Hashing It Out
Hashing It Out

Episode 58 · 3 years ago

Hashing It Out #58 - Stellar - Nikhil Saraf

ABOUT THIS EPISODE

International payments is not an easy challenge to tackle, and Stellar is working diligently to upend the current aging and fee-ridden systems for a more democratized method of international asset transfer. In this episode we have Nikhil Saraf, Principal Software Engineer at Stellar Development Foundation, to educate us on the core features of Stellar, how it achieves consensus, and the practicals of how Stellar can be integrated into regular business and consumer applications.

Links:

  • Keybase - Wallet for Stellar and end-to-end encrypted chat application: https://keybase.io/
  • Best way to contact me (via Keybase): https://keybase.io/nikhilsaraf
  • Coinbase Earn tutorial videos for Stellar: https://www.coinbase.com/earn/stellar
  • Stellar's Reddit community: https://www.reddit.com/r/Stellar
  • Stellar's chat groups (Keybase): https://keybase.io/team/stellar.public
  • Documentation for Stellar: https://www.stellar.org/developers/
  • Stellar Github Repos: https://github.com/stellar
  • Kelp, open-source market-making bot for Stellar DEX and centralized exchanges: https://github.com/stellar/kelp
  • blog post summarizing SCP: https://medium.com/stellar-development-foundation/on-worldwide-consensus-359e9eb3e949
  • SCP whitepaper (Simplified version): http://www.scs.stanford.edu/~dm/blog/simplified-scp.html
  • Full SCP Whitepaper (very detailed): https://www.stellar.org/papers/stellar-consensus-protocol.pdf
  • Talk explaining SCP: https://www.youtube.com/watch?v=vmwnhZmEZjc
  • Animated community video on the Stellar Consensus Protocol: https://www.lumenauts.com/lessons/stellar-consensus-protocol
  • Blog post about "The Liquidity Problem": https://medium.com/stellar-developers-blog/kelp-why-we-built-it-the-liquidity-problem-425ca5558c1d
  • Docker quickstart image to run a Stellar Node: https://hub.docker.com/r/stellar/quickstart/
  • Block Explorer for Stellar: https://stellar.expert/explorer/public
  • StellarX, frontend to the Stellar Decentralized Exchange: https://www.stellarx.com/

Donate!

Entering. Welcome to hashing it out, a podcast where we talked to the tech innovators behind blockchain infrastructure and decentralized networks. We dive into the weeds to get at why and how people build this technology and the problems they face along the way. Come listen and learn from the best in the business so you can join their ranks. Welcome back, everybody, to hashing it out. That's always I'm your host, Dr Corey Petty by trustee. Cohost, Colin Cuche is in the building. Say Hello everybody. Colin. Hello, everybody calling. Wait, I'm in building. Fun't see room tied of. Virtually we're virtually in the same room. Our hearts are in the same room. I got there. That's where. That's all that matters. Today we're bringing on. We're going to talk about stellar, stellar, luman's and everything that's teller's doing, and we have kill Saraf from the principal, I'm sorry, principal engineer, principle soft in principal soft engineer. I completely just blanked out from like a few seconds to go and I ask that question. That's what software engineer from stellar coming on to tell us everything about kind of how it works, what it's doing, where plans to go. So why don't you start off by the standard start off of how'd you get into the space and what makes you work on stellar? Yeah, so I think, yeah, this was going big way back into two thousand and thirteen. I was kind of casually following the blockchain space back then and bit going was around below a thousand dollars at that point and it kind of like went up to a thousand us. Okay, kind of it's doing its thing and it kind of like was around that price, strange for the next few years, and then come somewhere two thousand and seventeen, I was in the process of looking at new opportunities and Crypto was sort of picking back up as well, and I had the good fortune of learning about stellar at that point in time and I also kind of had a payments background. So a lot of things sort of came together and around October two thousand and seventeen is when I joined the team at Stello, like you're at that point specifically, but kind of the same community. I'm sorry. What was that though? What specifically, like you light here? Yeah, it was. It was the commercial side of stell though, but essentially still a working on this tell ecosystem, trying to work on cross board of payments using blockchain and just disputed computing. Got You, got you, got you, we say cross quarder payments. Is just so like. I'm going to I'm going to give you an emission here. My knowledge of seller is pretty much that it's it's it's a x ARP fork. Correct. Is that cur originally, originally, originally, but but if I'm not mistaken, most, if not all, the code right now is a completely different from the ripple code base. Let's let's kind of want to take this opportunity. I mean, I'm in the same boat. I said it myself. Relatively knowledgeable about blockchain and all that, all the players in space, but I don't know a lot about ripple and stellar in terms of how they actually work and if like it out one like, if you can make that kind of combination, you can say those two together, if that's even a reasonable thing to do anymore, and and how it differentiates itself from the other blockchain networks and what problem was trying to solve. It's like can we? I kind of want to take this opportunity to to just use this episode to try and fill that gap, because I feel as though a lot of people are in that same boat. Would you say that? So I think there's so I'm kind of being on the inside. I guess. A lot of this stuff is kind of kill for me. So I think I will take your suggestion that there's not a lot of clarity on what differentiates something like Stello from other blockchains and to sort of help clarify that a bit, I think, and sort of also diving into a bit of why I joined stellar in the first place, is that when sort of stemming from Bitcoin, one of the main things that people used bitcoin for originally was to transfer value across borders...

