloaded due to a drop in search traffic on sites. I thought to myself, “why not use an idling server to start a node?” At that time, a convenient script appeared for installing an LND node that isolated the system from the source texts. I used this to install nodes without affecting the system. Then the idea grew into the launch of several nodes because I had accumulations of bitcoins and free server capacities.

Along the way, I wanted to know how many transactions happen on the Lightning Network. I realized that this could be found out only if you run many nodes and channels so that most of the payments choose routes through your nodes.

Also, at that time, it seemed to me that I could even earn a little on commissions. Curiosity, the desire to support the network and merchants, and not having to depend on anyone when making payments — these were my motivations. I had experience managing a large number of servers. I had no problems starting twenty nodes, and then I launched five more.

The Block: Based on the growth of LN so far, are you bullish on its success? Is it too early to judge LN’s success?

This is not the first time I had to wait for the success of a technology. In my eyes, bitcoin has grown from $230 to $20,000. I have no doubt in Bitcoin and the Lightning Network. In any case, the Lightning Network is the same breakthrough in itself as Bitcoin. Micropayments are the future. Especially without waiting a long time. Over the year, I saw how many merchants on the Lightning Network appeared. I see no reason for this to slow down. It may take another year for LN to get the properties users need. The infrastructure for payments will be formed. But the basic infrastructure is already there, and it’s becoming significant. I am now so used to paying with LN that I use on-chain transactions only if the merchant does not accept LN or if it’s a payment above the current LN’s limits.

The Block: What do you think LN developers need to add to make LN more attractive for Bitcoin users?

I think it’ important for AMP payments to work. Of course, you need widespread support for this technology in mobile wallets. The AMP will play a big role in Bitcoin ATMs. Bitcoin ATM developers should pay attention to it. But even without AMP, the use of ATMs would be a very convenient improvement, especially if the ATM could issue coins.

As a person who constantly pushes Bitcoin to people, I would like to see the WOW effect in mobile wallets while allowing users to preserve control over their funds. By the WOW effect, I mean that a user could quickly take out their wallet, immediately send several satoshis, and feed chickens, for example.

That can already be done by using BlueWallet, but that’s still a custodial wallet. I see a way out here in hybrid wallets — after installation, the developer’s channels are used, but in parallel so that the channels open to the user’s wallet itself, and the funds move smoothly to this channel when it is open.

For the widespread adoption of the Lightning Network, it is important to have software that integrates the wallet with accounting. If I were a developer, I would put all my time into it. We need open software, simple, which will have a LN wallet inside it, but which at the same time will be connected to exchange rates, tracks the bitcoin price at the time of purchase and keeps a record of satoshis on a company’s balance sheet. And at the end of the month, a user must be able to form a standard statement for the accountant with the amounts at the beginning/end of the month in fiat currency and satoshis. The user should be able to do all this, taking into account regional standards.

People often complain that Bitcoin is too volatile, but this is not an obstacle to using it in accounting. Fiat currencies are also volatile, but the whole world knows how to count foreign currency in accounting. Of course, commercial billings can take on some of the problems here, but if we say that bitcoin gives independence from banks, you need to give this independence to merchants and make this decision simple and open source.

The Block: Do you plan on experimenting with other Layer 2 projects on other networks like Ethereum? What do you think about DeFi?

No. As a result of thought experiments, I came to the conclusion that none of the other cryptocurrencies make sense.

1. People are drawn to simplicity. If you put a bunch of applications under each cryptocurrency, no one will like it.
2. Bitcoin is highly distributed throughout the world, and it does not have a leader or a company running it. There is no one to put pressure on it by force.
3. It has huge hashpower, more distributed for natural reasons. It’s like super thick safe walls.
4. A large number of people and organizations keep their money in it — this is the key to the stability of the rates.
5. It has the largest and most distributed ecosystem for natural reasons (wallets, sellers, payers, ATMs, exchanges and Bitcoin brand)
6. Finally — everything that is not in bitcoin, and what may appear in other developments — everything can be added to bitcoin through consensus. Therefore, the argument that there is something in another cryptocurrency, but not in bitcoin, does not mean anything.

The Block: You said that you earn “5,000–10,000 sats per day” or “$20 a month maximum” from routing fees and that you spent more than one thousand dollars on opening and closing of the channels. Is that still the case?

Recently, I have reduced commissions to virtually zero. Earning $20 a month or zero is not a big difference. Having lowered the commission, I wanted to see how many payments go through my nodes per day in order to indirectly understand the number of payments in the Lightning Network. But having reduced and observed, I did not see a big correlation with the commissions. The number of payments was still 300–400 per day. Maybe a maximum of 600.

