Henrik Skogstrom, Founder and CEO of LN Capital, joins me to talk about the lightning network:
- Managing large lightning nodes
- Key challenges
- Monitoring
- Notifications
- Tools required
- Privacy vs. reliability
- New tech
Links:
- Twitter: @ohskogstrom
- Twitter: @LN_Capital
- Site: https://ln.capital/
MicroStrategy Bitcoin & Lightning for Corporations(code LIVERA): https://www.microstrategy.com/en/world-2023/bitcoin-for-corporations
Sponsors:
- Swan Bitcoin
- CoinKite.com(code LIVERA)
- Mempool.space
Stephan Livera links:
- Follow me on Twitter @stephanlivera
- Subscribe to the podcast
- Patreon @stephanlivera
Podcast Transcripts:
Stephan Livera 00:00:00
Henrik, welcome to the show.
Henrik Skogstrom 00:00:02
Thank you.
Stephan Livera 00:00:02
So I’ve been watching some of your threads and things you guys are working on over at LN Capital with interest and thought it would be interesting to have a chat with you. So, do you wanna just tell us a little bit about yourself and you know what? Guys are doing at LN capital.
Henrik Skogstrom 00:00:17
Yeah, sure. So, yeah, I’ve been in lightning industry since 2019 and through that I’ve worked sort of this entire network grow and as the purchaser of time, there was working with arcane and building like a payment system, and after I was done there, I realized just the lack of tooling and the what I started playing around with the routing node. For example, I saw that we had a very hard time managing. As the note grew so. So that’s where we started and started building a proper scalable node management system so.
00:00:58 Stephan Livera
And so then I guess that’s how you got started in lightning and then in terms of what you are working on now, what’s the main thing as I understand is it Torq that’s the main thing you guys are that’s the main product you. Guys have right, it’s.
Henrik Skogstrom 00:01:10
Exactly, So, we’re building Torq as a very powerful, scalable node management system for really large node operations. So, if you have multiple nodes, you have a thousand of channels and millions of dollars on your nodes, you need something that really gives you that control. That you need to run a proper organization. So, imagine doing that with just command line tools or simple Python scripts. It’s not a safe way to handle them.
Stephan Livera 00:01:36
Right. And I can recall from my earlier chat with people like 0 fee routing where he was talking through his difficulties with that. And I think from his perspective, at least at that time, he didn’t want to run a lot of extra software because he was worried that that’s introducing vulnerabilities and security flaws. So, he had a, you know, a concern from that perspective. But I think maybe if we were to zoom out a little and think about the Lightning network broadly, I’m curious to hear your thoughts as well. Because we see a lot of back and forth online about how much payment volume is going through on lightning, right, because people could argue to you and to me and now of course I’m a promoter and a supporter of lightning. And obviously as you are. But people come to us and say as an example, oh, look at the data on let’s say bit refill or coin cards, there’s not that many people paying with lightning. As an example. So perhaps the argument could be. It’s not that popular yet. I’m curious what you how you see that?
Henrik Skogstrom 00:02:34
Well, it depends. I mean, I’m not sure if that we feel is the best source to say how popular it is because that’s one use case, right? But we’ve seen that never grow steadily over the last years and especially last year, you can see that the allocation towards the big nodes, the large destinations. Exchanges, merchants and all this, that is increase. Missing so that also indicates that there is usage there. There is no point for routing nodes or small operators or the other operators to allocate capital to nodes where it’s not being used. So I think there is an increase in transaction volume and you can see that from wallets as well, like Walters, Toshi, transaction volume is increasing. Yeah, there’s been some reason why Kevin Rooke is mentioning some of them periodically as well. So, it is definitely increasing.
Stephan Livera 00:03:25
And I think that’s good to see and part of that as I’m seeing it is more integrations, right? So, we’re seeing if a big exchange turns on lightning, then maybe we start to see more volume coming from their customers who maybe they are not ready to let’s say, run their own lightning node or use some of the more non custodial lightning. Wallet applications, but they may be more willing to just use the inbuilt integration that comes with their exchange, whether it’s Bitfinex or Kraken or et cetera. And I believe Brian Armstrong from Clone Base has mentioned that they will be implementing lightning. Of course, as much as you know, you can say, well, we can, we can say our criticisms of coin base, but having them implement lightning would be, I think, a good thing as well in terms of building and growing the network and the possibilities there.
Henrik Skogstrom 00:04:19
Absolutely and I think there is going to be a wave of new exchanges, especially adding lightning in the coming year. I heard multiple places, different people are working on this, so it. I think we’re getting to the point where the technology and the main challenges are starting to get mapped out and people are starting to implement this on larger scale and this is of course where we come in as well. We’re starting to address this problem of running large node operations where until now people had built sort of this all these small tooling themselves. So, I think, yeah. We’re going to see a lot of new exchanges and places adding lightning.
Stephan Livera 00:04:54
So, let’s talk a little bit about that. So, you mentioned some of the challenges. So, can you talk us through what are some of the key challenges that let’s say I would face if I am an LSP, a lightning service provider or let’s say a routing node, what are some of the key challenges that I would face? That we would face.
