Economic Eras of Bitcoin, and Lightning Development,

Welcome to the Stephan Livera podcast.

Alright guys, welcome to the show, and my guest today is Rusty. Welcome, Rusty.

Welcome. Thank you. Thank you. Happy to be here.

Yeah, I've heard a lot about you, Rusty. So just to quickly introduce you to the listeners, Rusty is a very well-known and respected developer within Bitcoin and Lightning, and he previously came from a background of Linux kernel development. So he currently works at Blockstream as part of the Element project on the Blockstream Lightning implementation, known as C-Lightning. And on top of that, he's also written some great articles and given some really cool talks on Bitcoin development, such as his talk, "Future..." Technological directions in Bitcoin. but yeah, the, the thing I was hoping to start with today was your article, The Three Economic Eras of Bitcoin. So, in this article, you basically say the way the Bitcoin ecosystem will play out is written in the mathematics of its consensus rules. We should all know the three phases it will go through. Now, I thought this was a really prescient article, and you showed a lot of foresight. So, can you give us a bit of a background on this one, Rusty?

Yeah, sure. I think that came from with the, the events leading up to what became the big block size debate, and that was really that, you know, w-we knew this transition to fees was gonna happen, right? This was always written there, right? You know, mining rewards, the free money phase was gonna end, and we were gonna enter into this, you know, you know, the free offer is over, you know, you have to pay for your Bitcoin now. And I, I think it was, it was just, it was, it was kind of glossed over and grossly unappreciated in the early days. Partially, I think, 'cause it was so far away, but also because there's a kind of boosterism, you know, like, you know, "He with us or against us, it was, you know, this is the greatest thing since sliced bread," and, you know, any kind of message that, that, that contained caution or, you know, like "there be dragons" or, you know, "things ahead" It, it, it led to this kind of, this, this case where people were like, "But, but, but I thought Bitcoin was, you know, instant and free and everything else." I was like, "Well, whoa, whoa, whoa, you know, that, that, it was never those things, right? Bitcoin was free when nobody wanted it." Satoshi made it very clear in the paper that the plan was to transition across to a fee basis, supported by inflation to begin with, but that was supposed to taper out, right? So, you know, when you have people going, "But, but hold on, you know, fees are too high and everything like that," you go, "Look, you know, they may be higher than, you know, you may prefer at the moment, but..." We were always gonna have this at some point, right? And the argument is, you know, well, sh- it shouldn't be now, it should be later. But with no centralized control to, to put their hand on the lever, it's gonna happen when it's gonna happen. And so people who are- Horrified by what Bitcoin had become. I, I was amazed that these people had, you know, like, like, you know, how did they get this impression that Bitcoin was gonna be like, you know, this, this, this free money system the whole time? And so, for that, that started me thinking, right, okay, so it was the first phase where basically, you know, nobody wanted Bitcoin, it was, it was completely free, you could just send as much as you wanted around the world, there was no fees, it was all great, right?

You know, the Satoshi's free money phase. but, you know, it was quite clear that that was gonna end. I think we've seen this with, with, you know, any service that starts free and then you try to charge for it, right? There's gonna be a certain number of users who go, "But I thought it was always gonna be free." and in Bitcoin's case, it was very clear that this was gonna end. and seeing, seeing people's reactions to that, you know, that transition, made me realize that, you know, we're, we're very early on in that change, right? You got a little taste of fees, right? but that's gonna be the norm if Bitcoin is to survive and become self-sustaining. The mining fees, the miners who secure the network, are gonna be paid by you and me on a transaction basis, right? Every transaction's gonna have to chip in something. Now, ideally, you know, we'll get as many transactions to, to burden as possible, but, you know, it's never gonna be enough. You do the math and you go, "Well, what if you have one cent transaction fees and we wanna like give the miners, pay the miners a billion dollars a year?" Well, you know, that It's gonna be huge, unattainable blocks. So it's, it's gonna be somewhere in the middle, right? Fees are gonna be, you know, higher than you would hope, I'd bet, you know, as low as, as low as they can be because we cram as many transactions in as we can. So that's, that's, that's where we're headed, and we're not there yet. So, you know, there is gonna be like this, this third phase, hence the, the three economic eras. at one point, it's gonna become Inevitable that on-chain is gonna cost, you know, real money, and it's, it, it, you're not gonna have these, you know, e-even when it was busy, you tend to get like, things would lull on the weekends, for example, right? So you could get cheaper transactions through on the weekends because, you know, traffic would drop again. you know, at some stage people will have optimized for all these tricks and we'll get a pretty steady flow, and it will keep fees high, and that has to happen because, you know, that, that's- Just paying for the network. and given people's reaction so far and their, their, their, how incredulous they were that this was the way Bitcoin worked, I don't think that message has really gotten out there. So I do anticipate that there will be a push as that becomes a reality for these businesses who've built their whole business on this idea that Bitcoin is free. the transactions are, are, are trivially small. they will push back against the idea, and they will, you know, I, I think the obvious thing for them to do is to advocate for a hard fork, which just continues to inflate Bitcoin.

Yeah, sure, sure.

you know, hey, why, why should the people using Bitcoin pay for it? Why shouldn't the people just holding Bitcoin pay for it, right? That's the, that's the inflation argument. You know, and so, you know, I think it's important to kind of make that Clear so people aren't blindsided by it. There's gonna be a very, very strong push to, to break the twenty-one million Bitcoin cap, and to continue to pay the miners. The miners will obviously be in favor of it because free money, right? and, you know, the, the big businesses who are transacting on the network You know, just, you know, the accountants there will go, "Hey, we are paying, you know, we're having to charge our customers, you know, what half a billion dollars a year?" you know, we could Pass those costs on to the rest of the network and not have to pay it ourselves, right? So they will definitely advocate for it. you know, and I think, you know, there's rational economic reasons to argue one way or the other, but it doesn't really matter because I think it's pretty clear that Bitcoin was defined, you know, and people bought into Bitcoin on this idea that there was a twenty-one million dollar limit. So I, I think there's gonna be extremely strong pushback, but that doesn't mean that people won't try. So I certainly expect this, this third phase where, as I said, I think in the article, you know, the third phase is gonna begin with civil war There is going to be this massive push, by interested parties, particularly those who f-built their business on cheap transactions, to keep them cheap and, you know, use inflation to-

Yeah, sure. And I think just to sort of keep it accessible for the sort of newbies, what we'll do is just kind of explain a little bit around how Bitcoin's structure is. could you just outline a little bit around what, what is that pressure that drives the fees up?