...because of relatively how quickly that could be done, and one of the main issues with that was the fact that Bitcoin is very volatile. So if you wanted to send someone a hundred dollars on the other end, by the time the other party received it, they would receive some value that was different from a hundred dollars worth of bitcoin. And to add to that, the fees was rising because of increased usage. So you're really not getting the the expected benefit of being able to send money across borders for cheap, and I think that's one thing that Stello has solved very well, because it's kind of also similar to colored Tokens on Bitcoin, where you can essentially have own tokens on Stello that are specifically begged to guardencies like dollars, euros, bounds or even crypto assets, or maybe even native, I seo Tokens, or any of the tokens one might imagine. And these stokens are essentially first class citizens on the network, and their value is derived from what someone who choose it tokens suggests the value is, and then the market sort of places they own interpreted value on these stokens, taking a Dook ount gone to body risk and any other factors. But essentially, if I send you one dollar token on the network, you can be pretty much guaranteed that you are getting something that is very close to a dollar, and this is done through way I understand it, this is that's the name. That's basically the main case for using something stellar. Instead, it allows for transferring any asset and a decentralized network right, basically like any value asset, you can peg that, you just said, any particular denomination or token. I guess you will to whatever you want and transfer that across the network. What's what you think about that? Is that who is validating these transactions and guaranteeing that type of thing? Because that the way that I think it. Sensus works inside of Stellar. It's very different than proof of work or or different kinds of proof of mistake. It's more of a, I guess, pbft type. It's it's just actis. Can we talked about that? Yeah, yeah, so the stellar consensus protocol was developed by Professor Mazzierre's from Stanford and it takes a very novel approach to consensus of particularly in the blockchain world. So it's more specifically federated Byzantine Agreement, which is very similar to pbft, and I think one our way in which I like to describe it is it's so it's a transitive closure of trust across different nodes. So, for example, if I, as a no, trust you, Cory, and you trust Colin, by the transitive property, I also trust Colin and that has to be taken into account when you sort of compute consensus, because what about the decision that you make on whatever it is that we're deciding? That is dependent on what colon decides, so I as an individual have to also consider that. So the network as a whole compute this transitive closure and that's what forms the actual network. And one way in which I like to think about this is the stellar consensus protocol is kind of like the digital representation of Human Trust. Now, at the end of the day, we cannot live in a or rather, if we have as trustics assumption, then that leads to a lot of inefficiencies because we are calculating the outcome of something that like, if I trust you and you give me a dollar, I know that I've got this dollar, it's a real dollar, all of that stuff. But if I run complex calculations to verify that you've actually given me a dollar when I trust you, that's kind of wasted to some degree. But if some untrusted party gives me a dollar, I might want to verify. Okay, what are the source of the funds? Is this valid? All of that stuff, and I think that the stellar consensus protocol does a very good job of sort of maintaining the concept of trust where it's necessary and then doing the computations of verification where it is needed most, and that's what allows the stellar consensus protocol to be very efficient and reach the transaction through, but that it has today of a thousand operations per...

...second, as well as a three to five second leger close time. How do you sneat? No, but like when you say that it immediately evokes like when you say it's a man as I have a model version of a human trust it almost so I've had friends betrayed my trust before. You know, I've had Byzantine friends. You know, Huh, what? What? What? What do you do if, like there's got to be some sort of reputation system built as of the the network as well? Right and that, but like how do you like redeem yourself if there's like an accident like which between trust, like the non trusted thing is actually, funnily enough, the part that I believe the best, like the easiest part, the part where you doubt everybody. Maybe that's just part of my personality. I feel like yeah, you could prove that, is that we already know how that works. Right. But like when I hear this whole trusted notes that set up, I'm like, okay, you know, you got to be able to kick people out of the network who are bad so does this just set up basically, like what is similar to, like the Click Network on a theoreum? It's a proof of authority kind of system between trusted nodes and and allow people to create these sort of like what sort I'm looking for because of the C Cory, I don't know these collections, know these cabals? Know these just consortiums, consortiums. That's where consortiums one of the see, where's incentive stellar which enables, you know, that kind of trust mechanism to occur. And then is that how that works? In my understanding, it better that way. Does that make sense? In short, so I think how did to deal with PBAT, that is, to deal with Byzantine behavior? I think that was the question calls trying to get to. Yeah, like how does it deal with that? Shit will not just busy. So there's two cases here. That's the thing that I get. I think I get the untrusted case, like I think I have an idea, like where you're going, that you can get a more detail if I you know, obviously, but what I'm really curious about is the trusted case, like how do you handle busines, tee behavior in a trusted scenario? Yeah, so, so you obviously have the you of the design. So we in a silicontrances protocol. You have quorum sets. So each note basically defines who they trust, and you need to trust three F plus one nodes, where F is a number of Byzantine nodes that you're willing to accept, which can include like malicious nodes as well as just network failures or whatever else. And so you can tolerate that amount of lying if you will, and in the event where you have more than such failures, your note would likely end up giving a bad result, which would then in turn mark you as an untrusted node. But for someone who is a vigilant supporter of the network, they would then recognize these issues and then change who they trust. Right. So who you trust is a very active and evolving aspect off the network, and if you find someone that you trust is constantly lying, maybe they've been hacked or whatever else, you would want to, for your own reputation sake, in that sense, adjust who you trust so you'll know it is not misbehaving. And what does it take to become a node to do this type of thing? And then what do you learn for proper behavior? On the network for validating transactions. Yeah, Sou so. In terms of setting up a node, it's pretty straightforward. You would just download the software, run it, choose. Anyone can do this. Anyone can do this. We even have DACA images that make it very easy to set up a node, and I think so. Could you repeat the second question? Yeah, what do you what do you earn? What's the point of becoming a validator on the network? Is there like, why would I want to participate in the network with other like proof of work systems, as particularly referring to mining in this instance, for proof of work? You're it's a selfless and it's that a selfissact. It's a selfish act. I'm participating in validating transactions so that I can earn the transaction fees and the mining reward of this thing. That's the reason why I do it. Most miners don't actually give a shit about most of the stuff they're doing. They're just trying to make money. Is that the same case for participating in the validation network? And Stellar Yeah, so Stella does not have any mining. There are no rewards for running a node other than the fact that you get reliable access to the network. So what we are trying to build is a is the Internet of payments, and by running a...