Henrik Skogstrom 00:05:11
Yes, they in in some way they are similar that they’re both trying to optimize stability and reliability of payments or forward. So depending on what you do but for Nell speed, one of the main things is of course having a high success rate on your payments. So, you need to monitor and make sure that the channels are well suited. For that you need to close peers or channels with peers that are unreliable and you also need to automate changes. In rebalancing or opening close channels or the fees depending on what’s happening with your with your node. So these are some of the things where automation is going to be an increasing factor in order to sold this.
Stephan Livera 00:05:56
And so let me just give a quick explainer for people who are new. I’ll try to, let’s try to keep it accessible for people. So, for people who aren’t really familiar, the way Lightning works is we open channels. It’s a network of payment channels. That’s what lightning is, and you can think of it like it’s an Abacus. And there are beads on that Abacus. And then let’s say I open a channel with you, Henrik, and you know as. As payments go through, we’re pushing at beads on that apparatus back and forth, and as you’re mentioning the large routing node operators or LSP’s have to manage. That balance because you might have an Abacus not just with, you know, one person, but with many. You might have 100 channels. You might have hundreds of channels, or in some cases thousands of channels, and so.
Stephan Livera 00:06:38
This is then about managing the liquidity as in the balances on the sides because in some cases you may have too much of the balance on my side of the Abacus or maybe there’s too much on your side of the Abacus and then we need to do operations to sort of swap some of that in or swap some of that out. And so that’s where some of this node management software comes in. So do you want to tell us a little bit about how, firstly, we monitor what’s going on? So as we said, we might have lots of channels, but maybe they are into a bad location, right? That you know it matters where your channel. Is because in lightning you if you’re trying to earn as an example, you want to have your channels in the direction that lots of other people are going to pay towards and you want to ideally kind of be in the middle because you’re clipping the ticket as a middleman. As an example
Stephan Livera 00:07:29
Now that’s if you’re trying to earn as a router. But if you are a service provider as you said, you want to have high reliability, right? So in this case, let’s say you are operating a wallet service for somebody else and they want to have a nice payment experience. Well, then you need to be able to deliver a nice payment experience and have, let’s say, a high reliability as in a high percentage chance that that payment is going to go through. You know quickly and with reasonable fees, let’s say. So I think those are probably some of the key components that the routing node operators are thinking about. Those are some of the key challenges. So can you tell us a little bit from a? Monitoring, perspective what does that look like for a routing node operator?
Henrik Skogstrom 00:08:12
Yeah, sure So essentially a lot of the tools that has been out there, it’s like you can ask for what is the balance in each of my channels right now. But it’s an important factor as well to see like what has the history of the balance beam. So if you open the channel and so the entire liquidity is drained and then all the money is left on the. There’s the other side of the channel, for example, indefinitely, and then you need to see that as the same with if sort of. Whenever you gain a bit of inbound outbound liquidity, it’s immediately drained. So having sort of this history of how a channel has behaved and also how a node has behaved over time is an important part to have. But I think it’s important to remember that we have a lot of different companies trying to simplify lightning for the end user. Either it’s a wallet or it’s a shop or individual users directly, but at some point it stops by how Lightning Network works and somebody needs to manage the channels in the end. And this has a tendency, I think, to grow and sort of become a business of scale.
Henrik Skogstrom 00:09:18
So when you have thousands of channels you can’t go through manually and look and inspect on every single channel, you need to be able to see an aggregate of this happening. You need to be able to quickly navigate through this math of channels. So this is an area where I think we’re still seeing a lot of development in how these large companies are operating on behalf of others and that’s the. Sort of development in the LSP space, lighting service provider space, we’re just starting to see that really catch up speed now like with the light spark coming on the market.
Stephan Livera 00:09:54
Yeah, I see and just as a quick example. The idea then is that you could presumably have tools that tell you if, let’s say, a lot of your channels are getting exhausted rather than because if you have hundreds of channels manually managing that, I mean it’s possible, but it takes a lot of manual work and over time this is this needs to be automated, correct? So could you tell us a little bit about how? That kind of monitoring would work.
Henrik Skogstrom 00:10:21
So what we’re doing is we’re building an automation workflow setup. So you can actually automate anything that node can do. So right now we have we have channel data coming in. We have we have rebalancing and fee changes and you can tag channels in order to group them. But we’re adding more and more. The goal is to automate absolutely everything in. You can do so what we’re seeing is that they can look at the history of the channel automatically and have certain decision factors to this and act on that.
Henrik Skogstrom 00:10:56
So you can open channels on request or if in given scenarios and you can close them in certain scenarios all based on sort of the experience that you build up as an operator over time. And of course this will evolve even further and more automation. And this days today everything is going to be at some point so, so but still to do this you need some sort of tool to have the means to change this even with.
Stephan Livera 00:11:21
The eyes, I say. So with talk, could you tell us? What implementations it’s supporting right? There’s, there’s LED, there’s core light. I think there’s Rust, Lightning, there’s eclair. I think Electrum has an implementation. There’s different implementation So what’s currently supported with talk?
Henrik Skogstrom 00:11:38
So currently we have Ong, but we started and we pretty far along with CLN. So yeah, it’s pretty soon we’re going to release that and we’re looking to start with LDK next. So Russ lightning and we’re building more towards the higher sort of bigger operation nodes. So those are the three we have on map so far.