Right. Okay. Well, that's, that's, that's actually not such a, such a sort of a beginner question really. So sort of from a high level, with, you know, the, the miners are rewarded by minting new coins, right? And every four years that halve. So it started at fifty Bitcoin, went to twenty five, it's now twelve and a half, and in another two years, it'll drop to six point two five. The other money that, that miners get is from Bitcoin fees, right? Every transaction pays a fee. Now, because there's a limited amount of transactions they can cram into each block, obviously miners choose the one that pay the most fees per weight, right? So given the size of the transaction, you know, they, they take the ones that'll basically give them the most money. and that means when there is pressure, for, you know, there's, there's, when there's more transactions that will fit in a block, that leads to this kind of bidding war where, you know, those who pay most will get them Their, their transactions mined first, you know, they'll be processed first. now, particularly when we went through the last fee spike, when, you know, transaction volume exceeded the block capacity, and that's before we had the block size increase that was SegWit. We saw some crazy, you know, it peaked at something, you know, average transactions paying things like fifty-five US and stuff like that. That was actually pretty unusual, but we did see some spikes, and a part of that was the infrastructure wasn't there. People didn't know it couldn't deal with it, right? So you had websites basically saying, "This is the transaction fee. If you want your money, that's what you're gonna have to pay." they didn't give you an option, hey, you know, if, if you're prepared to wait a little bit, perhaps we could try a lower fee, you know? it was very, very primitive because there had never been this crisis before. So as a result, fees spiked. I mean, imagine, you know, you've got sixty bucks in this website, it says, you know, do you want your money? It's gonna cost fifty-five dollars to get out. You complain But it doesn't present you with any other options. So damn it, that's what you pay, right? And we saw people doing that, they were complaining about transaction fees, but they were still paying it. you know, things have gotten a lot better with stuff like that, you know, with, with replaced by fee, so that you can, you know, once again, like you used to be able to in the original Bitcoin days, send a transaction And then replace it, you know, if you go, "Okay, well, I, I kind of lowballed that fee, I've been waiting half an hour, still not through, let's, let's replace it with a, you know, a slightly more generous one." which gives you a lot more ability to control your fees, and, and software's getting more sophisticated as well. So, you know, next time we have a fee spike, we're hopefully in a much better place to deal with this, and it won't be so-- You know, there'll still But for the rest of us, I think, you know, having more sophisticated software will definitely help with this sort of bidding process.

sure, sure. How about-- Sorry, go on.

No, no, please.

Yeah, I was just gonna say, related to that, could you talk, could you talk a little bit about how fee estimation was very basic in the past and how that's gotten better?

Right. So, so fee estimation, I would argue first version is still pretty basic. You're trying to guess, so the problem is blocks don't come in every ten minutes, right? They come in, you know, approximately, you know, average every ten minutes, but, you know, it-- there's some probability that it'll take forty minutes before someone gets lucky, right? Or it could come in really fast. So you can tell just from that, it's almost impossible to guess how much you would have to pay to get in, you know, the, the, the, you know, some like the second block or something. You don't know how long it'll be, you don't know, you know, how, how much things will pile up, right? you're kind of trying to get in there so you're, you know- So you're in that first chunk of, of transactions, the best transactions. You don't need to be the best, you just need to be good enough that the miner will go, "Great, I'll, I can include that in your block." So you're trying to hit that cutoff. and over time, more people come in with more transactions, and if it's been a while between blocks, you know, that's gonna increase, people are gonna go, "Well, I really, really want this in," and they'll start bidding higher and higher. so it is With software, but more importantly, the ability to replace things by fee means you can start low, you can guess conservatively, go, "Oh, you know, I'm, I'm gonna guess and hope this slides in, and then I can replace it later if I have to," right? and that was missing, In a lot of software, last time we had the fee spike, so you kind of went, "I don't wanna get stuck with this transaction that never confirmed," so damn it, I'm gonna go cautious and I'm gonna bid high on the fees. And when everyone does that, of course, you create this problem that, you know, it's- If everyone's bidding high, you have no choice but to bid high as well. So I think that itself will, will significantly change the factors, when we see fee pressure.

Sure, sure. And I think there were some other factors you mentioned in the article in the second era, Satoshi's subsidy, and you mentioned a few underlying reasons behind why Bitcoin developers were reluctant to so, so-called naively increase the on-chain block size or block weight, and a few reasons you mentioned were It's a one-off bump hazard. The devs want to follow the community and also software and services we're still preparing, and that also transitions should be gradual. Can you outline a little bit of the thinking on that?

Yeah, so, You know, the, the obvious thing was, yeah, well, you know, crap, fees are too high, we should enlarge, you know, we should enlarge the blocks, we should just allow more transactions in every block, right? and The pushback against that was, well, that we knew fees were gonna get, you know, uncomfortably high at some point, right? Nobody's gonna be happy pla-paying a billion dollars a year to miners, right? So You know, it might, you might, the, the argument then becomes, "Well, but not now, right? It's just, yeah, just, just not yet, right?" But that argument is a very slippery slope, right? When are people gonna turn out, "No, no, now is a good time to have high fees?" They're not going to, right? So, You know, there's always been an argument you should try to kick the can down the road. So it was, you know, that was a reason not to do it, but also, doing the enlargement itself is problematic because previously when we had seen, A block size increase, which, which had happened in a couple of previous occasions, it had definitely driven, huge amounts of centralization, and the reason is that, as a miner, the, the longer it takes for your blocks to propagate the more chance that somebody else pips you and gets in before you do. So bigger blocks actually make it harder to profitably mine. The different, the, the-- but if you're a really big miner, then that gives you a significant advantage over everyone else, because your chances are you're gonna find the next block yourself, in which case obviously you, you, you don't worry about how long it takes for the rest of the world to find out about it. Your block. And so we definitely saw in practice that miners all jumped onto the hugest pool they could find when blocks jumped, and they started, you know, and they started feeling that they were had more chance of losing out, because of that. This, this block propagation problem. We spent a lot of, the developers have certainly spent a lot of effort trying to improve block propagation, but they're very cautious about that. The other thing is that enlarging, doing a naive, you know, just double it kind of thing would be the first backwards incompatible change since Bitcoin's introduction Right? A so-called hard fork. previously everything has been backwards compatible. Like if you can dig out a zero point three two version of Bitcoin, in theory at least, with a couple of bug fixes, you can sync that To today's Bitcoin, right? It will accept the chain, it's new, it won't get upset at any point. So it's a big deal to force everyone to upgrade, you know? You have been allowed to run, you know, old versions of Bitcoin, they will still work. So, the developers feel that that's a big deal. It has to be really, really critical. to force everybody, to change their software. but the other thing which was really interesting is, you know, despite the fact that we knew fees were coming, they were gonna do this and everything else, very few people proved to be prepared for it. you know, Bitcoin Core had fee estimation and they pushed, you know, r-replaced by fee and everything else, but everyone else was pretty much caught with their pants down. so it's pretty clear that unless there's actual pressure, you know, people will defer it and they'll be working on other things stuff, so, you know, large companies that should know better, still these days are not batching transactions or doing other things, they're just grossly unprepared for the next fee spike So it's pretty horrible to see, you know, to sort of treat it this way, but you go, "Well, if we, if we give them relief now, then they won't ever do that work, right? They're never gonna actually get their act together, and, you know, so, so there's, there's like a second order of kicking the can down the road where they start relying on the stuff that you're, you know, of the fact that you will just save them again next time." and they'll never actually, you know, do the optimizations that they need to handle this case.

Sure. And in, in some ways, they were-- you know, the large miners were, in a sense, getting themselves a, would get themselves a bit of a protective moat if the blocks are larger, and some businesses are able to basically socialize their costs Rather than do the engineering work required to, you know, use up less of the block chain, so to speak. That's right. So could you outline some of the, some of the techniques that businesses can, or perhaps in your view, maybe they should be using to try and minimize the impact on the blockchain, such as transaction batching, et cetera? Yeah.

