Hashing It Out
Hashing It Out

Episode 76 · 2 years ago

Hashing It Out #77-Tendermint & Cosmos w/ Brent, Chjango , and Sunny

ABOUT THIS EPISODE

Sunny, Brent, and Chjango of Tendermint breakdown the components of the Cosmos Network from the IBC, Cosmos SDK, and the Cosmos Hub. Sunny covers what the future of Cosmos will be within the coming year. Chjango talks about Cosmos test network named Game of Zones which launches May 1 2020.

Links:

Donate to Hashing It Out

Now entering the Yo, Yo, Yo. I got some news for you. Good News. See, status is just officially launched version one of its mobile application, a private, secure communication tool that combines a decentralized messenger web, three bowser and a crypto wallet all in one package. You download one thing, you get three. Lucky you. It's available now on the APP and play store for truly private and secure communication. All right, so install it, then check it out, then have it be in the top rated usage on your screen time report on Sundays. All right, join the public channel, Hashtag TPP, where you can get some s and t that's status to to play with some of the cool features like Ns registration, DAP curation and the decentralized sticker market. That's right. So go grab the APP. It's easy to do. If you have an IPHONE, it's on IOS. If you have an Android, it's on the play store. If you have a windows phone, what's wrong with you? You could download it in your mobile APP store or at status that I am TPP. I'll say that one more time. You can download it in your mobile APP store or at status that I am TPP. No black bear, he's welcome to hashing it out, a podcast where we talked to the tech innovators behind blockchain infrastructure and decentralized networks. We dive into the weed to get at why and how people build this technology, the problems they face along the way. Come listen and learn from the best in the business so you can join their ranks. And we're back another episode of hashing it out. Today it's me, Colin Cuche, introducing the entire podcast, going solo again Corey Petty. Of course she could not make it tonight, but fortunately we have three very wonderful guests with us this evening. We have Chango, the famous Chango unchained from Cosmos, headed community. We have Sonny, I go wo Agar Wald for not that correctly, Agar will, yeah, enough, I murdered Ned, I don't know. And we have brought all right, researcher at Cosmos, and Brent Zoos, head us bet our special projects lead at a cosmos as well. It's come over and kind of talk to us, talk to us about their their platform. You know, why don't you guys introduce yourselves a little more so that people can know who you are? Let's start with you, Chanko. I entered into the blockchain space in two thousand and fourteen as a Bitcoiner, pretty much when I realized we were in for another recession and I thought that Bitcoin was sort of the answer to that. So that was sort of my foray into cryptocurrency and eventually I just thought that I didn't know enough and the best way to teach myself was to do research and technical writing, and that's basically how I came to become a writer, and then my writing was kind of discovered by somebody who was hiring at tenement in two thousand and seventeen, and then that's kind of how I went into on Cosmos. That's the quickd dirty, cool wait question. Given there were basically in a recession now, do you think that Bitcoin is still the answer? She's the thing. I'm...

...like one of those hoddlers that think, you know, it's Bitcoin all the way and until I'm proven wrong, then I'm going to keep, you know, believing in Bitcoin, which means that, given that we are now in probably the biggest recessions in our lifetimes, if bitcoin does not become the store of value that I imagine it would become, then it never happened and at that point I will stop leaving in it and I'm giving a time window of two years from now. So calling it right now. Say If bitcoy doesn't go to like a hundred kid in two years, you know, and then, yeah, then that also on the pole and say, okay, you know what I was wrong. Okay, okay, we could talk about that. So I don't think I agree with that, but I want to get everybody. It's too sort of give there say who they are. So sunny wants you go next. Yeah, sure, yeah, my name is Sonny. I got into the CRYPTO space probably in like similar time period, like in two thousand and fifteen ish, and I was just studying computer science and political economy and I'm like wow, this is a cool way of combining these two ideas, and so I started getting that for a while, sort of death, you know, sorry, with a bunch of bitcoin stuff tatter class at Berkeley. Got Into summertherium stuff, work Internet consensus for a summer, which is where I'm friend, but I guess you'll talk about that. And then I joined Cosmos in the summer of two thousand and seventeen because I was kind of annoyed with the state of blocking development on a theium and so I can't and I'm like wow, this cause most things, really seems to be the right way of doing it. And I also run a I also run a crypt of podcast called epicenter. So that's right. I guess sometimes I just I have this bad happen when I'm on other people's podcast. I just start like butting in an accently, like putting in the host hat myself and studying a questions. So sorry if I do that. So good, so good about you, Brett. Hey, it's it's great to be here. Thanks. Thanks for having us. Yeah, this is Brent. I work on special projects on the Tenement team building on the Cosmos Network. So my background spend more on the Theorem side. So I I joined consensus and the Software Development Company in early two thousand and sixteen and, you know, hung around the office at first, you know, and the early and the early months, and eventually joined on as a as a member of the of the company on May two thousand and sixteen. Interestingly, you know, I had a chance to work with a lot of great people. Starting Out, I was as actually a Joe Lubin's personal assistant, so I got to see a lot of the inner workings of the company and just checking out, you know, how they're growing. Spend a few years there and eventually spend some time with a protocol benchmarking company called White Block, and then I ended up joining the tenderment team and, you know, it's it's been pretty great having the opportunity to see the cosmos ecosystem evolved and developed and all the iterations of the aform. It's just been a pretty tremendous opportunity and really have to be here. It's awesome and there's there's there's a few things we want to talk about. We just got this. It's we got to scratch now. Why do you think Bitcoin is a store of value? However, recent events not changed your mind about that. First of all, its second of all, why would you think that's be good with? Because I just completely do not think that it is anything remotely even close to storm value or even big gold or any of that. I just can't it just doesn't add up. So I'm just curious what led you to that so I could get more information to drive...