...node, you are getting access to the network as well as supporting the network in growing. So, essentially, so the benefits of being a part of the network are that you get to decide who you trust and all of that stuff. Now, if you just want to be a, let's just call it, a consumer of the stellar network, you would end up routing all of your transactions through just one entity, right. So in some sense you are making a claim that I trust this one entity one hundred percent. Yeah, but that may not be the actual representation of your own innate trust. So if you actually wanted, if you are running a big business, that is support, that is making use of the stellar network. You have a business incentive to run a node so that you can make sure that you are not duped by someone who you are handing off all of this trust to, and that ensures that all of your financial transactions are being drouted correctly and your business is secure, because at the end of the day, when things go wrong, you want to make sure that you have a well set up payment environment because at the end of the day, we are dealing with rule money. Yeah, but my question was, like it's a it's a traditional consensus or the condditional. It's similar to a traditional distributed consensus mechanism, and those are limited to maybe little over a hundred validators. How are you how are you dealing with that, like who's doing the validating and how are you getting either? Are you getting past that? Are you give a subset of people that are only participate in the actual validation of transactions? Is it is the whole network not agreeing on the same state, only the people who are operating with a particular transaction? Is that we're maybe that's the question. What's the data structure people are agreeing upon? What are people actually coming to consensus about within the network? Yeah, so that's a great question. So essentially, the the each ledger is a state and the movement from one Leedgit to the next is a an application off transactions on that stage. Okay, you said generally. Let's just that's a that's one, I think, fundamental difference that seller has that other blockchain networks do not. There are multiple ledgers that are passed around. There's not a single canonical ledger that you're adding state to. Correct. So, so I think the the Lego is the Guardians State of the network. Okay, and when I say moving from one ledger to the next, where I specifically refer to is the set of transactions that effect the change of the state of the blockchain. Okay, of the Stellu Blockchain, okay, and that results in, say, Lego and plus one. And essentially what all of those nodes are agreeing on is, number one, what is the set of transactions that we are including in this Lego, as well as the order of these transactions and the validity of these transactions. And then eventually, when that agreement is reached, then a lot of the then those transactions are applied with the guarantee that that will the application of those transactions is going to result in the same state. Given that state end was the same, you will always end up in the same state and thus one. Okay, so you're everyone is carrying around the same state. If you were all like network state of stellar and the dice. As a similar transaction that I say of doing one thing that changes the network state, everyone agrees upon that and then moves forward. So there is one exactly canonical ledger, which is maybe not put together the same way that my mental model of blockchain looks. We talked about what the block pain actually looks like the terms of like substructure. Is it a block change? Is actually a question. Is You can see this war in the deal t space or the block chain space. So definitely more in as a disputed ledger, but it's so. I think one thing that's very interesting about Stello's blockchain is that, given that you've verified the chain up until Ledger N, you don't really need to keep any history before that because unless you care about historical transactions, because you're guaranteed that the State is correct up until Ledger N. and essentially that's really all that matters. Is what out what is the value of each of these accounts and all the orders on the audiobook and things like that, although we do make that accessible...

...through other means outside of the block chain, or rather I don't want to. Yeah. So I think that there's definitely a lot of concepts out here and I guess an important point to bring up right now would be that stellar core, which is essentially what I would call as a node, is what's is written in c plus plus, and that's what handles the consensus off each of these nodes. It's highly optimized for disputed computing and not so much for getting data out of the network. It just essentially closes legos right. That's most of them. That's noted for structure these days. Is Not optimized for reading, as in like a database. You like database reads. It's just for doing its job and moving on exactly. And we then have a rise in, which is a Web site boll that presents develop a friend. The apis all of that stuff to get data out of the network, which also then communicates with History Archives that sell coal might put out. Okay, that makes sense, and so that's that's is that kind of where your business model, by the way, revolves around the fact that you have these kind of services that you can provide on top of the Women Blooman coin in the cellar network that would enable people who need that audit trail or that that ability to see where information is gone historically is provided. Are you purely a coin based like business model? So I think like in terms of the business model. So the Stellar Development Foundation is a nonprofit and a big pot of what we are trying to do is to provide these services for free to the world because we believe that it's making payment systems better. And Oh, a lot of these services anyone can build and use and a lot of people in the community up providing additional services on top of these for free to the community as well. So I think Stello is really going through a phase off trying to figure out what all of these services are the people want to use, and a lot of developers that on you to sell Comin and so to find these opportunities and try to Gob out their own niche in the ecosystem. That makes sense. So I guess I guess the the the the real question I have is like, if you're nonprofit, nonprofits still needs some way of you know, revenue generation is just purely donation based, or's this like like you're not taking any feeds on any of this? Is that correct? Now that's kind of where I wanted to go on that. Next question is that what what's the point of the stellar looming when you have a proof of work system? The the you had that proof of work system automatically gives you the inflation rate and the distribution rate of Dow Tokens entering the system and you understand how what the how people make money doing validation through providing that proof of work, and so that that's all kind of record, relatively known in terms of the incentive model of the people doing the validation. For stake systems, you stake, you stake money up front and then earn rewards based on doing the validation work. So you basically learn interest on your stake by doing the validation work. And transaction fees in some cases, and I know that there is a transaction fee relatively small, very small one for stellar lumans. Where does that money go and what is the point of Stellar Lumens in general? Yeah, so that's a great question. So when you submit transactions to the network, you are being a small fees, and of small fees is necessary to prevent people spamming the network, and that fees is then collected in a fee pool which is then disputed on a weekly basis. So you can expect anyone who is, I guess anyone who's using the network needs lumens to conduct these transactions, but it's not being wasted away by someone taking a big piece of that. It's being disputed back to the Stella Network itself. So it's still within that ecosystem and the whole idea is that you're not just creating a lot of these transactions that may and may not be meaningful, and if you were to do that, you would then be paying a fees for that. How is reduce a distributed yeah, you say, distributed back to the network. WHO and and...