Yeah. So transaction batching is the obvious one, right? If you're paying out regular payments, you can actually group those together because, you know, when you-- when, when, you know, so, so Happens is I don't have a ten, like a ten dollar output to spend, so I take a hundred dollar output, I pay you ten, and I pay ninety in change back to me, right? but if I'm paying two people at once, it scales quite well because I have, you know, ten dollars to you, ten dollars to someone else, and eighty dollars change to me, right? That's transaction batch Right? So you can get, you know, almost a factor of two improvement just by batching your transactions together. Sometimes that means just, you know, you only send out one big transaction every five minutes or something. Sometimes it means you do slightly smarter things where you send out like a really low fee transaction that pays to you, so you instantly see it, and then come along later with a replacement transaction that pays sort of a, you know, a more reasonable, higher fee, but, you know, batches all of the payments I've sent out previously. Now, that's The, the, you know, the technology to do that is still kind of, you know, really kind of sinking in, but that's definitely the way, you know, we will see the large organizations going, and many of them, many of them do, many exchanges already batch. Some of them, in fact, did that for quite a while. Surprisingly, some of the large organizations still don't batch their payments, and, it's not clear why they're not doing that. Maybe they're burning VC money and they don't care. maybe there's a point of pride in going, you know, hey, we're responsible for thirty percent of the transaction volume. even if that's a completely meaningless and in fact a negative statistic, you know, perhaps that's part of their pitch deck, right? When they're trying to raise more money, this, this is what they say, "Hey, you know, we're, we're dominant in the space 'cause we use more block space than everyone else." So, you know, it's, that's just a pet theory of mine, it may be completely false. but also- No, but that's a funny

way of putting it.

Yeah, but also some of these businesses are built on Bitcoin as if it's a finished thing, right? They kind of went, "Well, our business is, you know, doing KYC and doing, you know, getting, getting licenses and all that stuff," and the Bitcoin stuff is really, you know, we- it's something we sit on top of. Right? and this is, something we see, you know, the, the free rider kind of problem is something we see a lot in the open source world early on when people really didn't understand that if you're basing your business on something, you need to be involved in its development. you know, Bitcoin isn't a finished product. there are still improvements to be made, and if you're not in there, then one, you're gonna be unaware of how to use it well, but also, you know I think, you know, it, it develops this kind of mistrust between the people who are actually doing the work and you trying to earn money off them, and I think that makes, you know, i-it's unstable if you try to take it for granted. And I think this is a lesson that's slowly being learnt in the Bitcoin space, unfortunately, we're seeing more support of, you know- The core development, that so many companies are relying on for their, their valuations.

yeah. Alright. And now, in your article, one- Quote that you use from Satoshi Nakamoto is, "Once a predetermined number of coins have entered circulation, the incentive can transition entirely to transaction fees and be completely inflation-free." So now the question then is, do you think Satoshi, whoever he, she, they are, did-- do you think they thought it would happen this soon, or were they-- do you think they were thinking more like twenty-one forty?

Definitely before twenty-one forty, right? I mean, because it's an exponential decay, so, so, so twenty-one forty is the point at which, you know, the last Satoshi, block reward, is, is, is paid out, right? The last block subsidy is paid out, right? but it's, it's, it's insignificant long before then. In fact, you know, we're already down a quarter from the original, right? We'll be down to an eighth of the original amount when, in, in a couple of years

The, the fees in blocks being sort of, you know, comparable with half of, you know, the current block reward. So it's, it's quite possible that that becomes more normal, but certainly I would expect in the next ten years, so another like three halvings, by that point, you know, we really are getting down to an eighth of what you're getting paid now for each block. I think it's pretty clear that it was going to be in the sort of the, the ten to twenty-five year range of Bitcoin that this would happen. But of Bitcoin, right? If Bitcoin doubles in price, then it doesn't matter that the reward halves. And so there's, there's a huge external factor there, adding this big random component. But it was pretty clear, it would've been pretty clear at the time that, you know, After seven halvings, you know, you're down to a hundred and twenty-eighth of the original point, you know, that, that's, that's looking pretty lean, right? So, it was pretty clear that that transition would, would- Would happen, you know, within sort of my lifetime kind of thing, it was never gonna be like twenty-one hundred.

Yeah, sure, sure. And as you mentioned, if the Bitcoin price doubles every four years in fiat terms, obviously, this then at least counterbalances the block subsidy that's halving every four years, and then once you add in transaction fees, and I wonder whether that would be enough.

Yeah, so that's, that's interesting, but I mean, at some point it really does vanish, and you can't keep doubling forever. Yeah, sure. so, the other thing is that at the moment, you know, as a miner, most of your costs probably are in some kind of fiat currency, but, you know, as if Bitcoin becomes more successful, then that will change. So the, the tone of the conversation changes. If you're not converting all your Bitcoin, then suddenly, you know, I think that, you know, a Bitcoin is a Bitcoin, right but also we've seen the, you know, I mean, obviously the cost to run the network now is a lot, lot higher than it was when it was Satoshi and won- and had his first peer, whoever that was, right? So, You know, at the moment we're paying, I, I haven't done the sums recently, but, you know, we kind of said, I think I had some rough numbers based on the, the previous Bitcoin price in the article, on what we were paying for. You know, for, for mining, and I think it's, you know, it's well in excess of a billion dollars a year, yeah, yeah. So, you know, in practice, the price going up of Bitcoin just means that we spend more money on mining. so, you know, and there's a valid question of like, how much money do we need to spend on mining? How secure do we need the network to be? and the number is pretty high, if, if you're using real economic value across Bitcoin, you know, you need to make the

Even, even if Bitcoin gets, you know, even if Bitcoin goes up in value, then you could argue that, well, you need that value, you need that reward for miners because now you're protecting more stuff.

Yeah, that's a great point. And I think it's, it's an interesting piece that, that you're making a point that you're making there that- There are multiple moving pieces here. So obviously you got the price of Bitcoin, but then you've also got the so-called level of security that Bitcoin provides. I mean, at various points in Bitcoin's life or cycle, it may have been, quote unquote, over-providing security or under-providing the kind of socially desired level of security. So as you said, it, it could just simply provide slightly less security or more, depends on how much people want to spend for that. so the, yeah. The next thing I was curious to get your thoughts on is, could you outline just some of your thoughts on what happened with the late 2017, the No 2x saga?

Wow, okay, so that's, that's obviously a big topic, right?

Yeah. Oh, well, basically what I was trying to understand is, do you think the, you know, this coming battle could be resolved in a similar way to that No Two X saga?