...my own decisionmaking process. Yeah, so, you know, I do truly believe in bitcoin being sound money. You know, the economics are in place that you know, whether or not people actually use it as such is a different question, which is sort of what reality showed us. You know, people aren't necessarily using it as a store of value, but you know, as if you are like myself, using it as a store of value and preferring to travel with it and and storing it in a really very convenient way and in a very private way, it's you know, it's it has a lot of advantages. But Um, as we've seen with market, I I belong to the minority of people who are using bitcoin. You know, I after the two thousand and seventeen. I see O boom. We've seen a lot of institutional investors enter the market and so a lot of that has sort of flooded out the more niche subgroup within bitcoin users, which is people who do hoddle, and that's I believe that's part of the complexity of what we're seeing right now play out. But again, looking back at how people have used Bitcoin in times of social so social unrest and economic crisis, like in Cyprus, like in Venezuela, just like in in India when they got rid of load denominational cash, we're I think we're about, you know, several months, years away from having a larger population realize this and possibly flowing into to see seeing bitcoin as the as like a viable alternative to what they've currently have. You know, given that money is being inflated away at a Rabbit Pace, and I've heard that, but it's always it's always like a few months away. It's always been so, and only that. But like the truck directly with the the drop in USD in the stock market, it, like markets tank, bitcoin tank. That's not what a store value would do, and so I'm looking at the market dynamics. Also look at the fact that, you know, it's by definition, probably not a store value, and you appared to like goal. But either way it's it's really that while you're here. So I just wanted to scratch that Ed it you if I could, just to hear what you have to say on that. But you're here referring cosmos, tenderment, ether mint, IBC, these these these amazing projects that are that are coming out of your ecosystem, and I think there's so much to talk about there. So I want to really focus on that. So first off, let's describe what's the state of the IBC, because that's been one of the major tenants of Cosmos that I am aware of for for a long time now and it's a really difficult problem and I'm kind of curious how progress is coming on that. I last I heard it wasn't out yet. Is That's still true? Yeah, it's not out yet. So what's what's the status on IBC? We're gearing up for IBC launch with the second inside Abis Test Net that that we're we're we're doing. The first one was game mistakes last year and that was sort of the adversarial test net that we had before launching the cosmos hub, and the intent of that was to sort of stress tests the parameters and ensuring that there's, you know, nothing that could be broken before we launched something and we had to revert something later on, and so doing this a second time now. It's called game of...

...zones for pre empting the IBC launch, because we want to see what sort of attacks could happen in a live network where there's multiple chains now sort of transacting with one another. There's a lot of edge cases that are yet have yet to be explored and especially, you know, we're seeing multimillion dollar losses on DFI protocols on a theorem right now, and we kind of want to see we kind of want to prevent that from happening because, you know, once defy solutions flow into the cosmos ecosystem and the connected change there in, something like this could possibly happen and we want to mitigate that as much as we can. So yeah, so basically, another way to put it is we basically have tho like test net stage. So of right now, and so in the next couple weeks we'll see a lot of the projects who are building on Cosmos. We have like we're going on causes STK are going to be like how connect we're going to be creating like a parallel cosmos network, basically connecting a bunch of people's test nets together. So so the IBC, it's interbloshing communication is based. Is what it stands for, correct, and that acronym down right? Yeah. So, okay, so it's essentially it's a tomic commits across to independent chains. Or is it not? Atomic commits, like what? You shook your head. So I'm wondering, like what is Oh, is it's more it's more like a side chains right where all it's doing is it's sort of of is it's a base layer. What it's doing is it's just sending proofs of state between two chains. And then what we have on top of that is that what we called ICS has, which that for interchange standards, and so those are sort of these high level part like, you know, you could think of IBC as like TCPIP, but on top of that you need htp or SMTP or no FTP whatever, all these like high level protcols. So then you have these an ICS that are high level protocols. You could have one for the token transfers, could have one for like you know, evm, smart contract call. So you can do these different ics. Is Out that you can do stuff on top of it. But the base, fundamental layer is just this idea of doing proofs of state to each other. So the the the proofs of state or predefined in, I guess, the main cosmos hub. Or is this something that gives to Reagi, or can you actually declare what the proof of state is? Register that, yeah, create any sort of asset or any chain you want on your own in your own zone, I guess right. And then that would transfer out of your zone and into the main network, or at least prove to the name network that you possessed this as that correct. So I mean what what your chain kind of does is we have when you're when two chains as supporting IBC like first make a connection to each other, they basically make they have what's called channels, and so each channel is for a specific interchange standard. So when you do this handshake process, you're basically telling the other chain like, okay, we're using this channel. You can think of this like when I connect my my port to your the channels are kind of like ports on a server, right, and so we connect these sports. We're saying, okay, we're going to be talking this ics on this port, and so when they do a handshake, they kind of explain to each other these are the ICS has that I know and can speak and say. There's also the same with, like you know, the ICS is. So you can define different consensus protocols to each other. So let's say I I have a tenderment based chain and you have an avalanche based chain. I would need to have an ICS if I want to connect directly to your chain using my chain. I would need to support the ICS. That like explains to me how to do a like client proof to verify state proofs of your chain. What was the designed decision behind this? Why couldn't you just have people create arbitrary transaction types and have them be able to just validate whether or...