...how? Yes, so you can. So each account holder votes on whom they want the fees to go to. And there's this concept of inflation, which also is how new coins are created, which also is subjected to this vote, which then determines who gets these payments. So one very popular way to sort of vote for these these accounts, of who gets the LUMANS is through inflation pools, and there's one particular particularly popular inflishing pool called the luminots in fishing pool, which essentially computes how many votes it gets and then redispute those redispute those luments back to the voters. So you're essentially getting the fees essentially getting disputed proportionally back to the network on whoever is voting for this pool or any other way of sort of voting for beneficiary. So it's kind of like the people who are doing the governance of the ones that are benefiting from the network fees, or at least like they're. They're, they're their weight in voting is swaying where that money goes, as well as inflation. Well, but I think it's in some sense it's I would like to put it as each man for himself, because if I have like a thousand womens, I would get an equivalent proportion of a thousand womens worth of votes compared to the total vote. Voting, sort of all the votes, and because a thousand womans might be less competitive total womans in existence, at least ones that are voting, I would get a smaller proportion of the full payment, which includes fees as well as inflation. So in some sense, theoretically, I've done the calculations, I can expect at least an inflation one percent of my loomans on a yearly basis for any account that I hold. That is that governance done on chain, like this voting? Is that all done on chainers? That done at a band of the seller network? Yeah, so that that voting is done on chain when the voting operation is invoked. So you essentially set who you're voting for on your account. So there's no it's not like you are actively casting a vote when there's a voting round called or something like that. You just set it and by default it votes for you to it votes your account to make payments to this other account, which might be your own, and that is taken under account and consider for all the accounts when inflation is calculated. But I would like to point out that inflation is an active topic in the community right now and we are thinking of what in which direction we want to take inflation. So that's something that is not fully resolved yet. So it's still kind of up in the air. All right. So walking through what I would be do. I say I wanted to like, I wanted to participate, I buy a bunch of Lumens and I run a node and I just it just sits there. What can I expect to happen? I mean I set I set kind of preset where I want my tokens to vote on certain issues and then I just let it sit there. What can I expect to happen is that there's nothing going to happen. Am I going to earn interest on that node? And the I guess, amount of Lumans that are voting on various issues, like what happens there. Yeah, so essentially, if you would need to have a use case for the stellar network in order to sort of give you a good justification to set up a note. So there's no reason to just sit on tokens really instellar. So you can use tokens for various different things. For example, Stella has a decent less exchange, so you could use the lumens to pay for fees to do treating and essentially the cost of trading is new free, because the costs of transactions on still a very cheap. So if you compare that to treading on sentilized exchanges, but you might be paying a like point one, point one five, if you have a new account and you don't have sort of reduced feed years, you could be paying a fairly significant fees for each trade that you make, whereas on the CTELEN network your fees is like very, very cheap. So you can have theoretically thin or spreads and you can have essentially more, I guess, finer grain prices, and that's one thing that you could do on the network. You could also use so we...

...have, for example, as this one APP called pigs by, which is in a children's education APP that uses Stella. So you could sort of use applications like that, and we the list of applications on stellar or growing. And essentially, by holding and using Lumen's your sort of participating in the sole ecosystem of APPs that are trying to provide applications for cheaper sort of payment mechanism where fees are less. So if another example might be that if you want to make up. So we have this one company that did an ICO, I think, in two thousand and eighteen by game tokens to play games. Right. So if I want to do that and use my gred God. So if someone's paying three percent of fees on their credit card, that money is being taken out of the system, whereas if those payments some being made of a Stello, then that money remains in the system because the fees is significantly low. And that's another way in which you could use lumens. And Yeah, that is and Lemits, which means in essence you're still paying that three percent fee if use a credit card by Lumen's. Right, how's that not like? It's not like I can walk up someplace in with cash in hand, which, by the way, I probably got from an ATM, which it costs me two dollars anyway, and by and by a Lumens, like from like from a convenience store. Like what is you're not really bypassing the fees at the financial institutions of currently and implemented here. Is that? Like? Am I mistaken in that? Like? So, I think so. For example, you could always use something like Gemini, go from dollars to USDC and then from there go directly to Lumens. So the only gost that you're really paying is the spread to acquire the lumens on maybe a centralized exchange. So you are because you're using your bank account, you're bypassing the fees, you're we're waiting the seven days, whatever, however long it digs. But essentially now you have this drunk of Lumens and for every transaction you're not in, going that fees. Fair enough, you're not. So and then, yeah, for once you're in, you're in. That's across. That's across that world with all blockchain systems. But like I want to kind of want to go through the the life cycle of actually having someone do a cross quarter payment because, like the idea of enabling cross border payments is to serve the underserved or make the traditionals? That's a good question. Is your is your goal, your target market, to serve the underserved or to make the traditional system just much more efficient? Because, like we just said, once you're in the system it's quite efficient, but getting to the system you're still interacting with a lot of the traditional infrastructure what has the same fee structures and pain pain points. How do we scared? Like? How do we make it? How do we make it cheap from you know, some guy in that the live nowhere to some other guy in that will nowhere in a different country, without having to go through the same rigor morew and time problems of the traditional banking system? If, if, if, those on ramps are always slow. How to sell or fix that? Yeah, Sou so, I think I would say that both approaches are definitely on the table. So my let's talk about the case of solving the dissolved. So that's definitely financial solusion is a big Bot of Stello's mission. It's what drew me to Stello and with I think that through the rock chain finance is being revolutionized where a lot of people that don't have bank accounts are now using wallets right and Stella has a few of these wallets that people can use, a lot of which even have support for like smartphones, and in on the soft countries where smartphones are quite prevalent, and roid phones, for example, you could just download one of these applications. So, for example, or one of the more popular ones that I would personally that I that I used, is the key base wallet. So seller has a partnership with them and someone in a developing country could download this wallet and then request someone else for money or, having acquired money or Lumen's or dollars or or the local currency on the stellar network, make such payments and that enables them to transact very quickly without having to go to their bank account, which they may or may not even have, and that's something that I think as people start to use more of stellar,...

