Hashing It Out
Hashing It Out

Episode 36 · 2 years ago

Hashing It Out #36: Dan Robinson - HTLCs (And why they suck)

ABOUT THIS EPISODE

We have the pleasure to present one of the more entertaining speakers at SBC19, Dan Robinson. After two talks on Hash Time Locked Contracts (HTLCs), he drops a massive bomb in with a slide deck entitled, "HTLCs (And why they suck)." Lively and entertaining, we have him step through is thoughts on HTLCs and the alternate solutions to this technique that Interleger implements. Bombastic show chocked full information and analysis!

LInks:

Now, anjywelcome to hashing it outpacast, for we talk to the TEC in tevettors, BND, blocked in infrostructure and decentalize that one we dive into the weeds to get at Wy andhelp people bille this technology, the problems they fate on t e way, I'mlistening and learn from the best in the business. You can join OIR WRENK! Well, the back episode, thirty six ofhashing it out iner, host Dat, cory, petty and always on here with callingcruchet, say: Hello, everybody, Collin, hello, everybody, Colin Oqite, Sir Guess Today, Stad Robinson arrently,working at arodime and formerly at Inder Stellar in shame here to talk tous about Ach dl CS and why they may suck along with other types ofinteresting technology. That's currently in the pipeline M for helpingscale out a lot of the watching systems were in Tan what you start by givinghis introduction as to where you are and how you got into the space sure? Yes, so um thanks, you having meon Iam former securities lawyer D, security, litigator from New York, whoUm it's a refugee from that career and quit to become a blachine, protocalresearcher Um. So two and a half years ago I started achain Um, but before that o before that I'de beenI've been generally interested in Um and bicin. For a few years in etheriumine, the ttion, the lite paper came out and thev've been stort of followingthat and I went to work a chain partly in order to work on smart contractlanguages and Um phrotography and various and variousresearch there and UM ended up the past couple of years. I worked on. I had asmar contract language that I worked on a chain called Ivy, which is a acaticlanguige for Bik quin Um. I worked on some m some of the sort of Grt comitentchrotography, but a lot of what I'e work on recently relates to scaling, Umand particularly cament channel like and plazal like solutions for scaling,so I've Um h. You know I worked at when can gotAQUIR by Inter salary. I worked on payment channels on EA, O s, projecycalled starlight, and while I was working on that, I got familiar with the people. Um Work on Interlegere, which is a projectthat was started at ripple and got pretty interested in how they wereapproaching Bui, constructing a payment, nerworkand ways in which it differed from what people sort of the most h high profilepayment channel network in the space which is which is t e lighteningnetwork on Bickcoin M, and that led me to to really sort of like dive in on onthes particular question of other Ultimel tecnl question WB, whether touse a elsis in an these networks- and that was- and so I recently gave a talkand I've been going around just avangalizing against t e use of hlcs er,a solution used by lightening in favor of the of Um h the this different approach ofgeneraligitakes Hae to talk about anyway. So then Um that was wel. I wasan innerseller and then recently um about two weeks ago. I started as aresearch partner at Paradeim, which is an investment, firm focus on crypticcurrencies and I've been working, sort of in research,doing technicoldo diligence, andsourcing on investments and justtrying to keep up and follow the space and uh help us try to figure out whichway it's going and what the next big...

...thing is so first off, I want to say that youwere one of my top choices for first guest after SPC I was there. I saw youtalk. It was great youere passionate your Veryavocal, great job. On that talk, he really engaged the entire audienceand there was tons of questions afterwards that that were veryinsightful that uh people had, but it was. It was overall a very good talkand it was funny 'cause. You actually had to um two. I believe talks on each tlcs rihtbefore yours and it was likr a still see. An T was like why htl s Suck r e idle alike. I see that title and anew. It was on the purger, but I didn't like register with me that t was goingon N. I just kind like pusset out Lau, like that's bold, so yeah, so we havenot had a whole lot of discussion. We talk about thetopic swaps, like in aVey generalistic sensales pogram in the past, we've never really gotten intothe guts of it and the implementations behind it m. You know that's just single Tomasamultihop and that kind of thing, so I think it's really important for anaudience to get a sort of primer on what age t lcs are and what what a?What would they enable and then kind of lead into? Some of you know your knowledge Hof,some of the implementations that are out there and why they suck sure yeah. Absolutely so he lcs areconstruction. That's Um, I think you've been around for maybe a little fiveyears and they were originally designed to support atomic crosschain swaps. Soit's a way that you can do. The cononical example is is if someone'strading bicoin for light coin Um. I can send bickoin to someone and have themso like point to me and we can structure it in such a way that mytransaction wil go through. If theirs does so I'll only send the BITCOIN, ifI'm also receiving licorn, and then t l, those transactions succeed and it'sit's, it's a powerful construction and it's it's one that generally supports multiyou know I I cross Leger atomictransactions, even if those ledgers aren't um orn't public blockchain, soone so the way that they are used in the lightning network. An anotherpayment channel networks is it: they support Um atomic transactions whereeach of the ledgers are payment channels, a sayment channel. By the wayI on think I gat the same in channel. I think H, aman channels are fantasticand it's it's important to note that the lightning networwork and theseother Um thereare solutions are really two technologies Um that are sort ofcomposed into one, and one of those is payment channels which provide theseoffchained bilateral ledgers and the other is age. Cltis Wus to providecross ledger atomic transactions, so l Sezar woul allow Mulbi hoptrunsactionson the light new network Um, because the h you, if, if you got, for example,a payment, tro Malis, obob, Um and theyethey Ha paime channel with eahother, and you also have aieg ae channel between Bob and Charlie Um.Then if Alice wants to pay Charlie, she has to pape off, but only in such a waythat it only goes through if bobs payment to Charlie on their paymentchannel goes through and that O Lihtin provides all the hotpayments, and sothey use. They use a shllsesed to do this. But when I argue um the dark andwhat I've been sort of just like shouting into the face of everybody inthe industry that I meet to the test, Um eight months or O. is that there'sthere's an easier way to do it that doesn't have some of the downsides thatit Chelses have. So, let's get into what each? How e El sees work? Ouraudiencer is targeted towards engineers get is dirty and in the weeds, as youwant. How? How exactly do these work and how could somebody build like asmall model of any teelsee in there, yo w on their git of account and just playwith it themselves sure so at heart, Um, an HLC is part of a of a two pathcommit protocol that sort of resembles,...