...not these transactions were were these decisions they were making were falling any sort of state rules? Is possibly what what it? What is the I guess it's because you're using proofs. So be is this correct? Because you have to register the proof from, you know, these spokes to the center, like the hub, then then basically these they have to kind of check in every once in a while. Is that correct? So, for not, like there's the hub. Isn't really important here, right, like sure, the hobb is. You can ignore that. IBC is sort of just a point, two point thing it does, okay, allows to change, to talk to each other. So change, they can talk to each other because they speak the same language. And the proof itself has to be kind of defined in this was you say, I see us interchange that you could think of them sort of like RC's right on a theory on where, you know, we just need a common definition of that okay, we're all going to use this standard for how to because basically, yeah, you like an ICS for token transurs would basically be like this is what the packet format would look like for a coken tracer. Will have these fields that this is a destination address, this is the amount to be encoded like this. So that's what sort of with ICS and capsulate, it's like basically defining these standards so that way different change can implement handlers for the for the same standards. Okay, okay, got it, and then enables these assets to be they're not actually like changing. Are they exchanging between two chains? They're not like you're just sit telling the other chain that this person does possess this Asst. is that correct or that? I mean that's like sort of up to the ICS. But yes, for the can transfer as juand that we're using. Yes, I mean what it will do is, well, what how token transfer IICs works is we take a token, we lock it in a module or, you know, you can give them a smart contract. We lock it into a special smart contract and that smart contract in state. It puts it outgoing packet. Right. It says like, okay, here's an ice here's an ICS twenty packet out going for this chain. Then you have what what called a relayer who will go ahead and relay that like basically take a like client proof of that packet and put and submitted to the other chain as a transaction, and then it will be able to verify that packet and then it'll mint you a new coin on that chain. So when you say proof, what is the nature of this proof? Is it like a mercal route or like's what's the actual yeah, yeah, it's like a it's like a it's like client proof. Right, total contain the proof of the consensus. Like that, this is a valid block. So Kennerman has a pretty you know, easy like to use a header verification. So so that's so, like you know, you and then you would have the merkle proof showing that this piece of state is in the chain. There is a in the IBC standard. There is a spot that you can also do validity proofs. So currently we don't do this right now, but in the future maybe, like you know, you could also just include a snark proof that it's also like, you know, kind of like quite codec kind of does. So you can just do a smartproof showing like hey, look, not only is this like you don't have to trust only the consensus protocol, you can also, because right now you kind of have to trust the validators of the other chain. Have that all the proper state transitions. If you have validity proofs that, you don't have to do that either. Got You. Okay. So the end result of this is basically some chain, a user at a dress a on, you know, chain, one wants to speak to once, to send an asset to user to address on chain. You know. Yeah, chain to to just be a chain to they stay, prepare a proof using because they know already in advance at these two chains have compatible. I see us right. So they're they are compatible with each other. They both support each other's,...

...you know, standard. Then they know they can make a commitment to that chain by preparing what validity? Yeah, just stay proves a stay proof to and send it as a transaction. Is just a separate transaction or or just like a different kind of method of providing this information. It's just a transaction to the chain, which then puts it in the chain, validates it and then that enables now does that chain, the validators of that chain. So they have the validator set. You know about a hundred, two hundred people on that chain? I guess, yes, they would. They would be able to validate that this is chain to mind. You that chain one is valid. They need to jump in the validator set there. How would they? I'm not quite sure how, without the full state, or at least a current state, of chain one, you could actually know for certain that the transition transaction that was submitted a chain to was accurate. You were because it has you have to submit the blockheaders as well. Oh, you did say. Okay, okay, okay, so you submit the block setters. So basically it's an on chain SPV client rather than an on chain the you know, it would not be scalable for the validators of chain to to be running a full node of chain one, but essentially what's going to be happening is in the state machine of chain to, there's a spb client of chain one. Okay, I get it, I get it. Okay, cool, all right, that makes sense. Cool, NEAT. And this is this is currently on in testing right now, so it's actually something we've got and play with today. Yep, very cool, very cool. Well, and what's the demo look like? You know, I haven't tried it yet. The demo currently is basically, you know this exactly this it allows you to it'll spin up to tenement based chains for you and then it will allow you to run a relayer between them and you can do these sort of a coken transfers between them. Okay, cool, all right, well, that's that's super neat and we do not recommend using this with real value quite yet. But yeah, so what is what is the security? Because concerns about this remaining because this is well so some of the early designs that I heard they've hashtimelock contracts and things like that. You know, it doesn't sound like you've gone with that approach in the end. And you know, running an SPV client on the chain. To me that sounds like it's kind of limit the number of chains that you could register with the system. So you probably have a pretty a reasonably high cap, though, being given that, you know, you could always just relay to really, you know, basically use chains as kind of like their own realisis to which was the original point of the whole design. Anyway, when you when you originally presented it, is that you would still have like all these chains coming off of the, say, the main hub, but they would they would have each themselves also be their own hubs and have the ability to have other chains. So passing things around this network wouldn't be a big deal, except for the fact that I think, what is it, six to seven seconds of time per block, so would add some delay to your transaction getting around with us. Still signific fetly better than, say, the current financial system. So I go, you're saying there. So. So, yeah, this is kind of where the hub comes in, right where the hub, the role of the hub in the no of of the cosmos hub in the network is to basically be this chain who's like jot like all these other chains. They want to use IBC, but they're really meant for some other application. Maybe the decks are, you know, who knows what they're? The point is that the...

