The Delphi Podcast Host and GP of Delphi Ventures Tom Shaughnessy hosts a galaxy brain group of leading Cosmos builders to discuss the Cosmos Network, value accrual, IBC, comparisons to other Layer-1s and much more.
- Zaki Manian: Co-Founder of IqLusion and SommFinance
- Sunny Aggarwal: Co-Founder of SikkaTech
- Jelena: Working on InformalInc and Cosmos
- Joe Dirtay: Communications at Cosmos
- Ethan: CEO of Informal and Co-Founder of Cosmos
- Jack Zampolin: Co-Founder of SommFinance and Pylon Validator
Every Delphi Podcast is dropped first as an audio interview for Delphi Digital Subscribers. Our members also have access to full interview transcripts. Join today to get our interviews, first.
- Zaki Manian: Co-Founder of IqLusion and SommFinance
- Sunny Aggarwal: Co-Founder of SikkaTech
- Jelena: Working on InformalInc and Cosmos
- Joe Dirtay: Communications at Cosmos
- Ethan: CEO of Informal and Co-Founder of Cosmos
- Jack Zampolin: Co-Founder of SommFinance and Pylon Validator
- Tom’s Twitter: https://twitter.com/Shaughnessy119
- Delphi Podcast Twitter: https://twitter.com/PodcastDelphi
- Tom’s Clubhouse: https://www.joinclubhouse.com/@toms119
- Our Video interviews Can Be Viewed Here: https://www.youtube.com/channel/UC9Yy99ZlQIX9-PdG_xHj43Q
- Access Delphi’s Research Here: http://members.delphidigital.io/
- Cosmos by From The Dust | https://soundcloud.com/ftdmusic
- Music promoted by https://www.free-stock-music.com
- Creative Commons Attribution 3.0 Unported License
(1:41) – (First Question) Joe’s Intro.
(2:23) – Jelena’s Intro.
(3:08) – Zaki’s Intro.
(3:18) – Jack’s Intro.
(3:29) – Sunny’s Intro.
(3:48) – Ethan’s Intro.
(4:43) – Cosmos Elevator Pitch.
(5:02) – What is different about Cosmos.
(8:19) – Insights about Cosmos Security / Benefits / Overpay.
(20:30) – How does the Validator side work.
(24:37) – What is the Inter-Blockchain Communications (IBC) protocol / How does it further the ecosystem of application-specific blockchains built on Cosmos.
(30:38) – Thoughts on handling a supreme ccale to run a decentralized exchange on Cosmos.
(33:59) – Why is Cosmos designed the way it is.
(36:19) – Does it really matter if people or app chains have to choose their security to pay less if they could just pay such a small amount on Ethereum.
(46:30) – Thoughts on the Value Accrual side with Cosmos and ATOM.
(53:18) – Who is Jack.
(53:55) – Who is Ethan.
(54:43) – Who is Jelena.
(54:53) – Who is Zaki.
(55:48) – Who is Sunny.
(56:25) – Who is Joe.
Hey, everyone. Welcome to Delphi’s Clubhouse. Today we have on quite the powerhouse of Cosmos stars, pun intended. We’re going to discuss everything that has to do with Cosmos. Pretty amped to dive in. We have a lot of people on the pod, so we’ll start clockwise. Joe, why don’t you introduce yourself and we’ll go to Jelena after you?
All right. Well, my internet name is Joe Dirte. I started as just a Cosmos community member and my role I guess you could say just grew over time, and then I got involved in a community and on-chain, governance-funded marketing campaign for Adam. That’s how I got in contact with Tom and that’s why this is happening. So, thank you.
You got involved in a multisig that changed your life. Is that what you’re saying, Joe?
More or less. Yeah. My first DAO experience has been very interesting.
Joe’s just winging it at this point. Here he is. Jelena, you’re up next.
Awesome. Yeah. Joe, thank you so much for putting this together. You’ve been crushing it with the Prop 34 stuff. Hi, everyone. My name’s Jelena. I work at Informal Systems which is a core contributor to the Cosmos project with Ethan who’s here in the room. Before that I worked with a couple other layer one blockchain projects. Most recently the Celo project, and then before that Dfinity, which is about to have their [inaudible 00:01:29] and that’s really exciting to see.
Before that I was actually working as a lobbyist here in Canada. I’m in Toronto and excited to be here to talk about Cosmos, IBC and all the fun stuff around ATOM. Thank, Tom.
Yeah, thanks for joining. Zaki, you’re up.
Unmute myself. Here, I’m ready to go. Hi, I’m Zaki. I’ve been a contributor to Cosmos basically forever. Here to hang out and talk about ATOM.
Hell yeah. Glad to have you. Jack, you’re up.
Hey, everyone. My name’s Jack Zampolin. I’m also a Cosmos person. Been doing core dev since about 2018 and working with Zaki on a new project now.
Yeah, that’s my name, Sunny. Been working on Cosmos for the last four years. Used to be at Tendermint and I run a validator called Sikka. Working on a project now called Osmosis and have my hands juggling many different things within the Cosmos ecosystem.
That’s awesome. Ethan, you’re last up.
Hi all, Ethan here. Co-founder of Cosmos. Today, now CEO of Informal Systems which is continuing to do core R&D in the Cosmos ecosystem with a focus on formal verification. We’re also running the Cephalopod Equipment Validator, so check that out if you’re looking for a place to delegate some ATOMs.
Yeah. Also Pylon validator. I need to get better at chilling my validator.
We’ll link to all of the [crosstalk 00:03:03]. We’ll definitely link to all of them in the show notes. I’ll have to use some randomization or alphabetize them or something so I don’t play favorites.
Who has the best pitch-
We’re renaming to [Asikka 00:03:16] by the way. Just heads up.