...one that you that you'd use UM in otherareas of engineering and sort of the the phases of a ofdoing natial CD Um, I mean Shell, be based. ONMISH indaction. Is thatactually ot, two hcs one on thit, it's a big Qain across bicin like Corn TradeUm, so you ca one age, you'll see on the bickle network and one on thelightning network, and then you complete them metopically. So the Isselsees a contract that locks up these funds in such a way that Um h it can be unlocked if at only if it's the Correspondin is shilty on the otherchain, is onlunch n sor. The Way to do that is, let's oppose Um Yovegot Alice,als has Bitcoyn and Bob has like coin Um. Actually, usually, usually I useCharlie is the one who has lihe corn because a we all know Charle dumbdollars, light corns pectacularly at the top of the market last year. Sowe'll say, Ali is trading her her coin to Charlie and Charlie, I s trading adumping Ouis likewin on Onalis, so Um. What happened first is Alice, creates NHLC with the big Qoin that she wants to descend and what an hl se looks like onthe Bik Cuein newwork. It's actually just a big CN address. So if you go toyou, you can actually use ivy, which is the m thecin smart, conject language Umthat hele develop a chain Um. If you go to Ivy Hyphen Lang, Dotboarg Um, youcan actually try the Ivy playground there and you can actually see what alot of these contracts look like concluding what an ag Chlc he lookslike Um. I thank she'll, see it's but a bit coint to my contract. It looks likeit's just a it's just an address. It's an address that C that can be spentfrom an one or more ways and an GE she'll see as an address that can be.That can be either Um satisfied by one party, the recipient, by providinga hash or can be etrieved by the sender, which is another publice Um Prov aftera certain amount of time. So it's either the recipient plus this, so I'msorry to Tash Pretty Mage, so so a preimagic secret that corresponds to ahash or the center can get it back after a certain amount of time and soalice locks up her m diccoring on the bicer network into a Indo, an HLC where she is a sentor Bobis a recipient, Um Thi secret. That only right now only she knows is I'm sorry,yeah o right now only h h knows is uh is the she hashes dad and then shemakes tat the Hash in the in the HC and there's some time out. So let's make itforty eight hours so she'll be able to get this back after forty eight hoursor um Bob will be able to get it aft. I'msorry! Chrley we'll be able to get it after Um by by providing this hashpretty much. Okay. So now, Charlie ses, that this happens on the bickey network,now Charie doesn't know the primage, so he in order to to get it, he has to setup the other side of the trade for Alice. So he creates an HDLC on thelike Cornin Network, where he dumps all his likecorn into it m there he's thesender. CHARLIES escender Alis is the recipient Um. The time out is twenty four hours,there's a shorter time out. Um and the same hatch is used to lock it. So thisone can be unlocked by Alice by providing the secret. Now Alice knowsthis secret. You Know Thet, she ses at Bo ret as she can go to the like winnetwork use, her peimage M and her under public. He to unlock this. Ishell see and she gets all the light coin. INSO doing she reveals the secret,the the dreeimage that it needed to unlock the other, the other transaction. So the T thasill e over on the Um Biki network, so charlie, can go to the bicreen network,collect Um the HECOINTIPO by showing this this h secret. That he's justlearned and he has until the end of that forty eight hour period, to do itand the reason es time I was staggered is that Alice might wait till the verylast minute to actually use her secret...

...to claim the Um the light coin, and weneed to give enough time for Charlie to then go use that and collect the BIoinit before that one times out and finally, the reason that there's timeouts on this is that is so that if one of the parties just disappears, theother one won't be stuck, you know they won't lose their money. They'lleventually be able to withdri Um and cancel the ICH ils. So that's neededthis ree e. that's a SCIP Article Yeah! So that's that's how it works. When thetwo parties are trading on public block chains and it's a similar principle, Idon't go into thes sumution to Tal B'cause. It gets a little complicatedbut UM. If you have this in in a payment channel, you do somethingyou're, Jo Seing, similar with both with both the parties and theyessentially create a payment channel inside the HLC is one of the. So youknow in a payment channel, Alis can have a balance boken over balance, andthen you can just have the H. Ye has to balance if you ecept the paymentchannel Oter at some particular time. The ISSHEL SE ends up with part of thatmoney and then, if there's a problem, they can exipt the whole paymentchannel to the main chain and then settle the ch he'll see just like theyif wer one in the mainshaine in the first place. So that's you know in veryeey're, not going o explain really how payiment channels work is. That's I sown discussion but Um. I that that's how tos work as a sort of a Abgeneral,Abstract Idea for Cross Ledger: Orcomic, conductions ight. So I'd like to stopreal quick Um, and your talk also witted to this quitein detail with pictures which may helps quite a bit. It's always furthe Tranix,mens things completely verbally. So if you have no idea what the elvijust saidand Um you want to know more, I recommend you go to the shownutes andwatch his talk. So that way you can kind of be up to date on M, where we gofrom here and and you don't get lost in the weeds along the way. Yeah W get atthis slide. There too, as well Ya, Sur ion an the transcripts, Yep rigt. So what's the problem with that o? What's the problem with theseshelters, Yeah Aer? Where should I start Um? The three ponds that go over in thetalk and ti sort of the major ones are that there's this liquidity denial ofservice attack with this griefing problem? That's the one that I think isthe most serious, there's a free option, problem Um and that particularly whenyou have this AC, this trad going across multiple currenties as I we didwith the Bick ine like coin trade and there's just sort of an inherentcomplexity and difficulty of inlementation sowll start actually witha preoption problem. An this one recently got his sort of Gon mainstreamand there was a post on like on the like group about it, and I think Ithink people people mostly acknowledge this- is the problem with multicurrency like me. So the problemhere is that remember we had this time out of twenty four hours for Alice toclaim Te Liecorn. In this example. Now I the Alis, we have to have the time outbe loy relatively long, because otherwise, if it's, if it's time's outtoo early Charlie Charli, could potentiallyi get bo Um of the H, I youknow like Alice, tries to reveal this but doesn't get it included. Forexample, Um try te could pretentially claim both of the O cancel hiand Ye anclaim, the otherone. So you need to have the time U' be relatively long, but the problem thereis Um. If, if they're long, you know thisgives Alice this choice to just wait until the very last second wait: Twentythree hours and fifty minutes um and decide whether to complete the wholetrade or cancel the whole trade, and she could make that decision based inpart on whether the price of Biccoin and likecorn has moved in that timeperiod. So the trade has become more advantageous for her. She might H, youknow complete the trade, but if it's gone against her, she might cancel it,and so this is what's known in Finances: AN OPTION! U and that there' There'sreal value for her from having this...