...hub is basically an application specific chain whose soul job is to basically optimize to making as many IBC connections as possible. So basically it should support like every possible ics so can like talk to us many different consensus protocols as possible. And so, basically, that way, if you have a chain you're like, that's doing. You know, you're building a deck. You don't have to constantly be upgrading your chain to support every single interchain standard because you know, maybe there's all sorts of you consensus protocols being built. Instead, you just have to connect to the hub and the hub can act as a translator between different chains that want to use that service. There's nothing about IDC that requires you to use the hub, but it's there as a service to other chains. Got You? Okay. So let's say one of these chains was running either mint and so either men is not a bridge to the main atherium network on its own. Correct. It is merely just a vm essential, a evum that will run us most network and it's a it. Basically, it's just does smart contracts. It will let you handle this the similar functionality that's available in the theium presently. Yep, is this? So this is entirely correct. There's nothing different about it. It'll still handle normal eath calls. It's kind of like what we did right Avo with Atherium, where you can still make straight up like eth calls to the the system and it'll it'll completely work as this correct. Yeah, basically, how would you transfer tokens, it would only be the main root etherium token. Correct, that could transfer to another chain using IBC. No, because we could have a so either mind can basically it's so there's ICS for token transfers and ethermid could basically choose to how to adapt to that. Its own statementsion to that. To that I see. So first what that means is, yes, if you it could use it. Could make sort of a special contract that acts as a an evm smart contract that act as an ICS crapper. So you send her, you can make it so you sent arc twenty tokens to that contract and it wraps the and it basically still creates an ics packet for them, just as it would for the native token as well. So what kind of challenges did you face? Are you actually using straight gath or like? How did you actually build either mint like? Is this like a Unique v RMBER like? What did you do there? So so just a heads up that that part, that that has happened. We have not done that yet. Where well, their token transfer like the USC S T I s to like creating outgoing IDC packet for that. But yeah, I mean I can tell you a little bit about Ethermin in general, though. So what we did there was you wanted a version of evm running on a fast finality chain based on tenderment core, and two or three years ago we had an original version of Eitherman, which is basically we took goetherium and slapped it right on top of tenderment core, and that worked quite well. But the problem was it we spent a lot of time, like writing the proof stake system and everything in the cosmos sdk. Well, maybe I guess we should have talked about the cosmstk a little bit first. Go for yeah, so the causes. S Dk was basically, you know, a framework to make it easy to write state machines. So, you know, it's a modular framework. It's all written, it's written go and so you can write your own state machine and there's these modules that you can write. We're like, you know, you can import function the kind of like, you know, a ruby on rail's ecosystem, where like there's open source modules and you can pull in whatever you need for your chain. So we spent a lot of time building like staking modules and governance modules and token transfer modules and all these like modules.

That, which is how we built the Cosmos Hobbit, which is what most of the change on cosmos are using, the cause Kj, and so, you know, we're like man the ritual vers in either mint. If we have. We spent all this time writing staking modules and stuff and governance modules and we can't really reuse that on on this original version of weatherman because it's not using cons Dk. So what we did was we basically decided to scrap that original version and we started, you know, we call it Etherman two point o, where what we did was we decided just to build an evm module in the Cosmos S dk, and that way it can still use all of the staking stuff and governance stuff as well. And so that's what the current version of either mint is. Let's just say. Yeah, I don't know. So actually our Ted Mint, I'm sorry at Ted Ted Yan over here. We actually we didn't. We just got it their stuff and W U making goetherium work fine with our staking but I don't know why. That would be a little different in our system than yours. But it is possible for for go. But I would agree that just taking a plane Jain evm would make things lighter. It would probably make things faster than just including the entire kitchen sing in and whatever you do. Said, think that's a good design decision. What made you choose the EVM, other than, you know, the smart contract support? What specifically about the EVM WE INTERESTED IN? Mostly just ANUC effects that it already has, like a large number of people who want to make that. I mean there's even to today there's this like a very high demand for a faster finality evm. A lot of people who have spent, you know, years and years developing stuff on top of the EVM and in slidity. They don't want to like rewrite all their stuff, so they but they want to be able to port it to something that uses tenderment. So that kind of that was kind of what it was. I mean, at the same time, there's a lot of there's a lot of experimentation being done with newer and probably better smart contracting systems and Cosmos as well. So we have this we have this project called cousin wis on, which is just like smart contracting system, like quizm based contracting. There's the is the agoric team. They've done a lot of like object capabilities based smart contracting, which was pretty cool yeah, there's something. There's some other ones. Dan Know what's funded by the inner chain foundation to launch packed, their smart contracting language that's compatible with tournament based trains. Yeah, actually forces, you know. Yeah, our our approach is like, you know, get as many smart contracting systems here and see which one's work and which one people want to use. Totally that makes sense. Hey, you guys, ever noticed that you'll be having an inperson conversation about flame, not cheetos or cats or something or emus, and you don't even own a cat. You might own an EMU, but the next thing you know on facebook you start getting ads for cat litter emugates. It sucks. You should be paranoid. Big Centralized Tech companies make money from your data and they feed it back to you. You are the product. You're not buying anything, you're buying more of the things that you do. So, but guess what? Now you can take some control back with the status chat APP. The status APP lets you chat, transact and use web three dapps. Status is decentralize, so even the messaging runs on the encrypted whisper procol. Status is also crazy about privacy and security, so you know your data and your identity are safe and if you...