Stephan Livera 00:11:58
I see and can you give people an idea of the cost as well for those of you, those people who are?
Henrik Skogstrom 00:12:04
Right now Torq is free, so we’re not introducing any paid plans yet, but we’re going to soon.
Stephan Livera 00:12:10
I see Yeah, and But as you said, this is more of an enterprise level big node product as opposed to, let’s say the average, let’s say retail individual who just wants to run his lightning node. And so I guess that might be interesting as well. Could you contrast some of the different tooling that exists today? So as an example, how does talk differ from say, RTL or LD hub.
Henrik Skogstrom 00:12:32
So what we do is we collect all the data from the node in real time and make sure that all that is stored in a proper way in the time scale database or Postgres database with time scale plug in and makes it possible for us to aggregate and pull that information from any point in time and see what happened with your node. At any point in time almost instantly, you can also see that with new channels, if you have thousands of channels, those other tools, they can’t load them. Essentially, you try to refresh the page and it takes like 30 minutes. With Tori, you can see that instantly, and that’s because we build Torq to be able to handle any thousand node like.
Stephan Livera 00:13:12
That I see and one other question.Talk, I guess put it this way, is it running? Is the software running all locally on that node, or is it sort of requiring on, let’s say calling out to like a service that you are operating? You know does do you understand what I’m?
Henrik Skogstrom 00:13:26
Asking. Yeah, yeah. Yeah, sure. So yeah. Collect all data directly from the node. But we with L&D and with different implementations. There are some weaknesses. For example, with L&D it doesn’t give you. Information retrospectively about when a channel closed or opened, and there’s also some other scenarios where you can’t get certain types of information from the nodes that in that case we have a server that they reach out to you and get some additional information in order to sort of use that and fulfill it.
Stephan Livera 00:13:52
OK, Right to sort of give them more information about how to run their node as an example, and I’m also curious if you have any thoughts on this or maybe as a part of the product deals with this as you were saying, there are different kinds of nodes out there, right, that there might be some nodes who tend to be a sync, right? They tend to receive a lot of wins and then they end up being and that could be like a merchant maybe or maybe some of the swap servers maybe some of the swap servers have both ways. So is there an element there where you are categorizing nodes and saying this node tends to receive a lot? This node tends to send a lot and this is how, let’s say I as the routing node. Alright, I need to adjust for that, or at least plan my strategy around how I’m how I’m going to interact with that particular node.
Henrik Skogstrom 00:14:40
Yeah, exactly. So there are multiple ways of doing that. So of course most basic is you can tag nodes or channels, so you can tag actually node to being like sync. So you can do That, for example to loop. And every time you have a channel with them that that becomes a async channel and you can use that in the. Automation of course, as well, to act upon that in different ways. But we also have very advanced filtering system, so you can set up filtering just like you would do with SQL. Basically saying different criterias categorizing nodes. So you can either use that to tag them quickly or to drag the and in terms of uh, for example, rebalancing this I think is a very important aspect as well. Like as you have thousands of channels, there’s no longer like it becomes sort of categories of channels like they behave differently. Are you open to a shop for an exchange or a wallet or a like a looping? Oil These of course have distinct behavior. So yeah, having to that system in order to quickly label them is really important.
Stephan Livera 00:15:46
I say yeah, so being able to categorize the different kinds of nodes or channels out there as you as you mentioned. And so then yeah, it just means you have to because you might, depending on who you are as well, because you might be a merchant. Let’s say you’re a large merchant who’s running this kind of software. You yourself might be kind of like more like a a sink, let’s say, because lots of coins are coming your way and every now and again you just need to swap out. Let’s say out from lightning out to take that money on chain, let’s say so that you can, you know, do something with it or. You know, although I guess as more exchanges start to support lightning, well then you may actually be able to receive on lighting and then sell for Fiat on an exchange. If you’re a merchant who needs Fiat, let’s say to pay the bills. So I guess it is a constantly shifting game in that way. Whereas maybe in the earlier years of lightning you had to do that. Whereas now maybe not as much. We’ll see.
Henrik Skogstrom 00:16:37
Yeah, that’s one of the interesting part is the with the industry as well, it’s evolving quite quickly. New additions and new possibilities are adding added constantly. So it is having a tool that’s a bit dynamic to this and constantly being sort of on the learning side of it is important. But as I said with merchants as well, they often end up being either a main sort of sink or a source. So that creates its own sort of problems and how we deal with that right now we’re seeing this increase in service providers selling liquidity or trying to offer that in in different ways and I think for a lot of merchants this increase in services is something that’s really going to drive adoption.
Stephan Livera 00:17:18
As well I see and so there are different kinds of LSP’s as well. So, I guess some LSP’s may exist as like a wallet back end, let’s say, and maybe others are operating as a swap. So, the so they are helping people swap off chain to on chain like as an example bolts dot exchange I think is 1 good example with the swap as a swap provider and then there are yeah, just different kinds of services that exist in lightning. So, could you talk a little bit to that or elaborate on the different kinds of services?