...time, perie that she can wait, watchthe price move back and forth and decide actually whether to completethis trade and that's not necessarily necessarily something that Charlie umone or to actually agree to o. He wasn't. He wasn't giving her. He wasn'ttrying to give her an option. He was s trying to do tay trade, and so becauseof this, the long time it takes for chtes settle Um. This can be the sortof a problem for that, and it means it means. Ultimately, it would be hard todo an economical trade there, where you wouldn't be able to Weparyorcounterpartyou, wouldn't be able to take advantage of you and take tomanage that reaption then Um. This is even worse, actually and I', going todo t n to talk, but is it even worse, if you have the money on a paymentchannel tbicause the PAYIN, the channel adds its own settlement delay to this,so it would have to be something like two days or even more Um that you'dhave to choose whether the hours would have to choose hether, the completer ornot, and so that c a lot of people think it it's real. It's pretty fatalfor at least a at least a very Um. Ah, you know well developed multi, uh,Multiaset, lightning network m. You know again, you could imagine peopletoa sort of eating these costs or you could imagine this reputation systemkind of taking care of it, but to really have it be very robust,Um, it's very hard to do. Thas using hlcs Youn't like the base there to beas genneralized as possible trust as rispossible that didn't have thes sidesof and of wesed to gain the system, Yep Um yeah, absolutely, and so that's youknow, that', that's a pree option prolem, but people a e point out quiterightly that this isbe't really much of a pong with uh, lightning network trade.That's entirely on te ICQI network, because Um yeah someone can wad to lassecond. But why Woll be there insentive too Um and they don't actually get anECONOMICAC manage because it's all being settled. Ent Bitcon, both sidesof the trader bid Cin M sor that that's where one of the size of the t coricorttrade is Alice ending to Bob and the other side is bobs, ended, Charlie andso um there. You don't have a reoption problem, but that's when we got into uhthe other onthe one. I consider more serious, which is this grieving problemand that's a way for somebody to potentially just denial. Observisattack Um a huge portion of a lightning network. You can relatively littlecapital. So it's almost exactly the same as as why I to describe but t themotivation here is different, so when Alice h decides if Alice waits, justthe whole period and either doesn't reveal the her pre image at all or um W, we told very last minute a at thevery last minute Um Charlie had now has his money stuck in this h in this you know I thisis she'll seehe's not able to get it out Um until she actually completes it, and sothat's so he just got it stuck for twenty four hours, and this might youknow this might be. This is sort of annoying, but I guess Char, you knowsigned up for this trade, but suppose this was a multihop, lightning networkpayment, a d you can do this on lighting- and this is you know, sort ofthe whole idea of lining is is possible to these multihop payments. Where youknow maybe it goes. Maybe take a realisic example. Suppose Allis is anaccounted corn base and Bob is an accountof crackand and rather thanhaving it having you know, they don't have channels h directly with eachother, even directly, with the same intermediary. Instead Alis makes a h payment that hops tocorn, to crack and and then to Bob, so if Bob doesn't reveal his preimage Umfor the whole twenty four hours that first toney four hours, if Yohas it,that means that not only is allisis money locked up for that period, butthere is a portion of the channel between coin base and crackand like if,if it's one bcoint being sent, then just coin Bas and crackin just can'tuse that bit point Bo the for thet entire twenty four hours- and this isyou- know there- lightning network equidity, it's not very cheap Um. Yet there S, there is o ow here, hugesuply of it and someone can for free Um. You Know Char, I'm sorry Wat. They yeahBob doesn't even pay to do the attack. They can just sort of grief the wholenetwork by not completing this H, this...

...payment and just the money locked up inhis channels along the path and her. That was just like a three hop payment.Suppose it was a twenty hot payment, they've locked up twenty times theamount of the coin that they had, they had to be locked up, have locked up fora day, and so I just locked up this one bit coin in twenty different placesalong the entire path, and so you can really potentially cripple the networkUm. I Doan an attack like this. I just b sort of locking up all TL iquidity onI on t a payment that doesn't need omplete, what's important about thataspect, is at's basically free to do so. It's not like the attacker doesn't payanysaing at the cost of doing that, because people would kind of argue thatit's not spam. If I, if it's paid for that same argument, that they're gratsquite often right exactly and so in in this case,yeah because lightning doesn't charge any F if the payment doesn't complete,Um and itwould be very difficult actually to make that kind to madecharg on op be theire work, because you jusn't cually, have have to charge tosee Um unconditionally even for payments that don't find a path, or you know it's a little sier thanthat, but it it's hard it's hard to actually make Yo tacto pay in this caseUm, certainly without making other people end up having to pay as well Um,so yeah. So this is. This is something that someone just you know like creckright or something somebody with a gred against the lightning network. Um coulddo potentially in order to to really makeit difficult for any of anyone t any intermediarris to operate on it, not just to be clear for the griefingproblem. You need to have a a relatively equal amount of capital. Youalso have to have somebody on the other side who actually decides to exchangewith you m. So M This could possibly be mitigated by social constructs ratherthan just simple Um. You kno th, rather than Rpta just using protocol layer sto.Yet Ye people choose not to engage with you because you're doing this kind ofstuff, then you know, then that's that'ssomething you can implement, but in an anonymous system it's very difficult tomaintain and it could also cause other sort ofgaming problems. The reputation systems are kindy sufficient. If we're tryingto build something very generic and global, I'm trying to fill itot ro yeah.Absolutely so it it's true y. u you could you could try to do a reputationsystem here? What makes that difficult? In addition to the fact that it wouldbe nice if we had a network that didn't require everybody to be to be known toeverybody else, wet makes ititiony difficult. It isn't just your immediatecounterparty that can do this attack on you if you route a payment and you'rein just the middle of like a twenty Hof payment and the end recipient is amalicious counter party or just somebody wh who you know who just goesoffline to get sick e. whatever Um, you don't even know them, you're, justrouting a payment to them, or maybe somebody like halfway down the down theINTA, do that that will still affect you that to ends up Youin you for along time. Um, you don't you know, you don't know this party they're, not yourimmediate ACOUNTER party, but you're still exposed to them. So you reallyhave to learn not only about your the people, you open channels with, buteverybody along the path of every payment that you route you're exposedto this attack from Um. So it's it's it's not it's not a kind of a localtrust. It'. Very it requires, is kind of global Um, global trust and globalreputation. System that, I think is ther is largely incompatible with.Maybe what we'd, like a payment channel network to be structured? Like or Anotheu centralized network, I mean it's T it's actually incompatrible in general,with the philosopy behind what we're building and that we shouldn' weshouldn't depend on reputation to be trustless like that, that', theopposite of of a decentralized. You know a trustless network M. it's it'sexactly building trust into the network and you know, then you got questions oredemption and just you know what, if something El up like? How do you? Howdo you like rectify your bad Carma? As a result of that, you know it's just not what it's typhical towhat we're trying to build here and wane to build a robust protocol andRebos protogales do not depend on comma yeah, but so I cou D. I could probablyargue that this is the consequence of an early implementation of somethingthat will potentially become better...