...want to verify that, they open source everything they build, so you can hack away at it yourself and find out why. It's the most secure distributed chat APP. If you couldn't tell, this is a sponsored content that status wrote. They paid us to say the last few things, and this too, status is transparent as a core principle. So they thought it would be right to tell you that and it's meant to give me them exposure get you to go download their chat APP from the APP store. Or status that I am kid, and also helps the bigcoin podcast network keep running. So we'll try it out. Download it, get set up and find our public jet channel, Hashtag TVP, and we'll give out a few tokens to let you try out all the features. Just had the status that I am get to get started one last time. Status that I am kid. What is Peggy exactly, and is it going to enable me to actually get real etherium onto this or is that something totally different? Yeah, so peggy is basically it's sort of like a hack of like, you know, getting IBC onto a theorium is kind of expensive because just it doesn't support the right pre compiles and stuff that we need. Right now. It's just too expensive to verify certain pedotypes that we that we need, and so peggy is sort of just a hack around it until we get like a proper IBC, impromtation on the theorem where the validators of a causemost STK chain. We have a control of multi say, on a theorium, which is not how like IVCC is supposed to really work, like the validators of a chain. Aren't us? We really run than anything on that other chain. But this is sort of just it's a hack to make it, you know, to get us interacting with the theoreum faster than it would take to write a proper IBC. So so there's a single multi sick contract which would reside on a theorium. That would enable so what would you do? Lock the Tokens in a in that contract and release tokens of that contract to enable assets to float in and out. But what is views as an oracle system for pricing between because I would assume that whatever you're doing would need to kind of keep your I guess you wouldn't actually, because you're literally going to be transferring an etherium onto a rapt and basically it's not wrapped, but you like an actual like lock to theorium onto say just chain, to or whatever, some some chain, and then that that etherium is a promise essentially on the main etherium chain. And so you can base this is this is correct. Okay, cool. So I'm guessing. I'm guessing. So I'm just like, yeah, it's you basically promising on the main etherium chain. The multisig contract, I guess, is like what, you know, like seven of ten or something like that, something where you really have a lot of, you know, people on the Supposic Sig. That sixty seven up a hundred. Perfect. Okay, that make that's exact. So that's that actually makes a lot of sense. So that's a that's a that's a full corum of a validator. So you paid, you put a hundred. Every validator puts a signal, gets a signature space and in the Multi Sitcom. Okay, cool, that makes a lot of sense. Unfortunately, I feel like I skipped a lot of details for the crowd, so maybe you could, you could explain to them a little bit why you chose sixty seven and and what's going on. Oh, I mean, basically, we just wanted so that it's like sixty seven have the same security assumptions as the chain itself, which is what IBC does right, and IBC, you're basically the light client. The security of an IBC connection is based off of this, you know whatever. The security of that other chain that you're interacting with is same thing here, where we just wanted to make it...

...like, you know, at least two thirds of the valid it is have to sign off on any peggy transactions as well, right on the mix. Perfect Sense. So what what is the expected use case of this? So I actually go why would I want to take my theory and put it onto cosmos and have a promise when I feel like that promise could get how confident are you in the for instance, the multisee contract that you're using? A lot of these contracts have been compromised before. Hell, it was parities own contract back in, you know, July, two thousand and seventeen that I got that got their multi Sike took out a lot of big players bank or got hit really hard by this one, for instance. You know, the these slidity smart contracts are colosely under attack, and yours would be a very high value target. What are you doing to secure these contracts? Are you sure that this is like are you like how simple is this contract, for instance, because if you get really complex, this gets really hairy really quick. So, like what's going on there with the security side of that? Yeah, so this is kind of why we wanted to start with peggy rather than a full blown IBC implementation, because a full going IBC implementation insolidity is like, you know, we'll be probably way more complex. It has to like understand how IBC connections work and the handshake and the channels and all that stuff, which is why we'd like to you know, that's why I was starting with Peggy, which is actually a pretty simplistic version of a peg. And Yeah, you're right, even that could be complex. You know, you think a multi sake is probably a pretty simple contract and you know, as you mentioned, you know we still messed up on those as well. So do you honest? We haven't got into the auditing stage of of this yet, but yeah, I guess we're just really kind of go for a very simple contract and just getting as many eyes on it as possible. There's a there's a there's a company in the cosmos ecosystem called informal, and they're really focused on like formal verification of things, especially of rust, and so, you know, maybe eventually might be able to get them to like see how they can help with this kind of stuff. But yeah, I don't really have a particularly interesting example for you that like that like, you know, the same as what everyone elsing yes and forget audits and hoping up Bas got you. Given all of these abilities, I do seem endurance market starting pop up. Mean, yeah, what? What? Why? When? They because, you know, the mark we've shown that we can't guarantee against these things. You know, there's just going to be some multimon multimillion dollar exploits. Yeah, like two of my ex roommates, they actually they created this project called open which is sort of uses a put options to do insurance on them. So yeah, I mean I think people, maybe people who use Peggy, might want to be able, might want to compose that with like taking out some sort of insurance thing on on chain. I was I was tiding with someone a few weeks ago. I'd be like we were talking about like trustlessness in blockchains and we were just talking about like how, like you can imagine. So, you know, companies like you know, let's take traditional financial systems, have like ninety nine point nine percent like security, right, and then we go to them and be like no, look, look how what trustless we are. We have like ninety nine point ninety nine percent security, and they're like I mean, and there was much like well, you know, who cares? We're insured in this case and we're not in short on your system. So because we're in short, this is what's trust less to us, not yours them. Yes, truth, truth, well, they're very skilled at hedging their bets on in that space. Anyway. The I will say that their security requirements tend to be a lot more stringent than even blockchain provide. You Frank with you there. So one of...