Henrik Skogstrom 00:17:52
Yeah, like this landscape is just growing in complexity and we I think also we’re seeing the sort of line between routing nodes and those pieces starting to blur a bit where a lot of the larger routing nodes are looking into becoming like liquidity providers, service providers in that sense. and yeah, I mean, there are some who just focus on selling on demand liquidity and some are like. It feels like all this right now is just a bit unclear. You have the payment processors who are leading into the ISP space as well depending on like how much of an UI do they have. So it is still in instead of quick development there, but I Think the commonality among all them is that they’re focusing on making it easier for people to just use lightning and not care about the entire complexity. And they try to hide that layer from and of course, that involves different things for different users of that those products. So I think natural, I think we’re just going to see in increase in the vocabulary around what LSP is like LSD X.
Stephan Livera 00:19:00
Why and so yeah, maybe it’ll be becoming more specific over time, but and now people are sort of being a bit general in a way. And I think that can happen where let’s say, a particular company is doing multiple roles. So, a quick example would be coin corner or Bitfinex where they are both an exchange and a merchant services provider. So they sort of also provide that sort of service of helping the merchant take lightning payment and then swap it for Fiat or on chain if they want Bitcoin on Jane. And then if you have other, you know people like, I think probably Ebix, Mercado is another example because they have X pay and then ebex Mercado and then maybe someone like open node is more. They’re probably known more for the payment side of it. They’re not really an exchange per say, so they just deal more with, like helping merchants take the payment and then receive either Fiat or, you know, Bitcoin on.
Stephan Livera 00:19:48
I mean, so I guess those are some of the different examples that are out there today. But as you said, it’s constantly evolving and changing and we may see more individuals or maybe large routing nodes to just have their own reason. Maybe they just want to be a routing node, or maybe they are selling channels. That’s like another whole concept as well, right? Because I know bit refill cell channels, LMB cells, channels and even LMD or Lightning Labs has their pool service as well, so you might be a channel sell seller.
Henrik Skogstrom 00:20:22
Exactly and that again is into the Lightning service provider space. I think some routing nodes are going to notice the difference when they enter that like service provider. They have these obligations to keep these channels open. They the customers rely on that inbound liquidity, for example, to run their business and it turns that sort of routing node from being almost like a fund to invest in sort of their own. Portfolio of channels into being somebody who like they have a service agreement in a way they have to give that that liquidity. So I think that is also something where the routing nodes doing that they’re entering much more into professional operation much quicker than they maybe.
Stephan Livera 00:21:04
And is that a service or like maybe there already is something in that for in talk in terms of your. So let’s say an LSP wants to use talk as how they manage their channels. They might then need to tag certain channels are saying no, that’s the channel I’ve sold. I can’t just close that I have to provide, let’s say one month or two months of guaranteed uptime on that Channel. Therefore don’t close this channel down and don’t include it in the automations that may automatically close the channel if it’s not.
Henrik Skogstrom 00:21:31
Yeah, exactly. So this again is where we can use tags. We’re going to expand this even further. But what you can say is that you can open the channel, then immediately tag it as being a channel you sold. You can give it, for example A tag that means that it listens for being open for X amount of months or days or whatever before it’s being. So we’re going to add more like automatic closing or channels and that can of course use these criterias, but also having that interaction, the features that we’re focusing now on are more towards the LSP space where we increase exactly what to explain there. So you can automate these types of interactions.
Stephan Livera 00:22:14
I see. And I also know that at the larger lightning node level there are sometimes people who are actually running multiple lightning nodes as part of their, you know, their overall operation. They may have one at the front and some sort of shadow nodes in the background. I’m curious how talk is going to deal with that and how you’re thinking about that.
Henrik Skogstrom 00:22:36
Yeah, exactly and I think that’s yeah, it’s necessary for reliability reasons. So Torq is built to handle multiple nodes. So you can add multiple today and you will see all channels in one. You can of course filter and see just data for each of the different nodes, but it’s yeah, it’s inherently built in to be. Serving that enterprise market that needs at least two nodes, often for uptime, reliability and as. We see with the river, for example, they have for like 2 public and two private nodes. So it’s really important feature for us.
Stephan Livera 00:23:11
And as I understand, the reason part of the reason for that is that if they need to do up upgrades or things like this, they sort of need to be able to have one up and the other down while they’re doing upgrades on this and still keep the availability going as Lightning network professionalize over time. They become, you know, it needs a certain level of reliability. They can’t just sort of say, oh, we’re down for half an hour, you know? Well, no, there’s payments that need to. So that’s one example. Another area that’s perhaps interesting is notifications as well, so I think that’s something you’re looking at in terms of the routing node operator being able to receive notifications when things go wrong or things go down. Obviously, if you’re a professional routing node operator or even 70 pro, you may need that.
Henrik Skogstrom 00:23:52
Yeah, exactly so we had we had for some time have had this feature where you can get notified of course, if your node is down, but also if the blockchain is out of sync on your node, or if the gossip of the node in the light network is out of sync as well, and we just now added the feature to have that connected directly from your Torq instance to your private telegram bot or your slack channel as well. So yeah, having that notification. Immediately, it’s important we’re also going to boost that more with the different types of alerts that you need, again through automation, so you can actually build workflows that says like if this channel closes then and send a message on Slack and warn us about.
Stephan Livera 00:24:35
That I say So then the node operator can quickly get a notification and go do some operation and try to fix things and.