...there's a few things that I could thinkhas optamizations. That would help not get rid of this problem, but minimizeit one would be like brouting optomization so like maybe enforcingsomething is lested optimizing, the route based on cost anda number of hots. So that way, you minimized it about that mouth thatsomeone could grief in that circumstance. I E twenty hops is a bitridiculous in any well connected that work and also you could do somethinglike that: Don'tknow what they call lighting their work, but basically youcan break up payments and route over multiple routes Um. So you can. You cannot have everything on I si channel but hold on. The other thing is like you,don't need multiple hops in order to have this problem and that if you wantto attack Cindat Orsom to say it admitigates, I I it minimizes theamount someone can grief, it doesn't get rid of the problem whatsoever, itjust it. It lowers the potential for someone to do massive harm or as muchharm based on the same amount of funds, yeah yeah, yeah H. I think I think I'malso very glad you brought up that tha point of Greatingupo pament Um in inoonpiezes, because that's that's what about to get you bout, the anelagerapproach, but I also want gress. When you mentioned trust. You know, Ithink I think it's okay. In some cases for the system to use a small amount oftrust Um to totadvantage, and you can actuallyget a lot more power, alsoe going to talk about with interleger. If you dotake into account the small ways in Wich, for example, immediae Channel Ccountof parties already have to trust each other Um, but the idea of sort ofglobal trust to one party that we all have to trust, um or everybody havingto know every part of the system. I think that is, you know, really sort ofpoor future for a system to have um se Oan, I think an I'll I'll. When I getthrough when I explain how the interlegure potocalls method ofEcotomic Payments Um works, I think Tryi to ing how you know you can see insome ways that it uses trust and Um. But I think in ways that that are verthat very much. Don't leadto t e to that consequences of trying to do it in this kind of globalepication way that it might be needed to Solva show the a Cho Sie Problem Gee. I think that that that that's t eequidity enial of service attack with the griefin attack Um there's one moreone more area which is complexity and so inbeddin I mentioned beforeembetting tament channels, an I'm sorry, hs uas inside of payment channels iskind of difficult. It's KINDOF, complicated, teeay divicalt toimplement Um Shell S, put some requirements on the base ledger likethat. It has to support Hash locks hase to support timeloks. If You CrossLedger Um hlcs like like M and paymen channels, t a you know they trydtocrossom one ligter to another. Those both have to suport the same hashfunction. Um they have to have relatively insic timers H, clocks, um the requirement it turns out to beactually relatively relatively difficult to try to implement thelightning network as it exists, um, because the the specification of it isvery bicoin specific and it really involves literally the structure ofthese Bicorintrandactions Thatd Anbee LCS. It's very hard to build somethingUm on a different currency, that's compatible with it, and you know. I'vetried M on Celar and I think I found actually that that's one wore areawhere I found in a later actild be much easier to tiure t how to use it on ondifferent substrates, so yess S, I can um if you've an we havp to talk moreabout th, downside, dividual, lcs or can go. You know straight to Geask USand talk about Um about what I think. Potentially it's a solution. I move for the solution. Kin Curious asto walike how you think you could we committed at this by getting rid of hissheelsees or even yet differently, YEP Gr. So it's IT turns out. I I Ilove this part I. This is not my idea that came from Evan Schwarzen TovonThomas Ripple, who are the inmatters of the Anreleger Protocol and YoknowaoPota call originally at hcs, but when they came up with this alternative idea,they realized that it was that it was a...

...lot easier and and didn't have thesdownsize, and so they sithe an alledeher to using that Um. So H, the basic way I work- and this is- Iusually pose it. U, in the form of a riddle, so if I want to send send you adollar and have you sen me a Yearou? Let's and let's ignore Um, ignore exchange right right now, but so sppos I want we want to do thattrade, but I want to do it in a way where you can't cheat me, and I can'tcheat you Um. So there's a problem if one par person goes Farist, the otherperson might not complete the Tradi m. So how uld you? How would you do it so I mean I hate to give away theanswer of this. What you're going to tat is what you doubt I said earlier isbreak it up. That's right, Yep Ho! Would that work. I would basically break my payment.Intoo well just say a hundred payments. So it's going to be on one cent per adollar and Sen. do that wait and then await you to do the same thing for meand then go back and forth until we complete basically the streamed paymentto each other ecomically yeah that that's exactly right and S. it's itkind of xly is a lot in common with a Game Theory Um concept of tit for tat,because Um, if I it turns, is into a repeated game where, if at any point Um,I try to cheat you by not sending you my penny, you just don't send me thenext penny and at worst Um this bplaying. This game is cost me onepenny Um. So that's that's that'! That's theessential core idea of it and that's. Instead of having this big trade thatwe try to complete as a blop Um atomically with some other trade, wejust put it up in the tinny payments and sent it a little bit at a time andwhat I love most about this, I think I think tha it. Definitely it very muchmitigates the the quitly Nial Service and the reoption probems, but what Ilove most about it is I simplicity, because you can do this on any paymentmedium that supports cheaf payments. You can do it, you could do it with UmUm. You do it on on change. You can do it in payment channels. You can do iton Benmo, you can do it with Bak wire transfers. You do it by growing penniesacross the Grand Canyon. Anything where I can just make tiny payment to you andyou can make Kiny bame parents back to me. Um can potentially support thisprotical, and so this I it's I similar to the way in which, like PCD, can bedone over ip or can be done in Um like homing pigeons or carriervigion. It's it's H, agnostic to the to sort ofthe details of the lower protocol, and so that that S, that's one of myfavorite things about anleture Um in this an thes particular approach andthe H T. of course it also. It also, like, I said, address wit, a preoptionproblem, because if you have a free option, it's only for a very um smallamount of of time, A and for a very small amount Um and you can steal a little from yourcounterparty, but we make it the payment, so small that you can. You canlimitid Arbertari Omeone Yo can make I onc otociate a time and you just sortof limitid to make sure that it's within the amount that you'reeconomically willing to lose from your from this particular counterparty thatyou have to train you and this the Potocolis by the way it set up. So it'snot that anyone along the chain of payments, if you ere a motopay, anconceal from you, it's only your immediae counter party and it'saftectable to you. You know that they that they were the one who h whocheated you in this case, so you can just close your channel with them Umand finally, yes, there'sthere's. It's ALO, Iquidity thanoservice problem,because these payments, the tinpayments uh yeah. They don't lock up capitalreally at all when they do only tiny amount alongthis pat it only for a very short amount of time, because the thingagainthe thing about t the reason that these time outs on the UM H, our leterinforcas Shl sees were so long is that they need to be settled on the mainShaine. And so you need to wait for the Paymane Chennel totle the mainchain can,and then you need to wait that Thir's parbly to be able to close that O,complete that that ie shell see and when you're not settling when you're,not when these are just sort of...