...the problems that a lot of these people are facing is just getting into, getting past compliance and regulatate regulation between such as KYC. It's also anti money anyone anti money laundering is an L and all these other weird compliance laws. You don't want to get fins and knock out on your door. Has Cosmos had given any thoughts to approaching these major organizations, because you did just bring them up. So Your Wall Street bankers, your your you know your Nas dex as this world. Have you approached them? What are you guys trying to apply with that? I know that's a little outside of the scope of the tip of the engineering on this but because you brought it up, I'm just kind of curious. What's going on in that that rain. Yeah, Brapper, think this. Sure. Yeah, happy, happy gives an input here now, I think. I think in the our initial philosophy right now is to focus on open source development and making sure that the protocol solid and, you know, we're creating good, useful software and things like Peggy, IBC ether ment are built and know, proper standards and iterated in the right ways. When it comes to, you know, more traditional systems, interacting of Wolfe thanks, interacting with, you know, traditional financial situations, you know we have the capabilities, but I think we're still kind of focusing more on just hardening the protocol and making it battle hardened and useful before being applied to pridginal system because, you know, many ways in this with this technology, were also trying to displace or replicate some of those systems. So I think one of my favorite experiences in this ECO system was as actually working with Sonny at a at a hackathon for at eath Waterloo, and basically, I don't know if sunny wants to tell the rest of the story, but you know, we we, we we wanted to create a side chain and plumentation that we thought that could be useful for as to a certain extent. Son If you want to do want to finish the story? Yeah, sure. So this is sort of a little bit of a troll project, but realize it could actually potentially be kind of useful, which is this is like joke that's like, Oh, you know, most se twenty is have the like centralized and stuff. You might as well just put them on Google sheets. So we made actually a google we made google sheets as a side chain to a theory m where you good use you can transfer token off up from a from atherium on onto a special google sheet that's like right protected, but then you can make transactions on that Google sheet and they're like verified with like the alot signatures, and then when you want to move back to the theorium, you can withdraw them and you have to do like an alot signature, which is like Ursa. And then we were able to verify that on atherium. It's expensive, but you know you can do it. Yeah, so I thought the future that was what was the part of bring that up front. What was the relation here? Oh, and I guess we end up calling it a sheet coin. So that does yeah, as absolute. But for the reason I brought that up is that you know that that in a relationship between the theorem blockchain and there's Google sheet side chain is the fundamental component of how you can replicate a traditional banking system like that. That relationship where you have value that exist in one essential ones ecosystem that can be manipulated and interacted within a fast finality ECOS or just google sheets up at the time, is indicative of what block chain is able to do in a broader ecosystem of you know, having having good financial value and Inter interactivity. So yeah, I...

...think. I think all that stuff in, you know, definitely in the future, though. You know, we're sticking to our guns here and we're making sure that we know, build all this stuff really well before we bring it bring it on to the bigger institutions of use and you know, and many ways we want this to be used for everyone. You know, want everyone the access to be able to access this. So we're still figuring out but a lot of excitement ahead. It was totally kind of curious. So, getting into the doubtailing off the financial institutions in the mainstream stuff, I knew that finance uses the d the SDK pretty heavily. Are they running a private version of this or do they are they fully attached to public cosmos hub and what's I actually don't know. It's what can you tell me, like what is there architecture? Like? Are they still running? Are they running the most up to take version? Like, what does finance look like from a cosmos standpoint? Yeah, so they're usually COSSS DK. They're not connected to the CONSTOS help. That's because no one is connected to the COSSOS UB right now. Like I said, like you know, IBC is not live yet and on any magnets, but from our discussions with that they seem pretty open to like wanting to like start integrating with IBC once it's like, you know, sort of rugging and has been test Gad. So yeah, I think that they they want to be able to use IBC, especially they want to stay. Also, they haven't want to use like Peggy and stuff because they want to like connect to bitcoin and etherium. So yeah, that's something that they're interesting to it. And the problem is that, like you mentioned, their codebase is actually private, so we don't know exactly what version of the COLSGOSS DK there there. You work wrong. But they've actually done a lot of great help in the ecosystem. They built a lot of like you know, they built a lot of like good javascript libraries for interacting with consoles, sdk based chains, and so they've been, you know, they've been contributing that, which is kind of Nice. Oh, that's cool. So I'm actually you said something that I didn't pick up on earlier and I apologize for that. The IBC is necessary for so I know that there's a lot of different validators out there, like a different chains and stuff running, running on Cosmos and I've heard about anycotally mostly. Are they so they're not able to connect with each other, register with each other in any sort of way right now. Is that? That's what's going on? So they're all literally just disparate chains. In the dream here is to take this, take IBC out of Beta and have them all connect to each other so they could float assets between each other. About how many of these disparate chains, disparate chains, are there that you know of right now? If you would ask me so, one that are built with the Cosmos S DK? I might have to guess off the top of my head, maybe around like five to ten of them which are on Mainnet, and then maybe about fifty to a hundred that are like in development, that people are building what's supposedly interesting chain that you've seen so far. Like what's what's kind of jumping out at you at this point? I guess some of that. I'm interested. Yeah, I mean I'm kind of I'm pretty interested in this. Some cool stuff. I mean this project called Kabba. They're kind of working on a version of maker sort of built on the COSMOSSDK, and so they have like a whole multipilateral system working and stuff, which is pretty cool and it's not life. I'm made, not yet, but they're working on that. And there's a project called, I'm really a big fan of agoric, and they're like smart contracting system that they're working on. SHAPESHIFT is working on this project called micro kick, which is sort of a interesting way of doing price oracles using short term...