Henrik Skogstrom 00:24:44
And there’s cases where you can’t do anything automatic, you just need to see like, oh, this edge case happen. OK, we need to deal with it and if we try to define all those edge cases, the rules it’s never gonna work. So that’s why we again we build it into this workflow where they. Can define this themselves.
Stephan Livera 00:25:00
And when it comes to operating a routing node, there may be those individuals who are trying to operate it profitably and so they want to understand what their costs are because they may be thinking, well, OK, how much is it going to cost me to do a rebalance or how much is it going to cost me to do a channel operation to close the channel and reopen? How much is it going to cost me for swap in or out. So could you talk to us a little bit about monitoring the costs of a routing node for that operator? And if talk is offering anything around that?
Henrik Skogstrom 00:25:30
Yes, this is actually where we started. We focused on writing notes. Beginning and we have a lot of historic metrics. They can see how much have they spent on different cost throughout a given time period. So you can select the month or whatever days you want, so you can see then sort of rebalancing costs the open and close cost so that for now we’re going to increase this as well. We haven’t looked into sort of prediction of cost sets, but I think it’s something where as the routing. Operator opening and closing cost prediction is something that you check the manual and it’s not necessarily that easy to predict.
Stephan Livera 00:26:11
I mean, that’s this is inherently going to be hard because of the temple shifting as well, right? Like even with all this crazy, you know, the ordinals inscriptions. Stuff which of course I discourage, but it’s there. But it seems to wax and wane, so it seems you know the Mempool cleared again recently, whereas people were betting on whether it would not clear. And of course you know, I think many of us have hypothesized that we won’t see real a lot of lightning adoption until we see a sustained feed market or block space market.
Although that said, even without the fee, even in a low fee environment, it is still a better UX when you’re paying online to pay with lightning, or especially in person if you’re paying in person. Lightning is just so much better I think there is an incentive there, but I think the real push for lightning will come when we see high fees.
Henrik Skogstrom 00:26:59
Yeah, it might be I think we have seen sort of a steady increase in capacity and I think that’s sort of a steady increase of use as well. Of course, when we see this huge increase in in fee costs, think that might push more providers to support non and might be why there is more providers working on it right now. But of course, in the next hype cycle, the fees might shoot through the roof. And then, like everybody just has to go on like. But yeah, I just think that the user experience is so much better on lightning anyway, so just happening with time.
Stephan Livera 00:27:36
Anyway, I see. Yeah, I agree with you but I think it’s probably fair to say my expectation for the last cycle was to see more lightning because of the I thought the fees would go higher basically right? Like I, you know, I was probably like many of us saying. Yeah, I think you know, at that time in maybe 2019 and 2020 that you know the kind of next cycle would see a lot more lightning use where I would say comparatively or just relatively it was a bit less than what I was expecting because and maybe for that for different reasons maybe more people were using custodial things or stable coins as opposed to using. Kind of lightning as I was anticipating but you know, every now and again we. Get humbled, right? Yeah, yeah.
Henrik Skogstrom 00:28:17
But also think that you gotta remember the state that lightning was in at the time. It was hard to do transactions. Tried to buy a $5 postage for stickers and it’s stock transactions again. At that point it was very, very early in like 2019. I mean, and I think the development that has happened between now and then is a huge factor to making it a reality and making we got to remember that lightning. Skate Bitcoin, but we still need to scale lightning right for the companies operating in. You see it So yeah, I think that’s the difference going to be difference between then and the next.
Stephan Livera 00:28:54
Cycle I say and I think you’re right. The we need integrations to be there as well for the benefit to come and if exchanges aren’t playing ball then it’s hard to get the benefit out of it. Now of course, there are some that are very forward. So, for example, a bit for next as an example, because they were very early with lightning of course, they were kind of plugged in and ready to go, but maybe other providers weren’t so ready now. Now we are starting to see more, of course we have, you know, Coin Corner, River Strike, Cash app, you know, lightning ready and so, we’re seeing that so that’s a good thing to see. And so maybe next cycle we will start to see more actual lightning use in that because people might already have integration. So, I think the other aspect of it is that people just you tend to just use wallets. That they already have they’re not as inclined to shift over to a new one, so maybe there’s an argument there that let’s, as an example, you might be less inclined to spend lightning unless you’re already earning over lightning also because. That’s, you know, it’s just it just a lot of people would just stick with. The wall they already have.
Henrik Skogstrom 00:29:55
Yeah, I think of course some of the major wall providers adding lightning is benefit. But yeah, just like you said, we we’re seeing sort of some of these exchanges starting to integrate it. We have it for next, we have truck. We have these OK X and such, but river had Alex from river, had a talk on advancing Bitcoin about. How they operate out of the node with four different nodes et cetera and like that, that is sort of a topic that is still new is sort of telling you bit about where the industry was a few years ago. So, I think just that makes a huge difference that this is becoming more people, more knowledgeable about how to operate this.
Stephan Livera 00:30:37
Also curious your view around centralization in lightning. So this has been a common criticism, this idea that it’ll be a hub and spoke model or it’ll be very centralized into very few providers and will that be some kind of floor in the Lightning network. I’m curious what you think or how you would respond to that kind of criticism?