...informal agreements, essentiallythesehe higher level agreements Um to complete these payments, nothing'sgetting forced on the ledger, and so you don't have these. Have these havesort of like bound by the time that I picks actually settle thisso g t that that's hs a lot more details like in the presentation, Imean Ineligis a really big thing and I think it's a really cool project andI'm this focusd on a very tiny mechanism of it that I just findincredibly beautiful. I think I I M I'm a bigger pan of this pit than even eventhe people at Inaleger, Um and there's like one that I obsess about likethethin you're talking about here is quite a broad idea and coul be use on almost all current change, theirproduction right now, not ony, not only every every chain, Umand opeccially Chans that ride potocalls, but I I've actually justthoght a there's just a lot of different of different protocols. Evennot blotcing, not even you know, Mi protocalls. They can kind of takeadvantage of this inside and this is oething. I think this this predat. Youknow this, it's an older idea, but it was on the Soundr all o looking intothose Um s, porti calls were ch for exchanging information. If I want togive you a private key and you and an exchange for you giving me a privatekey Um, there's no way to do that. atomically Um, like with you know youcouldn't do that with like an h if you like protocol, so the way that the sortof interletury way to do it is. I give you one bit and prove that it's aparticular bit of my Pu private key, of course par paticulary public Y, a dmean you give me one bit like aibit. I mean literally a zero one, and we dothis over and over and either part party abort at any point, Um the H, then the other party only weknow that we know that the other party has the t most. The dionest party hasone bit advantage in brute forcin, the rest of it. It's that's T', a bit of atangent but Y. I really like itthats definilly dependent up on ethe aimacityof the transfer. You need to be able, like E, said, that proof, ninvolved anddoingsomething, like that for general information, has to be pretty robusteda D indefficient like it's Nice Oo, payments because eer every payment isit. Is it self just self, enclosed, playment? That's true that th nd! Thatis true, but I think there is there's other areas. I mean there', there'sinteresting analogys to this two. You can do pro different things ofprobalistic payments anyway. But yes, I think I think it is a. It is a veryrobust Um mechanism is, I particular design idea and Yoa. I just have an bigfan of it: IANDSOPREF for the fludters out therethat they be listen to this and say lightning's dead M. is this a problemwith lightning at its core, or is it something that can basically justimplement this? On top of it's? It's current routing, frameworkr Orshe'lsees a fundamental aspect of hell, whitening wheil and can't possibly workrighsofers ofal. I ant Yo, I'm a big fan of lightning and I think I thinkultimally the networs going to work out Um possibl Qi, possibly by by adoptingsome ideas from this Um, but I I think you know- and I think a lot of people'sdoubts about lightning network are tots generally about the about thepossibility of payment channel Merick, and at least I would like to be hopefulabout payment, gennal networks, ind general, because I think it's it'sreally cool design for scaling m. that's the question whout whether we'rstoing to pundemental tol light, meaning no in fact Um Soin- that not a lot ofpeople know it is that when you make a really small payment on lightning, it'salready doing. Basically this. If I make a one sitution payment, I'mlightening it just wouldn't be economical to put that in a in N HLC Mand try to settle it on the main chain because sell se. I mean it would be below the dustlimit for anoutput on the on the chain and just the the cost of adding that extra weight to thattransaction Um. So you PIK multipl transaction, to settle in ll C I it'tat 's, something like fifty cents is, is the m point below which- and youknow it pends- on a lot of factors, but it's around fifty cents, the pointbelow which basically economically, you can't settle an GE he'll, see on Chanuntil payment's, smaller than it tends on your lightning nop configuration butpayment small and then than around that amount. Ire just done almost exactlythe way that I astribed M in a sense...

...that it's done. I you just sort of basically make thepayment, but you still sort of have this coordination among the among thewhole, the parties in the route, and so they still use an aiel see at a higherlayer and the EFACTI. Also what ineletter does in order to m kind ofcoordinate these parties along the the route but um? It's not it's notenforcible by the maintaining and ultimately your counterparty can stealforty cents from you um if they ch you, while wh you're doing this kind ofthing, and so one way I mean at this that be a huge problem and lightningthere's actually a relivtely easy way to to vidigate it, which is just to dojust pt payments up into these small moutes and just send them, and that canbe done at a higherlay or that can be done Um. I guess I guess recall thetransport layer, Aye Um, where yeah they they they would be able tokeep most of the architeture of lightning network and Um, but just sortof changed the way that they send payments Um. I do think I mean theresome other things about the way tha the Neworgis issign that I think auresooptimal for this kind of payment network, but Um. I think it's omething that it coulddefinitely evolve towards that if t ist turnes out to be a big problem for them. So what about h? What if I put out a bunch of like fivedollar transactions- and I nickel and dined the last forty cents of eachtransaction- and I do this in bulk, using your system Um? Wouldn't that be away for me to sort ofacquire a bunch of funds that I shouldn't acquire Um and you know h.How would you mitigate that particular problem right, but the way the protocalworks that the Promot, the AAn cant, get exactly into the EEALS, how themaltehoopratical works, but you can check out Um ineleg. You can look attheir spack and se and sort of learn more about it. Um. The way it worksensures that the only party that you canceal from is your immediate party othe person that you have a payment channel with Um, and if you do that, it's detectable bythat counterparty, and so as soon as they notice, you steal some amount fromany payment. They can close your their channel with you and, and you you'resort of done doing business with them, and so yes, you can exit scam. Yourchannel counterparty ind this way. But it's not it's not really like me tobe a very pofitable thing for you and in fact, the way lighting doesn't in it.His is a possible Um way to to to do these. Things is instead of instead ofwill payment is pending having to be held by the sender. You can iknow the th Dedoil eves Ge get alittle Um complicated, but you can there's a structure in such a way that,instead of me being able to feel from you, which is that either is able toGriefe the other, so thire's payment, rather than Welt an transit rather thanbeing held by one of us. I can steal from the other, it goes into the minorfes Um, and so that's what lightning actuallydoes so, there's no way to actually profit from this. It's just that youcan sort of cause the other party to lose a little more Um. That's one tradeof that's PEDICTI genal ago us the basic idea he is, there isn't Um. Theywouldn't be really way for somebody to make a lot of money from doing this,but even if they don't make money if they make money by closing off thecompetitor. So let's just say: Somebodey wants tothrow op a cepetitorsay coinbase Um. You know this is where the socialaspect of the world being not all crypto comes to the play, and you wantto attack a coinbase you. Can it a bunch of five dollar transactions tocling base through R, whatever N, let's to say, crack, and that's probably abetter example, Um c you could interact with Crackii in some way that that'smalicious. They could decide no no longer to interact with you personally,but you just knowu of another anonymous account and keep doing it and you'reattacking your competitor. Now I believe that is a little harder in this.These pretilious situations, due to the social aspects, also having barriersentry, but it does prevent sort of like an oper platform befor that kind ofstuff to lerfully, Lig, Tlat kind of thing right. Well, so be the this isactually already a problem in some type of problem enlightening is that one ofthe parties, the fundor of the channel,...