Right. Well, that was-- it was kind of heartening to see, To see some people come out of the woodwork and say, "Well, no, no, no, look, this is-- that is not how we run, run Bitcoin." And, you know, some of that, I think some of those lessons have been learned now. as I said, you know, some of the big businesses are now more aware of how to, to interact with a real open source project, You know, so, so I previously worked for IBM, and when IBM started to get involved in Linux and stuff, they had this real, you know, who, who's in charge problem, right? Like, if you've got this top-down corporate hierarchy, it's used to dealing with another top-down organization And, you know, it did their head in to try to, to, to, you know, they wanted to have a deal with an entity, right? But this chaotic group of ever-shifting developers, Really took some effort, and the only the fact that there was a significant amount of money involved, you know, did I think they really went through that exercise of, of, of figuring out how to, how to deal with that? And, you know, there's a whole industry now around Linux that, that now knows how to deal, with open source developments and, and open source developers and the kind of the, the chaotic, models that they have. and I think we're seeing the same lessons learned, in the Bitcoin space, right? So I think you can put, look at, like the No2X, stuff as, you know, an interesting lesson in how not to interact with the community and, and the res-responses that you will get if you try that. so I, I do think that, that it was interesting, it, it is heartening to see people stand up and say, "No, actually, this isn't how Bitcoin works, right? You can't just, you know, sign a deal with someone and say, "Right, we've changed the rules of Bitcoin." And it's good that that, that that isn't how it works, but also the understanding that that's not how it works means that, you know, people learn how to interact with, with the sort of the Bitcoin ecosystem, and because, you know, if, if, if things gridlock, then at least we get Bitcoin as it is today, right? It doesn't get any better, but it doesn't get any worse. You still keep your Bitcoin, right? and so that is the default outcome of any activity in the Bitcoin space. If there's no overwhelming consensus, nothing changes. And really, that's, you know, As sucky as that might be, as sucky as it might have been if we hadn't got SegWit, if it had, you know, if it really had been this untenable, you know, complete lack of consensus, Bitcoin would still be there Right? It would still work, just, you know, the same as before. It wouldn't have some improvements, wouldn't have some tweaks, wouldn't have some things that we really wanted, but, you know, it would have kept going. and I think that lesson is really important And somebody in the heat of that whole, you know, everyone was fretting about what was gonna happen, somebody kind of pulled me aside and said, "You know, Bitcoin will survive this. worst thing that happens is nothing changes. And really, when you're looking at defaults That's probably the best default you can have. If there's no consensus, if everyone doesn't agree, nothing moves. and I think that, that has now somewhat been learned, although I suspect, you know, as Bitcoin becomes more popular, more people come on board, you know? They may have to learn that lesson two or three times before it sticks.

Sure. Yeah. And I think we may, you know, we may see more debates and battles come, come in Bitcoin as, you know, these kinds of things come back and back again. and I think it was interesting to see many people in the community were- Well, community, quote unquote, were, you know, using things like hats to sort of signal their, you know, that their align, their allegiances. do you think Sampson will make a killing selling hats with twenty-one million on them?

Look, it's, it is quite possible, that, that, you know, we will see the same kind of hat-driven, you know, barometer of, of community, community feeling out there. It's, I hadn't thought that, but, you know, maybe I should get in line for, for one of those twenty-one million hats. Yeah. So, it, it was important that You know, people felt empowered, right? So, so people may have these opinions and may want, not want things to change, but they may also go, "Well, you know, how, how am I gonna oppose some of these, you know, giant companies and things?" And I think developing that sense of community made people feel like, "Oh, actually, there are enough of us that we can actually make this work." so I think the social side of, of, of hats and stuff like that actually is important Important, but people feel they're not just, well, you know, tilting against windmills and knowing that there is a, there's a broad base of support for this stuff. I mean, I honestly hope that, that I'm wrong about the third era and that people will learn their lessons before then, but I suspect, given what we've seen, people will give it a red hot go.

Yeah, and it may be that the fact that you have now pointed this out and the fact that people are talking about it helps people learn in advance for next time. So

So I published this was try to inoculate people against this, so that they, you know, they recognize what it is when it comes. it won't, you know, we, we may well be gised and look, you know, we got these Nobel Prize winning economists and they suggest that the system is far better if we have one percent, you know, inflation and things like that, but, I think it's important to see where, you know, where the motivation is, from those players so you can kind of- See through the rationalization of what they're doing, and so I think that is important.

Yeah, sure, sure. Okay. Well, I think that's a, you know, that's a good way to finish off that particular part of the discussion. Let's now change to discussing about your current work in terms of Lightning and the Elements project at Blockstream. but first, Rusty, if we could just keep this, you know, accessible for the intelligent layman, can you just give a quick background on what is Lightning and what are the key benefits for Bitcoin?

Right, okay, that's, that's, certainly something, something we can do. So, Bitcoin is this, this, you know, great, great system of, you know, money transfer and everything else, but it is a heavy system, right? So, every time you wanna spend, you know, move some Bitcoin, spend it, give it to someone else, everyone on the Bitcoin network, you know, finds out about it, they all validate it, you know, it goes to everyone, it all happens, and then, you know, ten minutes later

It's under enough blocks and yeah, definitely it's, it's pretty much nailed there. We accept that now. that's a really, really heavy process. So, you know, and we talked about why you can't just scale it up and that, you know, large blocks tend to, you know- cause centralization and you'll end up with a couple of big miners and then why were you, why were you doing all this work in the first place? Turns out that about, three and a half years ago, A paper was released detailing system you can build on top of this, where two parties will basically put some Bitcoin in and then they'll basically be able to trade it between themselves and they'll just by swapping transactions, so, so, they put some Bitcoin in, they both have to sign in order to spend it, and they agree that, you know, say ten bucks is going to you, Stefan, and, and, you know, I'm, I'm taking like, you know, and, and I'm taking twenty bucks, and then I pay you, and so we change it, okay, let's have a new transaction, it pays you twenty, it pays me ten, and we can shuffle money back and forth between ourselves all the time just by swapping these, you know, these signed Bitcoin transactions, knowing

that at any

You know, and as long as you're still around, we can just, you know, swap these transactions back and forth. So this was called a payment channel, and so the idea was that two parties could basically pay each other rapidly back and forth, eventually closing out by s- you know, spending the last transaction that they traded with the other party.

Yeah, sure.

So this, this was like, okay, that's pretty cool. Like, you know, so that, that's one way of amortizing costs, right? 'Cause we only gonna have to have like one Bitcoin trans-- or two Bitcoin transactions, one to open this channel and then one to close it, and we can do like hundreds or even thousands of transactions. That's great if I wanna pay you hundreds or thousands of times, and there are some models where that actually works, but it doesn't let me pay anyone else. The Lightning paper did a couple of interesting things. One is originally these channels were bi-- were single directional. I could pay you, but you couldn't use the same channel to pay me. You had to have a separate one. So it's like, you know, single one-way channels. The Lightning paper introduced the, how to, a method of doing these, these two-way channels, which was kind of interesting. But more importantly, I think they introduced this idea that if I've got a channel to you, and you have a channel to Joe over there, I can actually use you to pay Joe without trusting you Right? I don't have to send you a dollar and say, "Hey, can you pay this to Joe?" There's this way of making sure that, I can do a conditional payment to you and says, "Okay, cool, here's a dollar." If you pay ninety-nine cents to Joe, he'll give you the secret, and that will let you get the money off me, alright? And this is all written in, like, you know, the, the scripting language, the smart contracts that go on the blockchain, so it's really, really hard, you know, as long as Bitcoin still works, right? You can guarantee that this payment will work. It will either go through or it will fail, and the only way it will succeed is if you actually make the payment onwards to Joe. And so these channels now form a whole network. You can