Henrik Skogstrom 00:30:58
I mean, there’s nothing sort of in the centralization that blocks others from participating in the network. So that’s one thing the other is that you have a ton of large businesses and exchanges and walls around the world who will operate large nodes and these will be quite. Decentralized in in that term but if you look at the liquidity, so a lot of people are concerned about liquidity and animation or capacity centralization so you can see that some of the biggest nodes, I think if you take the very top nodes like 2% of the entire network holds 85% at least of the capacity. This is actually much less of a problem that you might think because as an individual user. You don’t have $20,000 on your node. It doesn’t make sense at all, but for a operated, an exchange or whatever, like having spent $20,000 on you know it is. It’s tiny you might have. Right? So of course the liquidity distribution there is going to be seem insane and seem very centralized. But as a user and as a shop, and whoever needs it to be truly, truly decentralized, they can still use it and it works perfectly. We have enough routing nodes everywhere and there is businesses around the world with sufficiently. Different jurisdictions for this not to be a problem. So if you see a band like the like mining band in China or something, these nodes and lot of businesses operating them. And going to move So yeah, I don’t see it as a big problem, yeah.
Stephan Livera 00:32:28
And I think that’s totally fair, to be honest. I think it’s more just a, it’s a common criticism that I see and just, you know, wanted to get your perspective there also in terms of new lighting technology, is there anything you’re particularly excited to see come?
Henrik Skogstrom 00:32:46
I mean, splicing is really exciting. I think that gives a lot of options in order to handle liquidity for these large operators. And also of course improved handling of opening and closing multiple channels and collaboration among channel opening, closing and there is a lot happening and as as easy we write a lot of threads about future technology and lightning and explaining different concepts simply. And I think it’s like it’s really hard to keep up. So yeah I think splicing is one of the coolest things on the on the horizon.
Stephan Livera 00:33:17
Right and so for listeners who aren’t. You know my understanding of the splicing is that instead of, let’s say closing and reopening a channel, we can do one on chain transaction that resizes the channel. So let’s say, you know, say I’m operating a node and I have a channel with you. Henrik and I need to resize it larger. I could have a splice transaction that does that and then I guess the real power also could be where? To be operations could be batched and maybe that’s where the. Well, sort of big benefits start to come because let’s say you are running a big node and you want to, you know, do a lot of operations at once. Maybe there’s more potential for that also so do you want to just explain a little bit about what that future could look like?
Henrik Skogstrom 00:34:00
I mean, it just gives more optionality. Say like opening multiple channels, distributing channel capacity, or aggregating it back again. So, it’s just that the tool set is increasing So, I think that’s the most important part of but yeah, it needs to be built into each other. The implement implementations on the Yeah SDK and all this and I think that’s one of the areas where I would wish to see that we had even more resources that these like everybody wants the technology today right now, right. But of course this is financial. Infrastructure we’re building and it takes time, and it is something you need to be careful with.
Stephan Livera 00:34:44
And that actually that reminds me, I need to get dusty on and talk with him about splicing. I know he’s been working on that over at in the block stream core Lightning team. He’s been working on splicing a lot and I guess it’s probably fair to say that the core Lightning team are probably furthest ahead on the splicing technology. So that would be interesting. I’m curious if you have any thoughts on peer swap. That’s something also coming out of Blockstream, which is an interesting technique and as I understand you, you might have a node and instead of put it this way, instead of using LND or lightning. Tabs are pool server you are becoming your own swap server and so these different nodes could peer swap with each other and say as an example you know maybe our channel is a little bit lopsided. Let’s say I have a channel with you, Henrik, and it’s lopsided to maybe your side, and so then maybe what I would do is. You know, I could pay you some on lightning and then you could pay me on chain and we could balance it in that way and so that’s an interesting concept and it’s quite decentralized, so would actually be a pretty big benefit. I’m curious if you have any thoughts on that and maybe if there’s anything you’re planning to. Build out for that also
Henrik Skogstrom 00:35:56
So we don’t have any specific plans just yet, but it is like looping and sort of transporting liquidity, other places without rebalancing sort of circular way is something that. I think with peer swap I’m not sure where the protocol is at now, but it needs to be quite efficient. Instead of pooling transactions, etcetera which essentially loop is doing now when they close channel UM or they do loops out, they often batch them together in in larger transactions and I think for something like peers to work we need to have sort of that handling as well. So I think this is something that we’re going to see development in together with the LSP space and is going to be part of the LSP, LSP space as well. So I think what? We’re watching this closely and we’re going to to build out the services and the part of the software where these operators need solutions the most, but I think it’s going to be really interesting to see how this parallel this protocols are developing that are not part of the core Lightning sort of the main lightning. I think this is a really crucial point in in sort of the history of lightning. Where to what degree do you define these protocols or to what degree do companies are define their own thing and we might end up with a lot of solutions like we have a chat today where there’s no, there’s no protocol for it, it could be, but there’s not. So we have all these ton of different chat protocols So yeah, I hope that we see more of these general protocols in lightning, get acceptance.
Stephan Livera 00:37:24
Yeah, you’re right and by the way, I accidentally I think I said the wrong one. I said pool, actually like I should have said like as you said, what we were talking about is like, yeah, is, you know, you said it correctly I.
Henrik Skogstrom 00:37:30
Luke. Yeah, yeah.
Stephan Livera 00:37:35
Said it wrong also curious if you have any thoughts on Music 2 and taproot.