...has to pay fews when they open achannel Um someone has to ha has to fun the the tranductions that close thechannel they've sor of put enough money in there that that those peeds can betaken out Um and we wos you know in general- is thatthe party that's funding the channel, the one that's taking the risk and andpaying the cost is the one that is less trusted Um, so the one that's maybeanonymous, and so you'd. Imagine that w you could imagine. The thane wouldprobably be true in a network powered by interledger. Is that when you havethese these tiny payments, one of the parties have to go first right, one ofthe parties. There is the way th the risk of the other party cheating and ifit's coin base versus you know like anonymous. Five. Five five read at fourM, just some some anonymous user in this channel, then most likely tosay it's probit's. Let's to have the user, be the one T to bear the risk ofcoin base deciding to Axnecan them, so you wouldn't be able to just create abunch of channels with with some partyand and Steel Tinda bets from themor Greef them for a little bit, because they wouldn't h, they wouldn't open.They wouldn't when they opene a channel with you, you're going to be the onefunding it and you'd be the one Um io would have to sort of have the Resk of them, cheating you to imagine like in in most channels,one of the parties, s going to be the more well known one, the more the morecrossworthy one Ond, the other one, maybe e, even sudonymously, well knownand the other one is. Maybe I maybe sort of the newcomer and you want thenewcomeware to be the one who altintally pays any Riss Cau e thechannel was bad yeah, but we would definitely have to solve that problemon a more peer to peer basis if we're going to create sort of like the globalxchange network. Ere kind of idealizing at this point is what I'm trying to getut. That's possibly true. I think I mean I I think it may well Um it may. It would already be a geatimmovement if we have an whee. All you have is local trust. Where all you haveis you know. I happen to to trust this person because I follow them on twitteror something like that. I'm willing to bet that they're not going to steal adollar for me, so I'm Soi'manly to open a channel with them Um and so the you know, I think, there'sthere's a lot of other sentralizing forces in M inlightnng. I think we weWana, really want to dosign the system to avoid that and that's a lot of of myconcern about the liquidity on know serve as attack, but I think ultimatelyyou know you have to have some trust with your channel counter party. Anyway,you have to have the trust, because the channel costs SOE money to open soyou're spending somebody to open and to close it Um, and so I think so. I thinka this doesn't really maturially change and I think in many cases you're tryingto channel cont of party you might be willing to trust them. Fer much largermat. So what did you GE? I? There was areally intriguing question at the end of at the end of your Talkat SBC. Itwas about H, ramping up the payment mechanism so that it examically grows. What were your thoughts on that afterthat? After that comment- and I do see it as possibly of mechanism formitigating some of these issues- we're talking about about the fractionalpayments and stuff like that right, so so the big, the big issue witfractional paintments M, with thewit these with T extrem, favors, pactitizedther in metod as it the payments packets. If the packages are small inthe payment will take, a large payn will take Loum, and so the H, you knowthe the problem. F You're, trying to send Youdo Emesei, like you, can maybedo Cen Paymin a second or twenty Paes, a second Um at most and if you're, onlydoing fifty cents. It's about ten dollars a second in thropet on thisHowdi you consend through this, do this titua channel Um. One way to so one ofmy one of my answers of that is just that. I think sending extremely largepayments on if hencannl network may not be a good idea. You might as well justgo to the maintrain boy. That's the whole an you would just go to the mainChai. You have options. You Yep, I I think th. I think that's right. I thinkreally. The killer use case Fortam Channel thatworks is is mostly forsmaller payments M. another point that it makeas. I think, if you have a ifyou, if you have much loer e payment, sometimes maybe it is find a way to afew seconds to that, but another I yet. I think this was. This is greatpointthat Theveid worck raised in the Um at...

...after the talk was that Um? If you havea long running relationship, a long running channel Um, where you've made alot of payments over this channel and you paid a lot of fees over it, you canum raise the willingness, your willinges t, to loseto the other party to be by some fraction of the fees that they've paidyou right. So if you uh, or that or of the value that you've run from thechannel, so have you imagined that you know like like initially, I only trustsomebody for fifty cents, but then we they've sent thousands of dollarsthrough the channel and Um paid. You know paid like, like you know,ten to dollars a feet to me. If, then later they they grief me, they feel adollar for me Um. If and I close the channel. Ultimately, I've only lost asmall percentage of the lifetime value work out, Om that channel so overtime.You develop this kind of his cur channel reputation, Um and that allowsyou to slowly raise this H. th th, this Um anwisk Ta, Tha Moi, daat Yep Yep! U,and what's what's interesting, is you know Sa d? I say slowly, but this isthat it is actually if you're charging a per a percentage um of each of eachpayment that goes through as ces and if you wil, if you're saying I'mwilling to cap it at. Like, I don't know you know five percent of the totalfree that I've received over theres. Some percentage of the fees that I'vereceived over the whole course of this channel that works out to wouldactually be an expidential gross in this in this band with so you know thethe overtime Ou, eventually the the Thi Panwar can get ar it pretty high, justthe amount that you're willing to have in flight. Even if you you're onlysaying I'm charging five percent pes on a I or you know, one percent and paymesto go through, and I'm only willing to lose five percent of the Ligft TimeToyou that I've collected on this. It's still that so expenital growth andstill can get pretty high and still once again like that in inconjenction, with I you're going to do large amounts of payments or l e largevalue payments in a small amount of time to use the main gener. I mean thatthis is a these. These we or two solutions are really for eithermicrotransactions or strubable payments. In a lot of ways. I think that's right,yeah H, there is no one scaling solution. It's it's! It's a you know:conlomeration of a bunch of different solutions used t an there is waysdepending on what you'd need that payment to do. Yeah, I think that's Thy other people like that should be kind of driven home a little more. There isno like one stop shop for all financial use. Cases were able to Avrad differentways of sending money around and rounding based on how you're supposedto be interacting with that human on a financial level. That's right- and Ithink I think one example of this. So one thing that came te channel networkin general, UM and michae payments, extreme extremin, paatized, perens-well, don't work for is nonvungerable tokens right. If I have accrypto Kitty,there's no way for me to to Split Thi script, O Kitty up into Indo a bunch ofTine apuses and send it to somebody BEON PEC at a time, and also there's noway to have a payment channel at work in a CRIPTOC Kitty, because cipocatitcan only be could only be in one payment channel at at a time. There'sno theres, no liquidity there, there's noes, no o Ile. I I'm Goin to go ahead.Disagree, I'M NOT GOINGTO! Sgr Fo is just saying that t a that's if yous inthe base case as a cryptickity. That is true but C. Those lassets cun take upmany types of forms and some can have partial value. So when you're looking more towards thefuture and I'm domin very far in the future, here I could proceed thingswhere an entire group of assets is actually sold in a buttle, but they arereceived in piecemail and you can get partial deliveryn on this particularpieces of assets Um, in which case it would just be a supplicate of saying.The entire acid is the root of a Mirgle tree, and I received this keyse thispiece this piece on this piece. My signer also said that they sensor thispiece o this piece. This ion this piece, but this piece is missing that couldall be traced pretty pretty pretty shere forward in the form of acomposable and decomposable asset Um, but is that pratical today TUC o? AndI'm not really all that worried about...