The worst they can do is not forward your payment, and these things have a timeout. So, you know, worst things that happen, they time out, they, they, you know, and, and it fails. But the incentive is that you pay a tiny fee to them to try to succeed, so they only get paid if they succeed, so obviously it's in their interest to, to try to make sure this stuff works, and that produces this, what is now called the Lightning Network. So this paper dropped three and a half years ago of like, "Here's something cool we could do." But interestingly, the two people who wrote the paper, Tadge Dryja and Joseph Poon, weren't actually gonna implement it. They were like, "Well, it's kinda cool, but we have day jobs, right?" And that happened to be the time that I was joining Blockstream. And, I read the paper, couldn't make head or tails of it. It had a lot of, you know, it was like really, really dense, and kind of scattered. But finally, after the third time I read it, I figured out all the pieces. They kind of described all the trees, but never the forest. And so I wrote a series of blog posts on, "Hey, here's, here's how to understand the Lightning paper." and so I kind of pulled those pieces together so that anyone could kind of understand what the potential was here. and then when I was finally joining Blockstream, the CTO Greg Maxwell said We've decided you should work on Lightning, right? 'Cause, you know, you wrote some blog posts, you've got to be like an expert, right? So, so I was like, "Oh, okay." mainly because, It was one of those technologies that we really wanted to exist, right? The fact that we'd been shown it was possible to, to build this kind of thing on top of Bitcoin meant obviously we wanted it to exist, you know, it would make Bitcoin much more usable, it would be great for our clients and everything else, and since no one else was doing it, it made sense for us to step up and, and, you know, and, and try to push that technology forward. So, you know, we started playing with it, we did some refinements on the, you know, Taj and Joseph did actually end up forming a company, to develop Lightning, and then another company, Async, in Paris, decided also to do a Lightning implementation And then about, two years ago in Milan, we had a summit. Three, the sort of three groups kind of came together and thrashed out what the canonical spec for this would look like, because nobody really wanted to have three different implementations which didn't work together, right? We wanted a single network, so we had to agree on all the protocol details and everything else. and we were optimistic after we did that that, you know, it'll take six months for us to get something working, and of course, you know, this is software, so it took like, you know, eighteen months. but, you know, now we're at the stage where, you know, the lightning, lightning clients or something, they're all kind of in beta, beta, and they're all, you know- Caution's all over them, right? Don't put in any money you can't afford to lose. So think, think a network for beer money rather than, you know, you know, massive, you know, high finance. But this is something you can install and you can, you can make payments through the Lightning Network and, and at this stage, it's pretty much the, the motivation is, you know, you, you can encourage us to, you know Find the bugs and test stuff and, and play around with it. you know, and you can buy hats and, and, and stickers from the Blockstream Lightning store, and, you know, there are a few other places that you can, you know, you can spend your hard-earned Satoshi's and things like that. So we're definitely in that sort of, you know, alpha kind of early beta stage.

Yeah, sure. It's kind of contributing. Yeah. And then, so there are three main implementations or three big teams who are working on it. So I think there's Asyn, the, the French, the Paris team, and then you've got, the LND, software by Lightning Labs. oh, and for listeners, just check out episode nine of my podcast, that is with Brian Vu, he's the VP of product at Lightning Labs. And then A little bit about C Lightning and what you guys are working on and, maybe just tell us a little bit about how it's being used out in the wild.

Yeah. So, C Lightning, I mean, you know, I'm, I'm a C programmer, so, you know, we, I wrote Lightning D, and to differentiate from the others, we decided to call it like, we've kind of rebranded to C Lightning. Although the fact that it's implemented in C is really kind of a, you know, again, implementation detail for geeks, but, you know, that, that name seemed to have stuck. So, we're more server side, so everyone else is pretty good at making payments. we're optimized for kind of receiving payments. You know, you're running a storefront, you're most of the time you wanna receive payments. Not making a huge number, so we don't have a mobile client, for example, although there is a nice mobile remote control for it now, but you know, it's basically it's a, it's a server-class kind of, implementation, And most of our, our time is spent these days is, is, you know, trying to keep up with the spec, because there are a whole heap of options that you people, that we've added to the spec to support different use cases, as we've discovered, you know, pain points that people have and things like that so at the moment it's, it's very much sort of chasing, chasing those optional spec, spec things, you know, just improving UX, some performance work in there, you know, we run reasonably comfortably on a Raspberry Pi, but you do need a full Bitcoin node at the moment, and one of the, the big- areas of work is trying to, you know, how much of a light node can we, can we still run a, a, have decent security, and, and run at that point? And that, that work's being led, led really by the LND team from Lightning Labs. They're doing some really great work. The Neutrino, right? The Neutrino stuff is, is mind-blowing and it's, it's really independent of, of Lightning, it's just, a roast beef, Laloo, their, their, their head, head programmer basically went, you know, I hate all the, you know, the, the light client options that we have at the moment, we basically need a new protocol, and went ahead with that. So I'm really looking forward to that getting into Bitcoin Core and there's some great- Great theoretical work, and that will benefit all the boats, even if you're not using Lightning, I expect the next generation of light clients to use Neutrino really heavily, and it's much nicer for privacy, so, you know, there's been some side benefits like that, they've been just really fantastic to see. and it's certainly something that we're itching to, to implement as well. So try not to get distracted, but it is very cool tech. So, so that, that may come along. As an open source project, of course, you know, we get people coming by and contributing stuff, and we go, "Well, okay, we didn't plan to include that, but thanks. we'll put that in now, like way ahead of schedule." for example, you know, we have Tor support, and it was kind of, you know But, a contributor came by and basically dropped all support in our apps and, so that's, that was like, "Great, thank you, that's, that's included in the zero point six release," you know? Oh, very nice. So yeah, it does make it hard for me to kind of give a roadmap, you know, I can talk about the things that I'm working on, which is the stuff in front of me, but, you know, there's always these surprises. the other thing that's happening is, so, so, you know, not enough of my time, but, as much as I can spare goes into trying to maintain this back and kind of, you know, keep that, That collaboration going, because the spec is really the ground document on which all the implementations are based. you can't write a spec in isolation, you need to kind of, you know, play with things and go, "Okay, that didn't work," and, and that feed, that feeds back into the spec, and, and all three teams are very active, in, in getting that spec together. But, the one point zero beta spec has, has, you know, has been out for a while, and we're looking at kind of a one point with some, you know, some significant enhancements, all of which will be backwards compatible. and that will be kicked off in November, we're having the second summit, a Lightning Developer Summit, the final invitations went out, actually just this morning. Because I'm slacking, I was meant to do it last month. but, that will actually be held in Adelaide, so, it will be kind of cool to have all these, lightning, lightning devs, in my hometown, Get to show them around and spend a couple of days like, geeking out over the, the Lightning protocol.

Yeah, so, excellent. So what are some potential developments or features that you guys might be tossing around at that meetup?