...especially in developing countries, they would start to take advantage of that. And I think, coming to the other use case of revolutionizing the way things are done in sort of maybe like larger institutional approaches, Stello can be very useful for remitting money where, because one might hold stable coin back assets. You would not need to settle funds on a day to day basis. You could sort of settle them on a monthly or maybe even longer basis and just my holding these stokens from trusted entities, you are pretty much guaranteed that you will get that, maybe that's called it, a physical good work of money, at a later point in time. And if you are in the money remittance business, your model is such that things are netting out, you can be very smart about how and when you settle funds, with the idea that every single transaction that you're making is a lot cheaper than it would be in the traditional way of doing things. So so so read trading. I think both the options are on the table and I think for me personally, I believe that it's only a matter of time before people who can get value out of these systems learn about them and are able to find the value that something like the cellup blockchain provides. So they required infrastructure for getting into the systems a lot lower than what you would call traditional finance. So that way like getting into the system than May and starting to conduct yourself within the system for a long period of time is easier than saying, like waiting for people to set up a bank and then serve you in your and your underserved area exactly exactly. But that. How does that? How does that then? But against like it's I would I would say a good portion of the blockchain networks are trying to do that same thing and if they're not compatible then you're going to have an issue with who gets to do that first, who's how do you reach those people and have them use the stellar network versus the etherium or Bitcoin or something else? That's, I would say, trying to do a lot of the same stuff. Yeah, so I think, Um, so, just sort of comparing a theorem. So it's here M is. It's not a payment network, right. So there's certain things that that's Stello is better for and there certain things that a theorem is better for Competa Sello. So I think a big pot of that is finding that right use case. So if you want to use its theoreom as a payments network, they're going to be paying that. That higher fees for disputed competition. That can be avoided by using the stellar network. But if you are trying to do something that requires on chain disputed computing. Still, a may not be the right choice for that. So now, when I say that, I would like to point out that Stella has its own version off smart contracts. That is very different from the theorem smart contracts and they serve a very different purpose. But aside from that, Stello does not have it during complete language right. So there are a lot of things that you cannot do on stellar that you can do in a theorem. So I would say that if you are trying to achieve cross bad payments, you should consider your options. And on one hand you do have stello that is built expressly for cross border payments, which affords you a lot of native tools, such as issuing custom assets on the network when asset is a first class citizen, as opposed to on a theorem where you might have read your own smart contract and those contracts need to be audited, they could have bugs, all of that stuff. So then your thereum is a theorem is is. It is a poor comparison. Then I can let's go with like say, Bitcoin or ripple, which are two tube, two networks that are expectedly trying to make a global money thing. Yeah, so I think so. For bitcoin, for example, it's it's Bitcoin, right, like it's one token and and it also, I think you are also in some sense, then, we're bringing in consensus algorithms over there, because bitcoin is expensive to use. Right. So, yes, right now, that's the case. Right now, as well as like building things like colored coins and the initial implementations of things like colored coins on top of Bitcoin or also, expect, prohibitively expensive. Exactly, and I think a big pot of the wayste was designed was expressly trying to fix a lot of...