Henrik Skogstrom 00:37:40
Is that going to help or change anything? Yeah, I think so. I got to be honest that I have no time to look deeply into. Yet Jack on our team is running some excellent stuff on that, but yeah, not sure if I have too much to add on it should check.
Stephan Livera 00:37:55
Out the threads, right? And as I understand it could help in terms of. Today, when you open and close lightning channels, you can see or I think specifically when you close the Lightning channel, you can see that it’s a two of two multi sync on Jane, but in a music two taproot world. It all looks like a single signature standard transaction in the cooperative closed case. Now, in the uncooperative case. Then it you, you have to reveal some certain things about the transaction, but I think that that could be an interesting privacy, or at least one aspect of privacy. It’s not. It’s not the be all end all of privacy So that’s an interesting area also on the privacy question, I think. Maybe there’s a bit of a tension here because I’m starting to notice some of this in the discussion. And there are people in the community, and perhaps rightly so, they’re saying we should care more about reliability than privacy. And then there are the privacy guys who are saying no. Oh, that’s so bad you’re doing all the stuff that’s going to result in surveillance and decreased privacy. It sort of comes to that question of what is the Lightning network’s purpose, right? Is it there to make reliable? And cheaper payments? or is it also a, you know, a point or a purpose that it should be making transactions more private? I’m curious if you have any thoughts to add there.
Henrik Skogstrom 00:39:09
I mean, privacy is very hard to build into it retrospectively, so it’s like we should focus a lot on the privacy aspect. Some of this the new user of the Redmond in like gossip V2 eventually and yeah tap root opening channels et cetera. It’s really important and of course it takes time and I think the commercial interests are pushing towards adoption of technology as fast as possible, which have some side effects. So myself, for example, not the biggest fan of Lu URL compared to actually building Vault 12 and I think that again, it’s just this constant battle of somebody says we want to use it immediately, which I also support, right. I want people to use lighting immediately, but that constant battle with privacy and then later on as well, we’re going to see that privacy battle Ray, John and everybody wants privacy for themselves but then. Regulatory aspects or for society they some sort of the bad guys to be caught. So it’s yeah, it’s going to.
Stephan Livera 00:40:14
Be a war, and it’s also fair to say that sometimes what people say and what they do is very different, right? So there may be people who say they want privacy, but when it comes to when the rubber meets the road, are they willing to pay the extra price or have a less convenient experience? Maybe not and so they’ll say they want privacy, but actually in practice. They may not care that much, and so then it sort of goes to that question of if there are multiple competing providers or services or tools in the space, the ones that maybe don’t care as much about privacy will might win out just because people say, you know, if people say they care about privacy but don’t really. Then you know it just kind of leans a certain way and I think we’re seeing.
Henrik Skogstrom 00:40:53
So, I think we need to focus on that when we develop.
Stephan Livera 00:41:29
The core protocol, yeah, I say. But it does come to that question of what price are people willing to pay so for? Sample are people willing to go Torq only because that can have reliability concerns and you know, people like Warren from Blockstream have been speaking about how with all these Torq only node operators out there, it’s slowing down the overall network and that people are trying to make payments and it’s, you know, it’s slowing it down and there are, you know that’s where maybe that tension is coming because some people are saying. No, look, privacy is nice to have, but actually what’s most important is the reliability and being able to, you know, quickly send a payment. And if I’m standing there in the shop and I’m trying to make this lightning payment, and because of everyone doing Torq and you know, it’s going really slow and I can’t, I can’t get the payment through, you know, you can sort of understand the concern from their side also that if it’s not reliable people.
Henrik Skogstrom 00:42:17
Won’t use it as well, but also like more people should run turn out. It’s like a Torq this offering of relaying Torqmessages, right? So yeah, it is in every aspect there’s a convenience, privacy, trade off all the way. And yeah, we just need more people to try to do more good. But also I believe in building in defaults that enhances people’s privacy and benefit sort of altruistic mindset in building the protocol. It’s I think it’s the only way to solve this, like commercial interests are always going to push for the cheapest, fastest, most beneficial to them. So I guess the sort of the world of, yeah, protocol developers aren’t necessarily motivated by money in the same way and have that ability to push the industry towards.
Stephan Livera 00:43:04
The right direction, I say and that’s a fair point and as you say the corporate providers for lightning have an incentive to be less private, right? So they have an incentive so as an example, you may be a wallet provider who is trying to deliver high reliability and as part of that you may do more probing of the network and so then maybe there’s an argument there that, oh, you’re making the network less private if you’re doing all. Probing but at the same time, as you said, there could be new technology that comes, whether it’s a gossip V2 or something else that lowers the possibility or decreases the possibility for using probing and doing probing. So there’s kind of all kinds of arguments there, but it’s just, there are so many moving pieces here, it’s hard to, I guess concretely speak about it, but I guess we trying to explain a little bit about some of the contours of this discussion of what’s possible and what’s maybe unrealistic. So on the privacy conversation, I’m sorry, did you? Have something to.
Henrik Skogstrom 00:44:03
Add again, I think it’s just hoping that people do good is in the business sense. No, It’s just not I think it’s not going to pan out because we need to build technology that is resilient.