Right, so, it's pretty clear that, splicing in and splicing out are really important. So at the moment you have a channel, but if you wanna get funds out of the channel onto chain or you wanna beef the channel up, we decided for one point zero, like, yeah, yeah, you shut the channel down, you start a new one. but- In theory, you can have, you know, add funds into the channel from on-chain. You want, you wanna put some more money in, or hey, I actually wanna pay someone on-chain, I've got funds in my channel, can we, you know, splice payments out, And, and do that without any downtime. So instead of having to bring your channel down and then get a new, new, new transaction through and everything else, you can do that while you're still using the channel normally. and that's definitely something that will, you know, will be a, a hot topic week in November and exactly the details of how we do that and how that negotiation occurs, that's kind of important. And one of the other things that we're probably going to change is at the moment we, you know, I said you kind of, you trade back and forth these transactions, right? So you're holding this transaction so that if something goes wrong, you can just drop that on the blockchain and, and get your money that way. the problem is of course fees. We, you talked about this before, fee prediction, and, and we talked about, I talked about the difficulties of doing that. Well, the problem with having a, holding a transaction that's pre-signed by the other side and you're ready to go is you don't know when you're gonna need it. And, you know, fee predict-- you know, fee prediction is hard in the short term, but, you know, trying to do it in the long term, you might want this in a month's time, what are fees gonna be? is really hard. And so we kind of punted and went, "Well, we'll grossly overpay fees," but even that may be insufficient. And it turns out that, you know, you may not agree what gross area for paying fees is. The other side may go, "That's too low, I want, you know, the fees to be higher." And you may go, "No, that's ridiculous, I'm not paying ridiculously high fees, I think you're insane, right?" So, it turns out this is actually a significant problem So we're actually probably gonna change the basis, of the way the transactions work so that they will pay minimum fees and you'll basically attach a parent transaction, that at the time you want to spend it to basically push this transaction through, so that time, that way you don't have to guess in advance what the fees are gonna be. and that's what, that's one of our pay point, pain points at the moment. So that's- Something else gonna go, going to go through, in November, we've already got some discussion around that. But one of the other things on the flip side is the ability to split, payments. So at the moment, you've got to find, you know, you, you know, the case I wanna pay Joe, and I go, "Cool, I've got a connection to you, you've got a connection to Joe." you know? I've got to, you, you-- we've got to have capacity in those channels. You've got to have enough money in your channel to pay that to Joe, for example, and I don't have any visibility into that. When you talk about longer routes, it gets a lot harder, right? You gotta, you know? Can I find somebody who's got enough money to pay into Joe? so if you could split the payment into small pieces, and yet Joe only gets it if he gets all the pieces, it's still all or nothing. But you can split it somehow. then it lets you be much, much more flexible in the way you, you, you route payments. if something fails, you might try, well, let's try a smaller payment, maybe that'll work, and then I can pay the rest somewhere else. And that's called AMP,

it's another, another lightning contrived acronym, which is great. Atomic

Multi-Pub Payment, right?

Thank you, Atomic Multi-Pub Payment. So, you know, and that, that again came out of the Lightning Labs team, and so we're probably gonna finalize the way we do AMP, I imagine at the November meeting, and that's, that's pretty exciting for us, for us Lightning geeks. So, there are, there are other things that are further down the track. So, Christian Decker of Blockstream came up with this idea called L2, E L T double O. And it is, a more flexible, variant of Lightning that we can do with a soft fork in Bitcoin that we hope will come through at some point. We probably won't spend too much time discussing that in November because it does require a Bitcoin change, and, you know, The timeline for that has to be, you know, would it be at least six months before we can use it? So, you know, that may be something that we look into like a, a one point two or two point zero spec But probably won't feature heavily, in November, although I think that, that is incredibly exciting and, you know, it, it's, it's gonna keep us employed for quite a while, but, But yeah, it's, it's, it's probably, we've got more than enough to discuss in the two days on, on sort of the, sort of the ju-just the net- the, the things that we've learned from, from basically having the network working and, and how well that's working, You know, and there's, there's a whole lot of other minor protocol changes that, that we wanna make, and so, yeah, so, you know, it'll be, it'll be two good days of sort of geeking out and trying to, trying to-- we, we won't complete this back in that time, but we will definitely get a kick off and we'll get like a hit list of stuff that we really want in. and probably a list of things that we're gonna push off again, which is what we did last time. so yeah, I, you know, I, I'm really excited about it. It's, it, Milan was, was really energizing, very intense, And, you know, the result of that was the BOLT standards, the eleven documents that came out of that, which, you know, several hundred thousand words later. So, you know, we'll probably see a similar, A similar burst of, of, of verbiage, come out of this time.

Very cool. What about, another feature that I've heard of, that people talk about is dual funding. Can you talk a little bit about that?

Right. Okay. So dual funding, at the moment, the person who connects, pretty much, you know, the person who proposes the channel puts the money in, right? they also pay all the fees, by the way. They go, "Cool, you know, I'll put the money in and I'll, I'll Inform you, 'cause it's not costing me anything, right? but there are many cases where you actually-- and the problem with, with, with, you know, I set up a funnel to, I set up a channel to you, and I fund it all, I put, yeah, I put a hundred bucks in or whatever, you can't pay me with that channel now because You know, it's, it's paying a hundred bucks to me and zero to you, right? So, so, so, you know, you can't go, "Oh, we'll pay minus ten to me and pay a hundred ten to you," right? You can't do that. So, For a lot of scenarios it makes sense for me to go, "I want to open a channel," and you go, "Cool, well, I'll put, look, I'll put twenty bucks in, you put eighty bucks in. and that way at least, you know, I'll be able to pay you twenty dollars with the channel, right? It'll be a lot more balanced." And things like that. There are a number of cases where you actually want both sides to fund the channel, rather than just having one. the other way we could do it is splicing, we could start, you know, one side starts with it all open, then the other side, "Cool, but I want to splice this in." That's just a bit less efficient, so it does make sense to allow this case where you have this mutual negotiation, And again, it was something that was cut from one point zero spec 'cause it was like, you know, it was already getting pretty complicated, so definitely something that people want.

yeah. Yeah, sure. Okay. another question I've, I've got to ask you. Now, one of my prior guests, Nick Bhatia, so listeners, you can check out SLP episode seven. Now, he spoke about this concept of Lightning Network reference rate, and he-- what he was talking about there was this concept of earning interest, an interest rate on the Bitcoin that is Staked, so to speak, in or opened into, channel opened into a Lightning Network payment channel. Do you have any thoughts on that?

Yeah, so, so I found it fascinating because I'd never really thought of it in that way. I think it's wrong, but I do think it's interesting. So, it, it's wrong one because you don't have visibility into all the money in the network. you have So,

you do have, public nodes? Nodes, yes, the, the, here's, here's, here are my public channels, you know, I advertise them so that you can use them to route, you know, and you will pay me a small fee for it.

but there are many channels that, that aren't, that, that are, that are completely private, and they're only, you know, basically I use them for my own payments, but I don't really want anyone else to route through them, so they're, they're just private to me, and so, it's hard to script the network that way. But you also end up with this dual funded case where you go, "Well, I'm on the network anyway 'cause I wanna make payments, but hey, I might as well, I might as well route for other people 'cause I'm up anyway, right?" So, the, the cost For doing that, for me, is zero. So, the fee I'm gonna charge is going to be almost zero, right? It's gonna be as small as, as I can imagine. So what happens, I think, in, in real life, is that, fees drop, to, to like, you know, as close to zero as they can be. in fact, some people might just put zero fees in. They might just go, "Well, what the hell?" in fact, there's argument that if you want someone to rebalance your channel, if you go, "Hmm, you know, I really want, you know, there's too much on one side of the channel, I really want to rebalance it," I should just advertise zero fees and let other people route through it in order to do that work for

be a question of the risk that is involved with running a node, right? So, you know, how much do I worry about this node getting hacked? It's got these funds in it, you know, because basically, a lightning node is a hot wallet. Which is disturbing until you realize that anything that made Bitcoin more useful was gonna increase risk. If you wanna use your Bitcoin, it has to be in a hot wallet. So any technology that lets you use your Bitcoin means, you know, you're gonna be in a hot wallet at some point. So You know, so, so basically it should price in the risk of holding that Bitcoin and how much you value this Bitcoin. Unfortunately, we've learned that people are terrible at assessing risk. There's always somebody who will think they're smarter than you are, and they will judge their risk to be lower. so I'm pretty sure fees are going to zero, like, you know, maybe not technically zero, but, you know, very much in the noise on Lightning. so I think this idea that you will be able to make money by having a node that routes, is wrong because there'll always be someone who's, you know- Even less risk averse than you are, who will, who will try to undercut you, but also because there is enough economic activity and there are reasons for people to be on the network that don't involve them making fees of other people, so, they will just suppress the price. and so while I think it's an interesting idea that you could measure this, I think in practice it's going to be, you know, more a technical decision of what is the minimum fee, non-zero fee that you can represent in the protocol.