...those issues in a design that treats these colored coins or these stokens as first class citizens. So is that what the smart contracts and stellar for, is pegging those tokens or creating things like what would be like the equivalent of color coins on on Stellar? Well, my understanding from what you said earlier in the conversation is that you are trying to peg to actual currency. This is not a payment system where you can walk up to a convenience store and pays things for, but it's something you can kind of redeem in it in get your money back. Is that correctly? It doesn't help. So you have lumen and buy a pack of Gum. You know what I mean, you know it does it someth me does like you can use it as I would touch upon that as well. So first I would like to address Grey's question. Off What is question for Kats too, about bitcoin. He's got your talking about bitcoining color coins and like how the Smart Contract language yet? Yes, yeah, so, so smart contract. So if you want to create tokens on Stello, you don't need to create smart contracts in nutrition sense of a theorem. Nor do you need to create smart contracts in terms of seller. You literally just issue an API called to the stellar network and you get these tokens created right you submitlar transaction. Speaking more technically, you submitlar transaction that has the effect of creating these tokens right now. So it's so. What I want to want to point out out here is that the creation of new tokens is very, very native to stellar, such that it it's very cheap to do and from an API perspective, it's also very easy to do. And how you beg this value, this token, to whatever value you you sort of place on it, is by making the markets for that token against other assets, so maybe against some dollar tokens or euro tokens. And and also maybe the native women on the network. That is kind of what is allows you, as an issuer, to ensure that the value of the stoken is what you say it is. And essentially, if I'm creating a token and I'm calling it the euro tooken, as an asset issuer, I'm promising anyone who comes with this token to me, like on the network and gives it to me, I will give them a euro in their bank account and conversely, if they give me a you on their bank account, I will give them a urow on the network. Right. So I'm essentially the the translator of this digital token to the physical world. And anyone can do this for any asset, but in my case, if I do it, it's back by me, right, which is very similar to how money works in the world. But now, coming into Collins Question of well is, are these stokens only used to convert from physical to digital and back? I think if you wanted to buy a pack of Gum at the convenience store, you could very easily. What it's a question of, well, is the convenience store on the stelen network? Right, in the same way that a lot of people accept bitcoin for payment. This is convenience store accept women's or goodital dollars on the STELEN network? All of that? That's one question. Of course. If they do, then you can just make a payment by making a payment to their account. If they don't, I think there's definitely a very, very good business opportunity here that a lot of people in the community are sort of thinking about. Is, can can new businesses bridge this gap by being the intermediary where they receive the digital tokens, Lumen's dollars, euros, whatever else, and then make that payment to the convenience store either through their bank accounts, credit cards, whatever else. And I think this also brings up one one of my favorite features of cellar, which I think is extremely underrated, is the concept of part payments. So if I wanted, if I being based out of the US, only whole dollars and I want to travel to Europe and pay a convenience store to buy a backup come in Europe, I can pay in dollars and have the convenience store receive in euros and the network itself would perform the conversion of dollars to euros at a at a rate that is equal to or better than what I request, so I I don't have to go to a money exchange and pay your fees and...

...then a spread and all of that stuff. And given that stellar's decentralized exchange is very cheap to use, over time the rates would be, in theory, the most accurate with the smallest spreads. So essentially a lot of the fees that is being consumed by the middleman and traditional finance is going to go away by people using the stellar network, with people getting access to what would be considered to be the most accurate R it's given that this is a completely open system, assuming that it has use and liquidity, because like, like you said, that people have to then people have to put these assets on the network and back them themselves. If that's not being done, then centralized exchange exists but no one using it and you have bad market prices. Yeah, so that's it. That's a great point. So I think that's one of the issues that is based by pretty much a lot of exchanges when they first come up, and a lot of the changes that are popular today have obviously taken a big leap in solving that problem, the liquidity problem. In fact, we wrote a block post about this only last month, describing how exchanges have the sequidity problem what it is and to the point of people putting out these currencies that can be used. So, for example, recently why x announced that they're putting twenty six stable coins on the stellar Netwook, and I think with that comes the idea that they're going to support their own tokens in the best way that they can and that, over time means increased liquidity usage markets between a lot of these tokens, and I think it's really a chicken and egg problem, kind of like how you described. I'm trying to like, I'm trying to wrap my mind around this, because it's very like the way this is set up and potentially useful is very different than how digital scarcity works and a lot of other networks like that. The digital scarcity of bitcoin is bitcoin and that's what's that that's what holds value, and so the transfer of Bitcoin is the transfer of the value bitcoin and nothing else, and then you can use maybe the security of the network to do other things. If the room is a very different thing, and then it's a it's a kind of trustless, decentralized computing model and which you can build smart contracts that then create digital scarcities within the network itself. What stellar seems to be trying to do as allow people to introduce an asset natively that they that they have and then allow the distribution of that asset to didn't be to be put on the network. I can then transfer the ownership of that asset, whatever it is, throughout the network. They're in exchange, and so what the what the network is providing, is just the mechanism for transfer of ownership, not the actual asset itself. But that I work at all. But I support that model. So that's great. But, like I wonder, like that's requires a lot of trusts and people to put assets on the network and then back them. Do you need to trust each individual person who's putting these things on and that they have the asset, and not just saying it, which requires, I guess, maybe some institutionalization, which I'm not saying is bad, just that to feel like you're you need to have people that you can trust that then, if I buy something in the desert decentralized network, I want to make sure I can redeem that value. And if there's no trustless decentralized model of where that value comes from, that I'm trusting a bunch of individual people or institutions to do that. Am I am I getting this model a kind of correct in terms of how things kind of work? Yes, I think you are. And so I think the way I would like to describe this is this is obviously like a very, very long term thing, right, like if you go back to sort of the evolution of finance and all of that stuff, just sort of not looking too far back, just you know, I think that going from things from like, you know, people using like like agricultural products to sort of represent value, then the silver, then the notes, then do sort of innovations in money representation, like notes, better quality notes, holograms, all of that stuff, to sort of bank conglomerates, all of that to now the blockchain. It really is in a step up in technology in how money is captured and represents entered,...