...futures tradings. So yeah, there's something cool like defy stuff that's sort of being built right now. I don't train on grank. What do you guys think? Um, I like Iris there. You know, they've been sort of our sister projects since early two thousand and sixteen, two thousand and seventeen, and so they've been there with us since the beginning and they're building a lot of micro services on their chain, particularly with their their partners in China. So I'm curious to see what sort of interesting applications come out of that. Yeah, additionally, I mean defy. There's definitely a lot going on these days and the overall ecosystem. So you know we in the cosmos ecosystem, we open source to decks around St Blotching Week last year, basically utilizing a batch auction mechanism for facilitating trades as opposed to a central limit order book mechanism. You know, essentially what we want to do of this was figure out way to prevent sniping or a type of mechanical front running that exist and different sensualizing changes. So you know, this is a great experiment. We think that some sort of Peggy integrit so you know at the moment that the decks is only able to trade etherium, but we think that a peggy implementation connecting between a probabilistic finality blotching like a theorem, and past maality watching my cosmos integrated into the decks. You know, we could trade between a cosmos and Cosmos Adam and etherium pairs and yeah, you know, it's just in defy. In general, we see a lot of movement that that could intertwine with some developments that's happening at an art ecosystem. There's another trend that I've been noticing, which is people are trying to build on sensorable social media and so, you know, people are looking at okay, how do we build decentralized twitter and alternatives in that space? And so there's there's there's another project called desmos and they're built by, you know, the team validators called forrble, and they're based in Hong Kong. They're also trying to decentralize the social networking space, creating, yeah, creating a new social network. I haven't tried it yet, but that's you know, that's that's also in testament and you know, there's there's another one that we're building in house, which is a project within the Virgo project. It's called dither and, and that one trying to build a wallet that is associated with like a like a twitter handle, like COSMO's wallet that is associated with a twitter handle, and so you could do social networking while transacting as a social network. I was going to say the best Cyce to look forward if you go on a cite called COSMO knocks dot world. That's why it's just great site that kind of like has a list of all the projects building on Cosmos on like you know what stage they're at and I make sure the good hub and everything. So it's part of the best resource likes see more. So, yeah, I like that one because it's open source. And then anybody who happens to be building on causes, STK or Tenement is just creating a poor request to say, Hey, AD me here, we're currently building this. You know this is this is like we're testing head or we're about to go to mainnet. And again, yeah, that's a good resource, nice resource for this eat causes ecosystem. So I'm I've got one question that is kind of I keep hearing about...

...play a game mistakes and I'm not really clear on what that was structured like. I missed it. A kind of like passed past me and I was then I heard about it. Like I said, I'm not in the ecosystem, but I I'm associated with with with it, and I hear about it and it's typically like a minute too late. So can can you explain game mistakes? And now you've got this other thing, game of through zones. It's some kind of curious like what are these things? Can you describe them for the audience? So game mistakes was the first incentive ice test net of its kind. The the conversation that sort of preceded that was about, okay, you know, we have our idea is we have this ecosystem of sovereign chains and each of them is going to have to boot trap its security one way or another. And right now you know, so this is this is us in in two thousand and seventeen we're like, okay, had we have a handful of validators in our ecosystem, maybe five tops. How do we incentivizes validators to want to join our test net and actually want to try and break it? And so this idea of having this adversarial test net came up and and further to to incentive eyes even more players to come in, because we had a hundred slots to fill before we wanted to launch the hub, and so this this idea of attaching like a like a like a leaderboard came up, and that's basically how we came up with game of stakes. And it was like a multi month long, like marketing communications process. But yeah, after it was launched we ended up getting nearly all a hundred validator spot slots filled, and so that was the initial goal and it kind of you. Because of the success of game of stakes, it demonstrated to the industry that this is a good way for a lot of proof of stake networks to to like stress test their systems, because we don't know, you know, it's such a stake, is so new. So you do need something like an incentivised test net. And so game of zones is this year's version of game of stakes, and game of zones is our tempt to stress tests and ecosystem where there are multiple block chains, because in cosmos a zone is just an application specific block chain and so nothing IBC has permission list. There's nothing like stopping. If I if might change supports IBC in yours does they will. Just you know, you don't get this all. A governance vote required to approve an IBC connection. You any chain can kind of just like connect automatically, and so this might create like adversarial systems that we at the time what we want to test as well here. Sweet. Yeah, chains could just connect automatically to these. So explain that a little bit more. What would that process look like? They just go ahead and say, Hey, validators, I want to connect to you or like how does this like? What does it look to they even care if somebody connects to them? Like, should that be an issue? Know, the validators don't care because it just sort of something on the state machine. Right. I did. I just make a transaction saying look, here's the chain ID and some initial route of trust, whether it's a genesis file or some some route of trust, and you just say, okay, here are connect to this. You know, start tracking an eye. You know someone has been running a relayer for it, so maybe I'll be run the relayer or something, you know, and you know, as long as I'm just relaying packets between the two chain like relaying headers between the two chains, then you know that's an IBC connection. Dutch it. But you said earlier that the the number of BBC connections could kind of fill up, like there's a there's a limit to the number...

...that you can theoretically support. So is that a possible like attack factor for somebody if they just want to launch a bunch of machines and then try to Smith that a transaction or with the network just rejected and say I'm full, which is another kind of attack factor? I mean it will kind of just there's a gas like you know, submitting, submitting IBC connection like had ours and stuff has a gas cost associated with it. So yes, they could do that and they would end up a spamming the chain and having to pay a lot of gas to do that. Yes, I but mean eventually do we have to consider? Okay, what happens if they spam so much that like they fill up the data back, you know, the database of all about like now that the nodesag yeah, yes, that's a concern. That's like a larger concern. You could do that just by scamming normal transaction, right, and yeah, that's something. Yeah, but I thought the actual notes themselves had asked. Act is sort of like a light client to the the secondary chain, right. Did I misunderstand that? No, the light client is in the state machine. So you can consider it like there's a smart contract that is running the light client. Okay, okay, okay, that makes sense. Wow, that's actually pretty cool. So the smart contract. There's a smart contract in chain to which would wait? We know. Actually, how does that work? I'm not sure. Okay, so's there's an example. Are you from a have you ever in the old etherium project called btcu relate? Long time ago? Don't remember it that well, but I thought it was just rap BTC. I mean. So what BTC relay was during was it was just you know, you can't. So anyone can basically take the current Bitcoin blockheader and submit it to a smart contract on otherium and then, if they're smart contract, would verify the proof of work saying okay, yes, this is a valid hash of the previous block block and know it meets the difficulty threshold, and Yada, Yada, Yana. So that's kind of what IBC is. Basically the same thing, but in a much more generalized way where it doesn't count specifically for BBC, but it kind of works with any chain that also supports IBC. But like, so the point is that you know, the validators aren't run I mean in a in a way, technically, yes, the valiators are running it, because you know, the running the state machine. But it's like, you know, you wouldn't say that in the BTC case. You wouldn't say that the miners are the one running the BTC like client. Just the BTC, like client, is a contract on the evm and the miners are just running the evm. Right. But I thought, okay, so this is this is part. So you're committing every single blockhead or too to a contract and then put you don't have the ability to dwell, I guess you do to verify the proof of work for that block ahead. So okay, but then you have to have every single one kind of leading up to that. Wouldn't you need to keep a state history with the district? Because, like, I could just smell anything and just say this works and it should be fun. Right. So this is the tendament. The way tenderment, like clients work, is it? You know, they're pretty efficient in the sense that you don't have to transmit that many headers, because you can actually do this sort of as long as the ballotador set isn't changing by too much every block, like as long as the vallidator set hasn't changed by more than a third, you can actually skip a bunch of blocks. So I could say, look, here's the vout if the ballotist. Yeah, let's say Browgers, ABC and D are. The valigers are block ten and then a block hundred. They're still the same validators. You actually don't the way tenament, like clients work. You don't actually have to transmit the blocks in the middle. Just go ahead. Okay, how's that? How's IT rolled up? Like? How do you know that that's a that that's, you know, block one hundred, is a lot. Is is in fact block one hundred, and it is a valid block between because, well, you know, it's block hundred, because that information is in the blockheader. Yes, and you don't valid...