...it? But I do see a future where these kindof transactions don't just apply to currency Incan and just you know,fotapointinagers the they canapply to actual Um actual seralized assets thather stored of Blatching, Oe, N, Choiceo Yeah, and I think I think there there'sUm. I think if there's a token, that's onlyvaluable that can have the entire Ireis only one of them, and it's only usefulif you have the whole thing. Um like like, I don't Kn W. I e Ke a particularconcert ticket. It might be somewhat hard to to tocanize hat andter likewe're networkin it, because because it's not worth anything unless you'rehave te whole thing together, but I'm I'm much beay digress. thereuld be Ma,be a better way to to say that that's that's more generally, true is you canonly go down as far as Wath the animosity of that asset allows forRfokanimisity is the token you can't break it down. I would further, I wouldmaybe say to visibility, because because we use TAT imicity for theother things so oppen in the nomberstation, so I isibly O someting.Like that know exactly. I exactly what you're saying there wic what ere thePeople Tho findamental at an how Ar noring it down? Yes, just lating, NSTIST, isn't sufficient, like you do have the v Dou believe there are the nature.Beneties does allow for more complex structure out than what we are iently.I have someane of a more philosophical movement here, an something that Callinusually used to used to kind of ask, and a lot of our earlier shows wekindaf got away from it, but for here it seems it seems relativelyappropriate and that is um ine sense of problems like hl CS andwhy they may suck for these side of payment channels or or payment methods,and to difficulties we're having with solving real scalebal production level,H therrtsolutions for cross chain payments. Do you see the future of kind of our blotching future, goingmore towards like a one chain to rule them all we're having this Um widenetwork of of disperate chains, communicating O each other Yeso? It'sgreat question and I think Um one then Gonta say that is that lookingat right now with people we're talking about crosschain atomic transactions ora PA channel NEWWORK that that's bans, crass multiple chains, we're thinking,sort of multiple economic systems, multiple, you know the daimativeboansbic corner a thery and omything like that, but um in the future, it'squite possible that one network like aparium, might be actually composed ofa bunch of different sharts, and I think you know if you want atomiconductions from one shart to another, there's ways to do that: Thetare, sortof inherent and bilt into charting, but I think doing I the these kind of thesekind of cross chain swops of value. You can also be often be reallyadvantageous and I think, having Um this general, the general concept ofpayment channels that soult of are anywhere where my money, you know mymoney's in in the payment channel on some chain or another Um. I think it gives you a lot of freedomto change the architecture of just of layer. One and say it's not necessarilyjust in one gigantic Chan. It can be a lof if in change that doesn't mean thatthey're not able to enter people aren able to interact and send money towhoever they want to send money to ut that that o the Broder Point get it yet that theInnerleger ovision is that Um. You should be able to send money tosomebody and it doesn't matter if you both are using the same medi ere usingthe same currency Um. That's that's just an implementation detail, and so you know like right now. It's kind ofannoying that when I, if I've got sware cash on somebody else, is Benmo Um. Ithere's no way for me to just send square cash money to them that arrivesin their account as Benmo dollars or or if they're INORP. I want to arrive atZeuro. It's it's relatively hard to have a t, even today, to send paymentsin that way. Um! So yes Ri, I think that', that's the potential thatinnoleter has is just abstract. All this way in fact way how I'm holding mymoney, Howt you're, holding your money-...