Of course, I can be very mistaken, but I think that a maximum of 1000–1500 payments go to the network per day. The exact number is not known if users prefer to pay merchants through a direct channel. When I wrote that I spent a thousand dollars on channels, that amount was calculated very roughly. I took the approximate number of open channels and multiplied the channel opening transaction by the average network commission. The reality is a bit more complicated. The biggest losses start when the channel closes. If you can choose a small commission for the Bitcoin network to open a channel, then closing the channel, especially if it is force closing, the commission there is determined by contractual relations at the protocol level. In addition, closure can give rise to a long chain of transactions, the commission of which is paid by the initiator of the opening of the channel.

Since I opened a lot of channels at the beginning and then decided to close unused ones, I also paid these commissions. Only very recently, I took and wrote a script that considers a comprehensive balance of funds on nodes and compared it with what it should be. Roughly speaking, I calculated the difference between how much money was contributed to the nodes and how much they really have now. The difference is, to a greater extent those commissions that were paid when opening and closing channels.

To be honest, I did not expect such a figure. To be precise, it amounted to almost Pi at the moment — 3.14 BTC (~$31,000 at the time of writing). My script also revealed channels that were closed by a remote party for violation of protocol rules by my software. In such cases, the protocol withdraws all my funds of the channel in favor of the opposite side. It was a bug in the LND, which has already been fixed. But it is difficult to identify on the side of the offender. If the developers paid attention to this problem, they could identify such cases and mark them with a certain status, but they identify only those in which we tried to deceive us, and we took the funds. This happened four times, and I lost 0.29 BTC (~$2,900).

It turns out that I spent about 2.85 BTC (~$28,100) on opening and closing channels. Unless, of course, there are some losses from bugs that have not been found yet. But I think that the figure is very similar to the costs of commissions, since force-closing of the channel can take 0.001 BTC. There is also the opening of the channel, cooperative closure, etc. Multiplying these numbers by a couple of thousand, it’s easy to get to 2.85 BTC.

The Block: You also said that, on average, 200–300 transactions (and not more than 600) are processed through all of your nodes in a day. Is that still the case? Has there been growth or decline lately?

In general, this is still the case. It is difficult to say whether there is growth or not, because I periodically experiment with commissions, and they will certainly affect the number of transactions through my nodes.

The Block: Is LN viable if routing fees don’t outperform lending rates?

At the moment, I can’t even come close to making a profit on commissions. I think it’s too early to make money in LN (if you are about launching nodes). But I am convinced that the Lightning Network has great potential and will gradually grow due to new users.

You can already make quick payments of up to $400. This is the new paradigm. At first, the market is empty and not profitable, but as time passes and LN is used more and more, those that started early on will have a huge advantage.

The Block: What do you think about Square Crypto and their interest in LN?

I don’t know anything about it.

The Block: To what extent does LN and LN operators’ success rely on the demand to spend BTC?

Who knows how things will turn out in a few years? Maybe in a few years, the Lightning Network will become one of the popular ways to pay digitally.

The Block: What do you think about the HODL movement? Do you believe bitcoin should be spent? Why is it not being spent more?

I already wrote above that I am actively paying with bitcoin, and recently, mainly paying with the Lightning Network. I am not a supporter of only HODL. This is anyone’s business. But I believe that if HODL’ers spend a bit of bitcoin and again accept it as payment, they will only win in the coming year. Because this contributes to building the Bitcoin economy, and the bitcoin exchange rate indirectly depends on it.

The Block: One of the common criticisms of LN is that it will lead to centralization to a few large nodes, which would lead to the creation of payment hubs. Do you believe this is a valid criticism?

This cannot be in principle, because the payer’s software determines which route to send the payment to. If users are afraid of large hubs, they can create demand for software that will bypass such nodes. In addition, everyone can launch at least one node, put funds on it, and participate in decentralization. But, unfortunately, if this is happening now, it is very slow. This can be seen from the network capacity statistics — it grows very slowly.

The Block: What do you think is the biggest weakness of LN?

The weakness is that few users are familiar with this network. Every day about 300-350 thousand transactions is made on the blockchain. And in the lightning network - about 1000. If at least those people who use bitcoin would transfer their transactions to the lightning network - they already made my project profitable, for example

There is no particular reason not to do this, commissions are now lower, and bandwidth is easily scalable. In addition, payments are instant. There are still problems with backup channels. But in the latest mobile wallets, they have been successfully resolved (for example “BLW” wallet).

Other problems concern node operators, but here they are not problems of mass adoption. The infrastructure of nodes already allows for orders of magnitude more payments than now. Moreover, for this, you can not increase the capacity because the funds are distilled from one end of the channel to the other, and this process does not consume bitcoins from node operators. It’s like a circulatory system, and the body is already full of blood. It only remains for him to live an active life.