...block and pull it out and replace it with another number. So I can be like this is block because the header has the signatures of the validators on it. Okay, yeah, Oh, okay, I got it. Okay, cool. Ah. Yeah, sorry, maybe that wasn't clear enough. Yeah, that makes perfect sense. I should have thought again. Cool, okay, cool, no need. So. So really, you guys are about to really unleash the big, big, big, big unveiling. What's IBC guts going and what is your like a projected timeline and actually producing a fully fledged live ib like IBC protocol, the whole whole Shabang? What's going on? You step back really far with I do engineering. I don't do, you know, planning on timelines. Well, so, to give you an Amn game of zones is going to be launched. The start date is may first. It was just announced and the registrations are open. And as soon as that concludes, then there's going to be further testing and tweaking of the parameters for before we launch IBC and then we have to have like a governance vote on the hub and everything so to get the activated on the hub. So that's that's a multi month long process, but you know it's probably going to come before coronavirus ends. So there you go. That's a good so next week? Cool. Yeah, really. So what's kind of you excited in the space? It doesn't have to be cosmos really to what's really, what's really what you guys like looking at. What are you interested in? Oh God, I'm the biggest change hill there ever was. I just felt we just in it. We just released the podcast on that today. I would yeah, and I am very like last week's recording was was with handshake, blanking on his name, so terrible like that. It was a Tastian. Yeah, Tatian, and he was a really he was really good grasped. I'm really excited about that protocol as well. It's super cool. I want to steal it put on the AVA network. It was amazing, but it was. It was. It was definitely a really, really, really really interesting podcast. So to show my own stuff. You're listening to this for Cosmos. Listen to previous episode. Of all, for handshake. Jack is absolutely correct. It's amazing. What else you guys interested and I saw sonny seemed to have something. He was at you wanted to say. I'm really interested in sort of any a lot of the web of trust base things. I just think that I don't know, I had this like realization a few months ago that like trustlessness is a silly thing and because it's not worth building trust systems, because that's how the world is. The world has like a lay fabric of trust, and what we should be doing, is what tothology should be doing, is like leveraging existing trust relationships to make interaction that made you worn't possible, be for now possible. And so to that and I've been really interested in things like into ledger, which is just like payment protocol. I've been interested in sort of how to do some undercloudalized lending stuff using web of trust. Oh, I actually had this cool idea for how to do a web of trust based consensus protocol that is sort of a fork of avalanche, and so, because, and so, yeah, that's kind of one of the things I'm kind of interested in and maybe now that you guys have your code open source, I might actually start, might actually fork it. It's just like try testing, you not to see if it works. Well, you know, have the forky you can just take what you can fork the repoke like, yeah, you can just build your own...

...vms. Anyway, we'll talk off like the plays of the VM. The problem is what I want to do is not use proof of steak. I think you can do sort of that's a combination of stellar with avalanche. Oh, okay, I see, I see, that's an interesting choice. But Anyway, okay, cool, cool bred. What do you into these days? Sure, yeah, I got my eye on some of these steaking derivatives come out from and proved to stake ecosystems. Some good stuff coming from Kepler, excuse me, the Evert Evert Protocol team in Korea. They're also building a Meta mask, cold one for the the cosmos of the system. I like. I like some of the e two point road map. I know it's it's looking pretty far into the future, five to ten years. You know, that's that's a that's a lot of a lot of future outlook. But you know, it's kind of good seeing things laid out, you know, and you Havn't having some sort of expectations on the on the timeline and just sort of a shameless plug, here we are. We are having a get coin. Happened on starting April twenty of so now we're hoping to get some folks working on on some conmost related projects and definitely on the lookout for folks who are building cool. So, yeah, cool, awesome. Well, guys, thank you very much for coming on the program I'm excited to see IVC rollout. I really am looking forward to seeing how this all kind of plays out and as as things kind of coolsced come together and people start joining hands with each other on these separate block chains. I'm really, really excited to see what that looks like and how people start interacting with each other and a more global scale rather than just have these isolated, you know, single, one off chains finally have an assets flow across them. is going to be really interesting to see, really interesting watch. I'm really I'm really going to be paying attention to seeing how that goes out. So thank you very much for your work. Thank you for coming on the air. I was really great. I really appreciate you guys being here. Thank Youre.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (127)