It's gonna be, yeah. Yeah, yeah, fascinating. Yeah. I mean, and that you sort of touched on what my next question was gonna be. I was actually combing through some of your, documentation, and I found, there's a C-Lightning command called get route, and it actually incorporates a chance that a node would fail and it would cost you twenty percent per annum, then the risk factor would be twenty.

Yeah. So this is the delay, right? So, so what happens is if a node goes down halfway through, then you have to wait for the timeout. So, so I talked about how we chain payments. But, the way that works in practice is I say to you, "Cool, you have-- if you get me the secret in the next, you know, twenty blocks, I'll make this payment. Otherwise, you know, otherwise the money's going back to me." You turn around to, to Joe and go, "Cool, Joe, if you give me the secret in the next ten blocks, Then I'll give you the money, otherwise it comes back to me. Because you need to have time, if he gives the secret right at the end of the window that you've given him, you have to have the time to turn around to me and, you know, get through to me and, and take the money off me, right? So there's this kind of staggered delay. So that means that if I talk to you and then you vanish off the network, I now have to wait 20 blocks before I can-- before that's cancelled, right? So there's this kind of-- the risk is that somebody goes down at the wrong time and- You have to wait to get your money back, you know? And so you go, "Well, how much is my money worth? How long is that?" And that's why fees on Lightning are pro-proportional fees, right? Because it's like, "How much is at risk and how long would it be?" Right? So that's the, the, the basis for fee charging, which interestingly is very different from Bitcoin, where Bitcoin is charged by weight. I like to think of the Bitcoin network as a courier who doesn't care what they're sending, they just care how much Try to send it as a hundred pennies, they're gonna charge you a fortune. and the Bitcoin network works the same way, it's the weight of the transaction, number of inputs and outputs, and the size of the whole thing that they care about, not how much money's being transferred. Whereas in Liquid, sorry, in Lightning, we care, like it's, it's actually proportional to the amount being spent. Which is why Lightning makes sense for tiny, tiny payments, 'cause it's proportional, right? If you're sending a Satoshi through me, I'm gonna charge you a tiny fee 'cause, you know, if, if things get stuck and I have to wait a day, it's one Satoshi, I don't care. you know, a ten to the minus eighth of a bitcoin. So, so that's why Lightning is, is so, so nice for tiny, tiny payments. and it doesn't make so much sense when you get really big payments. One, 'cause you've got to find the channel capacity that you can route that through all these people, right? And the bigger it is, the fewer channels you've got to choose from. But also because, the fees are proportional, right? At some point, it'll be cheaper just to use a damn Bitcoin transaction.

sort of cohabitation where you get Bitcoin up here and you get, like, you know, Bitcoin via Lightning down here. so I think that's worth pointing out as well.

yeah, sure, sure. one other point I just wanted to clarify a little bit, from the discussion earlier around the risk factors, my understanding, correct me if I'm wrong, that is due to the, what we might call the cascading nature of the hash time lock contracts, on a multi-hop Yeah. Sort of journey, is that the reason?

Yeah, yeah, it's because you have to, you know, basically the time will, will drop across every, every node that you're sending. So you're sending through twenty nodes, and each one has decided they want like a five block Window between, you know, them offering money to the next hop, and then being able to accept it from the hop behind them. So, you know, they, they, they-- you wanna make sure there's a bit of a gap, right? Because what the way it actually works is they pay out before they connect, they receive the funds from behind them, right? They've got to make sure there's enough time there that they can do that. So they all incorporate a little, a little bit of buffer there, right? so, that- That of course is cumulative across the hops, so if you're at the end, that's kind of the worst case, you, you end up waiting the longest in the, in the absolute disaster case. so yeah, you basically just multiply how likely do you think that it is to happen by how much you think, you know, it would cost you to have those funds tied up. and that, you know, that, that's the risk factor in the get route, and it basically takes that into account and goes, okay, well, that node's actually asking for a really long delay, but it's a really small fee, so, you know, you have to figure out what's the time value of your money. in order to make that decision on, on which, which, which way you just, you, you route your payments.

Right, sure. That's fascinating. Yeah, it's really good. Thanks for the, the explanation on that. Okay. well, I think they're pretty much the key points I had on Lightning. The next area that I was keen to ask you about is around development and de-development practices. now, Rusty, you come from the Linux kernel development world, and now you're involved with Bitcoin and Lightning. Do you have any thoughts on how their

Yeah, so at, at, at the root, when you talk to developers, it's very much the same sort of open source culture, that, you know, has been shaped a lot by the Linux kernel culture, but Almost every project uses a very similar kind of, similar methodology, similar tools, similar thought processes on, on how and why you would do this. I mean, when I started my career, sort of in '97 really, full time as an open source programmer, that was really weird you know, it, it was actually more, it was more understandable to be a full-time open-source program at least I was being paid a normal salary, but, but, you know, why you would volunteer to, to hack on random things, Was not the norm at the time. It was seen as really unusual. There wasn't a clear model of why you would do this. how would you make money out of open source was such a common question. and now it's become pretty much the norm, you know? pe-people hack on this stuff one, 'cause it's fun, two, because, you know, they, they develop a network, and, and three, because it can turn into, like, you know, an employable activity and stuff like that. So, you know, it, it's It's always, it's always a fascinating experience and you learn a lot. So, th-those norms are tend to be across almost every open source project and, and certainly the Bitcoin project and, the Kernel project very much, similar in those ways. one of the differences is that, from, from sort of a surface point of view, you go, "Well, Linus is in charge of the kernel, he's like the, you know, he's, he's the big boss," and it's very clear in this, there is no big boss of, of, of Bitcoin, for example. But on a day-to-day level, it doesn't really make a huge amount of difference because, in practice, that, you know, there are maintainers who look after specific parts of the Linux kernel, and they're the, you know, whichever one you're working on, they're the person you're gonna be interacting with and going, you know, and, and, and, you know, submitting patches to and getting feedback from. I mean You know, I was dealing with the very core part of Linux, the module subsystem, so, you know, I was dealing directly with Linus sometimes, but pretty much, you know, he would leave you on your own, and, you know, unless something blew up, you would be left to, to your own devices, so it wasn't as sort of centralized as that may seem. so there's definitely a lot of similarities in the way the, the projects work, There's also a lot of similarities in the sort of, the early days of Linux and just the,