...and I think the way I would like to the way that I think about what Stella is trying to do is we're really trying to capitalize on the blockchain technology itself to specifically represent value and make it very, very easy and seamless to transfer right so we're we're doubling down on the financial aspect of blockchain right and I think personally, I think that the the solution itself is very novel and by fixing a lot of the inefficiencies in money transfer that existed in Bitcoin, you can you still need two players in finance, the banks, the institutions, the the users, the people, everyone to buy into the idea of blockchain and money representation on the blockchain. But I hope it happens and if and when that does happen, essentially what we've ended up with is a system that is way more efficient to work with, with a lot of the other human side of the dynamics staying the same. And tomorrow, if I don't trust Scam Bank to hold my money, I'm not going to hold their digital tokens on the network. That's not going to change. Right. I don't think blockchain can fix that. That because at the end day, if if I'm if someone's going to treat me, they're going to cheat me. Right. That's the whole idea of like a that's the whole idea of like Bitcoin, both being a payments network and and a token on that network. The value of Bitcoin is a fact that you can't get it anywhere else and it's not backed by anyone, whereas the seller seems to be doubling down on the network side and not so much the value of the token within the network. Yeah, so I think that's does definitely more. I don't specifically want want to talk about the value of the token. I like to stay away from that. I'm more tim more referencing the trust model of the token itself. Right, the digital scarcity isn't is it created by any one person. In bitcoin, at least, the digital scarcity bitcoin is created by the network itself, and so there's no reason to trust someone to make sure that there's liquidity or it's good or that it can be it's a part, it's an intrinsic part of the network, whereas assets with instellar like how our coins produced, like it's just to the interest of the monting. So there's a part of it. Was a premitted like is like. You know what I mean. Now. So, so the definitely so. The lumans themselves are premnted, with some tokens being created on an weekly annual basis at a pre specified rate, and those are just distributed with the voting mechanism you mentioned earlier. Correct. It's nothing exact. Okay, so this is essentially from a scarcity standpoint. The protocol only accepts certain amount of coins. Every like you, as somebody who is a you know, as a chief software engineer and at Stellar, can't just randomly get a bug up your ass stick something in the code that allows you to sort of like print coins. Correct, the protocol exactly exacting? Yeah, yeah, so okay, so we can, we can. We get, at least believe that much right. So a digital scarcity standpoint is very similar to how atherium, any proof of work system or any you know, concept of proof of stake, as it may potentially be, kind of kind of works. Is that an accurate statement? Yes. Does that answer your question, Corey, though? Because no, because the majority of the the majority of the value that's being transferred on stellar is not intrinsic to the system. It is something that you're Peggy, it's something that you're trusting someone to to present to the network that then gets that. Then it's transfer and movement around the network is facilitated by the stellar network, not the actual value that you're moving. That the it seems as though the benefit of the stellar network is to bring value to the network, create its digital representation and then trade that within the network, not that have something intrinsic to the system like other blockchain networks. Right. So you do have. So the Lumen itself is intrinsic to the system, right, but that's not like, if you what people are pegging value too. Well, so there's definitely some value ascribed to the lumen. So say, for example, if I have so so so, coming to the path payment concept that I mentioned e earlier, you can go so if you wanted to go from dollars to euros,...

...you could go from direct dollars to euros, or you could go from dollars to pounds to euros, or dollars to Lumens to pounds to euros, and that would effectively get you the best rate. But if you want to do that in the case of dollars, to do lumens two pounds to Uros, then needs to be enough pounds and Lumens on the network to hold the value that you want to send. And speaking specifically about any of these assets, a pounds, that needs to be that issue that is willing to put out. If I want to do a million dollar transaction, then needs to be that issue that is willing to put out a million dollars worth of fluquidity on the network so I can then complete that million dollar transform exactly a nail. Million Dollars. You nailed what I just said. Sorry, you're nailed what I've been trying to get to. Someone has to put up those things and I have to trust that they're doing that well in order for me to facilitate those, transpose those like translations from one thing to another to another. That's not a part of the network, that's a part of people bringing value to the network. Yes, but now you don't need to trust them specifically because when you do the spot payment you are never at the risk of holding one of these intermediate assets. So it's completely atomic and if it's not bounds that it might be someone else putting euros on any other currency on there. But I think that itself is the chicken and egg problem right now. Now, Stello by itself is the technology that allows you to do all of these really cool things and it's up for people to sort of Fusty, be aware of what Stello can do for you and realize that it can reduce costs in their business and increase the reach that they have with other botanos banks, all of that, and then that's hopefully, along with other reasons, is going to be convincing enough for them to then, as you said, bring value on to the network. And I really like what you said there and I think, all right, so, cory, I get where you're coming from in that there's like this idealized to than we have, where everything is a digitalized asset and everything had can be valued in some digital currency. But we still also exist in the real world and there are systems which currently, like our current system, functions all right, like it's not the best they can be anymore. You used to. We have a global finance that's put it that way. So it does some like there are ways you can litigate, you know, if somebody's not doing their job correctly. You know, there are ways that you can hit their luman reputation by form, by creating this like like litigation system, like you can actually there are there ways you can hook the real world into a system closer to stellar than Bitcoin, because there is probably also this element of KYC. If you're dumping a bunch of money in there, like you can't just yep back a million dollars in without saying I can't. I can't say, Hey, I have I'm representing a million dollars. Now. I started a node, I put a million dollars onto the network and you're, trust me, possible. There is a litigation process available to you. If you're using a system like like seller, it is more real in that it's not it's and I'm not saying that Bitcoin isn't real, I just think it's like another asset you could list on a network similar to Lumen's sils stellar. However, I would say that that one of the things that kind of early still bothers me about this and I'm sorry, I'm trying to like the the episodes getting a little long. We have to kind of like wrap up. There's one thing I really want to touch on. I hope I answered your your question or kind of pointed on some of the concerns you had their corey. Yeah, but like by marrying the real world stuff, you have to also take in real world usage considerations and transactual throughput in a thousand TPS is not meeting these a standards. How do you plan on fixing that? Problem because these can get up to fifty five tho transactions per second on their stress test right real world use case, black Friday, worst case scenario, everybody's buying stuff through visa. They can get up to five thousand. You're currently supporting one, Onezero TPS. What is your plan to increase the throughput globally? Yeah, so, so there's does definitely a lot of work being done to scale the network and I think that involves improving the blockchain. The cellar blockchain itself, like still a call specifically, but also they would solutions like lightning networks on Stillo, which we've in our the stell ecosystem. That sty light and essentially you you get all of the benefits...