...and you can just tell me this- is whereI hold my money. This is how I hold it is what I want to receive Um and thenjust like, send it and every all the inemerits intermediaries in between dothe work of transtranslating, what I have andto what they want to receive. I I do. I do think that we will Ltihave some kind of Multiploc Thain Future, even if it's all under somemomaginal system, um Ye. I think that that's important so that that's the keyphrase you put there at the end N, and so I'm Gointa, I have a obox got damnit. I GOING TA use it. So I I still I'm still in the one morblockchain, just the evolution of the world and the way it goes. I don't seeus, we don't have multiple internets, you know what I mean and what we'rerating is sort of a net of value. We might have individualies like thingsbut there's one protocol, which we all kind of adopted as are means oftransferring packets of data around as you're going to do this, I believewe're going to disame thing with value and I think it'll be very light andthin and nobody will have any problems, accepting it as the center of Truth andeverything you're talking about to me sounds better suited for lay or to andthat yeah okay, you can hold it in your own ledger, which is very privateand lock down, and only certain people can see it and superincrypted andthere's in one way you can get to it and that's because your company enablesit or this particular oveen organization controls it, but ultimately it inheritite truthvalue from one single source of truth and I believe, like just the wayevolution of society works. We all eventually adopt a sort of Linguifrancafor trade, and I I see block chain is beinganother t ar not just watching decential ase truth mechanisms as beingsort of another language. That is a language which computersspeak between each other. In order SA s, they could trust each other and Ibelieve we're going to build a tower of babble for that, and I think everything you're talking about.Yes, there probably will be entertained protocols, but their routing mechanismwill probably be some sort of main centere of truth. Well, so I I think what you justdescribe- T er, the there's, there's one way T at that: it all works outthat way, a D and it's because it's all on one R blocchin Um, on which we allhave consensus. But another is, I think, that the that you use the Internet asan alleg thin. Actually the interleger itself is- is Um, maybe more like that of sort of thisone thing that everybody is using Um and what it doe it connect. All ofthese different networks, O th, the Internet connected untwhat, were thenhad Orogdin thes computer networks using one protocall, by which anyonecould then o send back communicate with anyone else Um, and I do think, there'sa possibility that people end up storing money in a bunch of differentways and th. Some of it could be in bhanks. One of t could be on block Chaithing it could be on on Um. You know asterblock an or Oun e, more fullyfutured ones. So I think I I probably do disagree with you on that then,because from a U S perspective, this would be like Um. What intolegit givesyou is essential ability to abstract away all those differences, so I itactually potentially give V, give it a lot more room for for difference in aroom for diversity and in what the baselayer is Um, because, ultimately,if I want to tonvact with you, if I'm using an Innerleger, I don't have to bethe same language as u so to with another technological, more compaticalanalogy. If we all had universal translators, it's quite possible thatthe spread of English as the only language might actually reverse andpeople would m, because it Ould Bem there'd be much lower cost for peopleto comuegrated cross languages. We might actually destroy that tower ofbabble and end up scattering. You know the to the wind say: Hat the Y, everyonecoan to peak their own language, because when they need to come together,we can use those translators and they can all talk to each other at change.That's a change in the inthe Wat you're, considering base right when you startwith a one lotchange rule the mall framework you're, considering theBlachen, the ultimate baseslayer on...

...what people agree on and themultiblocchain world. It's it switched. It's a exchange, por a call whichbecomes more base than the actual leasure in hself. THINK THAT'S RIGHT!Yeah and I underthand that perspective. But my my my my when I look at what theworld needs and what it really wants. Is it once an oracle of what is trueand what is not true, and when you break things up intodifferent ledgers, you have to introduce an element. I Trust ThiSledger, and I don't think that the world really wants that. I think thatwe've desperately been searching for this one Oracle of truth th. This isthis is where our information is recorded. This is where things are trueand when you do an exchange protocol, you have to believe that the otherchain is adhering to a set of rules that you consider valid enough to doexchange and that audit process is taxing and costly and unnecessary, andI believe you could cut all that out if we just have one root root center oftruth, which dictates all of this, and then everything you're talking aboutmakes perfect sense to me, but as Laer to from a risk ofal. It is, I think,that's that's. It's a parsh way to move forwardbecause you have to get that single source correct and you coald Y H non. IRea Onop, if ten, I'm gein really funny with that everybody and everybody hasto agree on it, but I believe that that's a natural evolutionary process,tit would probably seem society. I meanin Aaer, just going to see. Thabenefits outweigh the cost of trying to determine whether or not you know thethe Farma chain is is accurate. You know what I mean I', I'm agnotic on the general question,but I think I definitely do think it camplain out the other way, becausewhen you just said there about, I need to actually gerify. what's t you know,F, I'm doing exchange with somebody. I need to verify their late measure.That's not actually true. I mean if I want to send money to somebody, I don'tactually care about whether, from my view, to my point of view they receivedit all t I care about is whether they're happy right wh. They receivedwhat they wanted to receive, and so it's a very subjective thing and I'mdoing business with somebody. I don't really care if we all agree onsomething all I care about is egreeing on something with the person that I'mdoing business with, and I think you know the economy really runs inthi kindof on this kind of local trust, Um and the the problem of sort of Thi ofglobal of globally. Connecting these these, these different pieces of localtrust. U Into something where, where we canset we're, can actually maniftrnsact ar people who are um H, who have different sources of truth from us.That's where these potocalls become become very useful, so AA. I think Ithink that's I see that being potentially where it works out. I thinkit's, I think, there's some aalogies here to the Internet, but that thatcould be a whole other discretion. I definitely encourage you if youhaven'nd, yet to have M Hav, avange Schwortz, Mos, Pon Thomis, on to talkmore about the broader inerlege revision. Yea Al Te could think a lotmore about that that was helpto. An baoer o make one more point in that.You know the way people value things is becomes very subjective at that pointtoo, and I think people don't like that either and it's very it doesn't helpautomation at all. You know values. An value is inherently subjective, no notby ne, not by Ne. I that's what I believe that I think that we arestriving towards more economic palibrium, hat's, always been kind oflike e Holy Grail, and a lot of that is just in determiningwhat the fuck is value fair. Well, we could go balls deep intoget inncredibly philosophical right now over to wrap it up. B'cause it's beenabout an hour. Are there any questions that Um? You hoped we would have asked you andwe didn't? No, I mean I think we got two M gotthrough a lot of them. The one thing I want t I want to say: BCAUSE, I wantntobe fared O that Um critics is Um. That's T er. ' There's one otherdownside of the inalleger approach that I forgot to mention Um, which is that H,you can sometimes have payments partially complete. It's just possiblethat I'm trying to send five dollars. Somebody andquily dries up in themiddle of this payment and I only end up sending two dollars: Uh Fiftand,some people avpe a lot of lihting...

...people for some reason. This justreally stretches them out this possibility of a payment partiallycompleting. Personally, I think I think you know there's a lot of way to fi tat.You can you can find another path, you can refund it. You can just go stettleon the main chain, but I also think you can never Li Yo uilt UNOU o a lttle. Isthis the same way why the Internet works now like if you drop pack likethe reliability layers built on top that's right and think an YTN? Can youhave a liquid enough n network, especially because you know nobody likegrieping the whole thing with hts, if Youve a liquid enough network, itreally shouldn't be a problem to ty to find a way to complete er or fin that,but yet I is sort of a downgia. It does require anything really. We arecatecting a lot O of how we think about these payments hat we're not trying toworsip adimisity as this as I halse idle. Just like the way the Internet worksnow stream. Everything yes, yeah, stick around a little bitand well we'll talk about tetal guests that we coan P, O o the Sha sh anthingcoming on and for the guests who are still listening and still with us.Thank you for joining us. If you like this, please hit subscribe that thelike buttons share with your friends on twitter. You oan find me at Atcuorpetti,UN, twitter and calling at at Colin Cuchet Co, Lli, N CUS CE and Damn.What's your chiugter name, Ohyeah, I'm at Dan Robinston entoener there you go,I to go si Dan and give them all your hat speech and tell them how much o lotdo no hate speech guys. I there's cory, that's Goan to be yourSallie! Do all right! That was great man.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (108)