you know, it was a toy, right? Nobody took it seriously. you know, it was, it was never gonna, you know, there were the true believers who thought, you know, it was gonna take over the world, and, and there was the rest of the world who really didn't, you know, didn't, didn't take it seriously at all. And you very much see that, in kind of the Bitcoin project, a lot of kind of misunderstanding of, of what, of what, Bitcoin developers are trying to do, And, and so, you know, that, that's kind of a fascinating, fascinating thing for me, because when I joined Linux, it was very much the, you know, the crazy long-haired hippies trying to, trying to do this stuff, you know, they're never gonna take over from, you know, the dominant players in, in computing. And, you know, obviously that happened, so, you know, there's, there's more, you know, certainly it's probably one of the widest, most widely deployed operating systems on the planet, given All the Android phones out there, certainly that run Linux. So, you know, we kinda won, and what's interesting there is that that's not the way most Linux developers thought we were gonna win, right? They thought that, you know, we-- it was gonna be the year of the Linux desktop, from nineteen ninety nine, and it was every year afterwards. It's like, when's it gonna be the year of the Linux desktop, right? When are we gonna take on Microsoft? And it was always this assumption that- The purpose of Linux was to basically be a better Windows, and, you know, that never happened, right? You may notice Microsoft's still in business, Windows is still a thing, and, and yet it was incredibly successful, On devices that didn't even exist back when the Linux project started. I think that's a really important lesson. It's very hard to predict the future like that. When you've got some interesting technology, it may just, you know Rout around existing technologies, you know, the idea that Bitcoin will kill Visa is one of these things that I think, you know, hold on, I've seen this tale before and it doesn't end how you expect. So, you know, I, I think, I, I think it's, so, so there are these elements definitely, like, in the Bitcoin project, but, you know, I think one of the things that I did, When, when we started with, you know, the collaboration in the Lightning side, particularly, was look at both the Linux ecosystem and the Bitcoin ecosystem particularly and say We want to deliberately set out to make something that's much more inclusive, much more friendly, much more collaborative, than certainly, you know, the-- there was some, some- Sort of, you know, bad blood in the, in the Bitcoin space at the time, and it was like, you know, hey, we kind of wanna try to avoid that if we can, and so I think you'll find like the Lightning, development community, particularly, we have the advantage of being much smaller, of course, and having, you know, not having the same level of pressure on us. But, we're extremely collaborative, and, you know, I, I, everyone gets along extremely well, and a number of people have remarked to me, you know, coming into the development community, just how much, you know, how nice everyone is to each other, and that was something that, that we explicitly, highlighted as something that we want, because, you know, we wanna work on this stuff, you know? I expect to be working on this stuff still in ten years, right? So, you know, I want it to be a nice place to work, and so that, that is, is one thing that we've kind of tried to deliberately, cultivate. on the Lightning side, and I think so far, Touchwood, we've done pretty well.

Yeah, that's some great insight into Bitcoin and particularly Lightning development. how about the recent news on the Linux code of conduct and the supposed, you know, the diversity, co-contrasted with meritocracy? Do you, do you have any thoughts on that?

Yeah, so this, this is, you know, this is, this is a debate that's been going on at least ten years, in sort of the open source community and, and things like that, as, you know We realize that, you know, on the surface everything's a meritocracy and everything else, but there are certainly, there are certainly advantages, for certain people in a meritocracy, in, in this- Sort of pure meritocracy. in practice, you know, I think it is, you know,

you know, people have unconscious biases that I do think play into that And even though they might think that they're operating purely for technical reasons and everything else, we know, that people are biased in various ways and that, that can accumulate. So the idea that, you know, you should compensate for that has been one of these, these, these bigger questions that's been over projects, like the Linux kernel. From, from my point of view, certainly the Linux, community, we're building a product that we hope will eventually be used by a massively broad range of people, and- And it's re- we're making design decisions, we're making protocol decisions and stuff now that affects them without them even knowing it, right? So, so we're making assumptions about how people are gonna use this and, and, and what their priorities are, that are gonna be baked into the system and kinda hard to change So I think just from that point of view, the more diverse input we have, at this early formative stage, the better fit that is gonna- be, for what people actually want to do. you know, I remember somebody saying to me that, trying to explain, you know, sound money, like why Bitcoin is so important To an American is like explaining water to a fish. if you've got s-money that works almost all the time and it hasn't had, you know, major inflation collapses and everything else, you don't understand Why these things are important, and I think,

we kind of, i-it's, it's easy to kind of fall into that trap of having, you know, Of making assumptions about how things will be used because you have a very fixed small experience, and I think getting, getting a broader, set of viewpoints in is really, really critical. And Just saying we're a meritocracy and basically just, you know, continuing to interact with only people the same as yourself, doesn't get you there. So it is something that I think is, you know, certainly, you know, I've become more aware of in my work in the last ten years, and I think this is seeping through to different people. This is actually important thing that doesn't just happen, but from a personal point of view, You know, I'm always kind of horrified that, you know, geeks generally were these people who weren't treated well in school and they were kind of the outcasts and everything else. and surely this should be the last group of people who are gonna then turn around and form their own clique and not let other people in for random reasons. So, you know, I think we have an obligation if anybody basically really wants to work on lightning and stuff like that, my job is to help them. you know, and, and I feel really strongly that that, you know, the, the number of people have given me breaks and given me assistance when I probably didn't- didn't deserve it, or they didn't think, you know, I certainly didn't have the experience to justify the amount of time that they spent on me, and so I feel, you know, when somebody comes along and they wanna be part of the community and they're, and they're, You know, a-and, and they have something to add. you know, it's really important that we, you know, that we, on the one hand, we milk that, but also that we, you know, that we encourage that and that we, you know, treat them with the kind of respect that, that we would treat them When we came in, that, that helped us stay. So, you know, I, this is, as I said, this is a conversation that's been going on for a long time, and, you know, it's interesting to see that Linus has finally kind of, you know, come around perhaps to this view that there is some, some input in this. And I think, you know, different people are at different points of that progression.

Yeah, sure. Okay. Well, that's, that's been a fa-- it's been a fascinating conversation. I really enjoyed it, Rusty. where can everyone find you online? And do you have any other, you know, projects or things that you want the listeners to come and find you at?

Right, so, usually people follow me at, at rusty twit, at rusty underscore twit on Twitter, and that usually just shouts out to my, you know, medium blog and things like that, you know, but, I'm usually around on sort of various media, but, Twitter's usually the best way to follow along with the Lightning stuff. I just, I guess, I wanna emphasize that we're still in the early days of this, right? There are, still proposals come forward and go, "Wow, I didn't know we could do that. That's actually really cool." Cool. I think that's what keeps it so exciting, keeps everyone kind of jazzed about this, and, you know, this is definitely something that is, looking, just looking at the adoption that we're seeing and, and how positive everything is going at the moment, I feel is, you know, has got a huge future. In front of it, and I think, you know, it's an incredibly exciting thing to get involved in.

Yeah, fascinating. I'm, yeah, I'm really excited myself as well. Alright, well, I think that's pretty much it. Thanks very much, Rusty. Thanks for coming on. Well, thank you for having me. Alright, that was my conversation with Rusty Russell. I hope you guys enjoyed that discussion on the economic eras of Bitcoin, and if you're interested, you can find the links for this on the show notes page. Go to my website, stephanlivera dot com, that's s t e p h a n l i v e r a dot com, search s l p twenty three, and you can find all the links related to this episode. we also discussed the Lightning Network and some of the coming advances. and also if you liked this episode, you may be interested in some of the earlier episodes, such as episode nine with Brian Vu, the VP of Product at Lightning Labs, and also episode seven with Nick Bhatia. Otherwise, I hope you guys enjoyed the episode. If you got value out of it, please do share it on social media, as that really helps me out. That's it from me, guys. Thanks, and I'll speak to you next time.