...off state channels through that. Right. That makes sense. And essentially, in the case of you know, Black Friday sales whatever else, say through Amazon, Walmart, whatever, each account might have direct channels with Amazon or to Athot body or whatever else, and eventually a lot of that would get setted on the network at this constricted rate off a thousand transactions per second, so a lot if the business can still be conducted. I think that in general the blockchain approach to payments is very different from the traditional sort of approach to payments and we're going to have to sort of realize at which point, with providing value and in the case of visa, everything is going through, let's go and go with, say, Theo one, right. There's no need for lay or two because it's centralized. Right. So the last question I have then, is probably, from a user experience standpoint, the most important, it sounds like, in that's finality. HM, how quickly can you come to a decision on the network? That's something is locked in and absolutely correct. And somebody can walk away with their starbucks coffee after paying in Lumens or dollar representation on the seller network, and I mean yeah, Walk Away with their starbucks coffee and I got if to sit there for a minute and wait for enough confirmations to actually say I got this coffee, or is this something that that you guys can come to agreement on really quick yeah, so, so with cello you don't have the concept of confirmations. So it's really just a question of three to five seconds. Your Leger moves to the next lego number, next state, and either the transaction is included in that on not. Now, if you have a a leger that is overcrowded, your transaction may not get included for the next three to five ledgers. And again, at that point we are operating in a whole different wall. And that's probably the point where you want to stop thinking about using Leo two solutions. But in the typicult case, when you're interacting directly with Leo one, typically that would be three to five seconds. And Yeah, you can, you then, once you get that confirmation, Colm that's a that's another. That's a relic of not a relict, but that's a that's a consequence of the consensus mechanism being used, that it provides finality. So because you've provides finality, you have to deal with probabilistic finality. That other like prop work, change shoots. There's no such thing as a as a fork in terms of like to the network. Just make making two just decisions at the same time and then working off and then coming back to the same one. So you have re reorganizations. There's that's that doesn't exist in it's a network like stellar, based on the way that network comes to consensus. That you know, pbft provides finality. I understand, I understand, but when you're talking on a global sense and the potential for double spens still kind of exists and you know right, and that's what you try and resolve. You know there's still an ordering problem. That's I'm not I just wasn't sure of that. That's if that makes sense. So pbft does provide finality, but like at the same time, you know, things can happen out of order and then they kind of like if, like the whole network doesn't see your transaction wants it has to kind of hop through. What I understand is that correct. You have trusted nodes, who then they trust, who then they trust, and you could have an automatic payment go off from Adobe in California while you're visiting a place in Russia and make a payment at a coffee store in Russia. You know three nodes out. They'll say it looks good to me, but then that payment from adobe conflicts and then suddenly you've got a double spend issue. That's what I thought was going on, just because it's kind of a global thing, but I might be misunderstanding like the so I think. So that's where the tron sit of closure off the network comes in, because at the end of the day, all the nodes are voting on the set of transactions, which includes verifying and validating such potentially conflicting transactions. So you have the concept of Gorham sets and got him into section and things like that, which goes into the details off the stellar consensus protocol. But essentially there would need to be one note that is knowledgeable about both the adobe transaction as well as the coffee puchase to then say hey, these two all conflicting, I don't allow this to be included in the transactions that and therefore you have to pick either one or the other. There's a couple of really cool kind of visual representations of hell these closures work that I found online. I'll try and find them and put him in the show notes, but I think we...

...need to start wrapping up. It's there. Is there any questions you wish we would have asked that? We didn't. So I would love to sort of talk about the project that I do spend most of my time on. It's this project called Kelp, which is a market making bought for the Stelen network and essentially it's completely open source, written and go. You can download it, you can run it if you are issuing tokens, stable coins, whatever else. On the stelen network, you can literally get this up and running and start offering bids and asks for the token and then create this liquidity. So anyone who's interested in doing token issuances on the network, I would encourage you to take a look at this project and ll find other resources to create this liquidity. But essentially what we try and do is we try and not only give you the ability to issue these tokens, but also the mechanism by which you can create this liquidity so you really get up and running with whatever token you're trying to provide and allow for any use case that you have as an application on top of the stelen network. Was it call again, Help K LP? Okay, okay, I make sure I had showed us to that. And and is that just an like basically open system? Anyone can participate in that and start doing that then build on top of yeah, it's yeah, you can build on top fit. It's extremely develop a friendly. It's very modulo and we have a lot of improvements coming out for that as well. Towards by the end of the CEO. We have a you are coming out very soon and yeah's cortly written in goal and and the plan is that by the end of the year there will be a significantly more sophisticated strategy of developer experience where people can write strategy is not just in goal and but also in Python, Joba script, whatever language that you want. And there's also support for trading on Sampliz exchanges. So in case you have your token this on Stello and also want to list on sentlize exchanges, you get the same experience for making markets upperating on simplize exchanges as you do on stellar nice and it's completely free. That helps to one of the like liquid issues I think we were talking about earlier in terms of the chicken of the egg problem. Exactly, exactly all right, where people could work, people contact you and learn more. So yeah, I think the best place of contact me would be on key Bas. So keybas also happens to be one of stellar spotners and has a lumin wallet built in and it traned application and you can find me the a key based AIO and it cuts her off. Awesome. Thanks for coming on. I appreciate that. I feel like I have a better understanding of stellar and what it's trying to do and where it's going now. Thank you so much for having me on Guardy and Colin. It was it was great.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (126)