Stephan Livera 00:44:15
And in on that theme as well, government regulation, right? There have been concerns raised about this kind of thing, saying look, if a lot of people are building out custodial service. Is that opening the door for more government? I don’t know your business is in Norway, right? Yeah, or I am right, I see Yeah, but as an example I know the EU or I think Norway is not in the EU, but it’s in the EEA or something. Yeah, it’s a bit complicated I sometimes struggle to understand the exactly all that, but I understand in the EU there’s a lot of. Regulation coming, things like Micah and you know EU crypto asset sort of rules that could apply or you know, do you see what kind of risks do you see there from government regulation, whether that’s in the EU or let’s say in Norway?
Henrik Skogstrom 00:45:03
I mean, it’s a very interesting topic. There’s going to be some crazy attempts at regulation of technology that they don’t understand. It’s going to take time before they understand it. I mean, for people who hold assets or money for other people and who transact. On behalf of other people, they’re going to be. Subject to the same regulation that every other in this like payment industry participant is going to be, but the border sort of it gets muddied and difficult. Of course, when you enter into self custody or sort of these half ways of custody and especially routing nodes, routing nodes. Actually, they don’t know anything about who they’re forwarding for, and this is really interesting. Spain’s regulatory in terms of regulation. And yeah, I think it’s just going to take a few years until there’s clearly on it. And I think basically the conclusion is going to be that it’s not possible to really regulate A routing node in that sense so or social nodes.
Stephan Livera 00:46:06
Either interesting and so then, yeah, because I’ve seen different arguments made that you know it’s better if it’s more self custody because. It’s harder to regulate that component of it, whereas if you are a custodial provider, maybe there’s more regulations that could apply and hit you, whether that’s on the XAML side or something else. So, do you see any possible responses? Like should the Community be out there trying to advocate? Should people be building tools? What kinds of things do you see as helping on that particular front?
Henrik Skogstrom 00:46:36
I think that there is sort of a natural balance in the development there because there is a lot of people concerned about this, that are building self custody setups and solutions and making it easier and easier for people to opt out in that sense and you got to remember. But if you don’t need it as a right now, it might be OK to use a custodial wallet or exchange partner, et cetera. But if you have an efficient way of opening up and jumping into alternative, the alternative self custody, that’s really important factor because then you get. A lot of the efficiency benefits, but still don’t. Not locking yourself out the worst case scenario is if you’re building towards a world where everything is custodial and everything. Just because convenience, everything ends up in only custodial services, then we’re in a really bad spot because then we no longer have the alternative for the case where the world turns bad. So, in Norway, for example, we’re fortunate to have a very stable, predictable. Government, we have a lot of trust, very low corruption, all this. So the custodial sort of system financial system for us works really, really well, which is also why we see lower lightning adoption, of course, and Bitcoin adoption in Norway and users. But yeah, I mean it’s not likely to change quickly in Norway, but a lot of other countries. To very quickly change So yeah, we need to account for both and build. Towards the world where we have.
Stephan Livera 00:48:01
Self custody and the other angle. I’m curious to hear your thoughts on is around stable coins, right, so. Has been. Now I know there was a legal battle about Taro.So Lightning labs have. I don’t know if it’s still called Taro or not. I’m just gonna call it Taro for now. I know there was a legal battle between them and Tari Labs, which is like some Monero L2, shitcoin, whatever thing, but basically there’s this question and in fairness, mentioned, Taro also RGB is out there. Do you have any thoughts on whether? If they are. If bringing quote unquote bringing stable coins to lightning and I will note here that some people challenge that notion, whether it’s quote unquote on lightning, is it just you know. But anyway let’s just put that aside. Does bringing stable coins to lightning create unwanted attention? Government risk, or do you see that like it’s an opportunity because there may be more users of lightning and that’s going to grow our network.
Henrik Skogstrom 00:48:55
Yeah, I think it’s just, it’s positive. It brings more opportunities as a more, more use of it. And whenever we see. When we get to a certain level where the number of users and the we’re just going to bring enough attention to this at some point anyway that they’re going to be. This sort of scramble for regulation and I mean it happens to anything like AI now is going through this massive like people screaming in regulation and one sort of something limited because it’s scary, right? It’s something new that changes society around us, so I don’t see sort of the RGB and stablecoin on lightning as something that. Sort of makes that person anyway. Right?
Stephan Livera 00:49:35
OK, fair enough. OK So, yeah, I think we’ve spoken through a lot of things, but essentially the Lightning network is growing. People need tools to manage that. Whether you know if you are an LSP or a routing node in particular. Obviously if you are just more like a retail or you know individual user, maybe you are. Just using, you know, apps that deal with the complexity for you. So, do you have any closing thoughts? And for listeners on Lightning network and where you think things are?
Henrik Skogstrom 00:50:05
So, if you are an exchange or wallet prior who doesn’t have lightning yet, you need to start experimenting and learning about the network at least. So yeah, I just recommend everybody to start running a node if you are someone interested in lightning, just to get that exposure to technology.
Stephan Livera 00:50:49
Fantastic! Well, listeners, I’ll put the links in the show notes, so it’s http://www.Ln.capital and the Twitter account is @LN_Capital. So links will be in the show notes Henrick. Thank you for joining me today.
Henrik Skogstrom 00:50:57
Thank you for having me.