Oh wow. Okay. You went there then.
So we’re first alphabetically.
That’s a good way to play it. So, who has the best 30-second pitch on what Cosmos is? I’m sure you guys all have your own, but who has it?
Ethan or Jack should go for it [crosstalk 00:03:43].
Well, the five-second pitch is it’s the internet of blockchain [inaudible 00:03:46].
The same sort of permissionless network structure that the network has, but applied to a network of connected blockchains.
If somebody’s brand new to Cosmos, they’re a huge ETH bull, they’re a huge Bitcoin bull, what do you tell them is different about Cosmos?
I think the best way to phrase it is the blockchain ecosystem started as all these sovereign chains with each of them with their own applications. You had Bitcoin doing money, you had Namecoin doing BNS, you had Sia doing storage. But these were all siloed. You couldn’t use… At the end of the day, everyone wants to use Bitcoin for everything, right? It’s like, I want to use Bitcoin to buy my DNS’ name. I want to use Bitcoin to pay for storage. I don’t want to use all these random other coins.
Ethereum came along and kind of solved that, but it came with all these… It did it by putting everything on one blockchain, which comes with all of its own problems. So Cosmos is basically saying hey, how can we get all those composability benefits that Ethereum provided but still have not the drawbacks that Ethereum came with? I guess we can go into those drawbacks. Mostly around scalability, technical, social scalability. Things like that.
Yeah. Let’s dive right into it and let’s keep this illustrator for people. If all of DeFi was launched on Cosmos right now, what exactly would that look like? Would every app have its own chain? Would they all have the same security level? How would they all talk to each other? Everyone knows Ethereum. Most know Cosmos, but how do we put that into fine-grain terms for people?
I’ll take a crack at that. If all of DeFi was running on top of Cosmos right now, you would have chains that specialized in being DEXs. You would have chains that specialize, like what Sunny’s working on or the Gravity DEX that’s coming to the Cosmos hub. You would have chains that specialize in lending. You would have chains that specialize in stable coins. You might have a futures chain or an options chain, each running their own [inaudible 00:06:10]. You would imagine that all of these things would be specialized and there would be both token transfers, which is what we call ICS20 which is a standard for moving tokens between blockchains, but there also could be more application-specific protocols or things like, you could have orders and exchanges actually happening on the IBC layer and yet to be invented protocols for things like sticking derivatives and asset collateralization across chains, could all be running on top of IBC protocols.
When we talk about all of this stuff and if you’ve been watching the blockchain space, this actually seems like the way the world is going anyway. More and more application-specific blockchains are coming. Compound is launching their own. Lots of projects are launching Cosmos SDK chains. The Cosmos vision of what DeFi looks like does seem to be the inevitable future of DeFi anyways.
That’s awesome. Really brings in to [inaudible 00:07:19]. I guess the next question on this is, every app can have its own application-specific blockchain and they all get to pick their own level of security, whereas on Ethereum you have this one slate of security, but that has its own trade-offs. What are the differences on being able to choose your own level of security and how does a developer actually do that? Sorry for not pointing out people to ask. I’m just lobbing these out there.
Yeah. Bucky, do you want to cover shared security a little bit?
Sure. I’ll just say that these are pretty early days for that security model. Right now it’s a matter of, there are a set of I don’t know, maybe 100 or a few hundred professional validators out there who are looking for chains to validate on and participate in their fundraiser to get tokens and then to help secure the network. The model Cosmos has gone for initially is yeah, this very separate, every chain for itself to find its own validator set ideally to draw it from the community of supporters that care about that particular application it’s hosting and its token and its tokenomics and so on.
We’re seeing already a lot of the validator companies are running across multiple chains and we’re already, there’s ongoing R&D efforts to take advantage of that and to reduce the overhead for new applications to source validator sets and to create more of a direct marketplace so that this is actually more of a formalized process and isn’t this ad hoc thing that’s happening in the background. So that there would actually be explicit protocols for discovering validators and attracting them to validate your app. But we’re also working on shared security protocols to make it so that you actually don’t necessarily need to find a separate validator set. You can leverage validators from the Cosmos hub and have that be part of your chain launch, let’s say.
This is going to evolve considerably over the year and the next few years as the validator marketplaces start to emerge, but I think in the longer term, the essence is to reduce the layer of abstraction between infrastructure and applications because so much gets lost when you abstract the way the infrastructure. It’s just like some anonymous cloud of miners or Jeff Bezos’ empire. So the goal is to bring that infrastructure closer to the users and surface it into the applications so that it’s something you actually are supposed to care about.
A lot of people think that’s a bad thing and it just puts more burden on users to think about things they don’t care about. But that’s how you solve the critical problems we have in the world that are brought about by the layer of abstraction between infrastructure and applications. If users aren’t given the option to reckon with it, then nothing’s going to change and we’re going to have the same kind of technology empires that we created in Web 2.0 and Web 3.0, and that’s something we don’t want.
If I can jump in there and add some color to what Ethan said [inaudible 00:10:43] super-important about shared security or what we’re calling interchain staking in the Cosmos world, the idea is that Cosmos have validators, will be able to opt in to validating chains that come online. Chains with the biggest foreign market cap. Maybe not as recognized in the marketplace, but the beautiful thing is it’s still opt in. I think that really speaks to the philosophy here on Cosmos, which is this idea of sovereignty.
There are chains out there that exist sovereign independent of the Cosmos hub. The extent to which they want to interoperate with the hub is totally choose your own adventure. I think that’s something that, to Tom’s questions earlier, sort of comparing it to Ethereum or with the other chains out in the ecosystem I think is really distinct.
I think one of the reasons I joined to this ecosystem is usually we think of economics generally in everyday life a something jurisdictionally defined. We’ve got a Canadian economy, we have a European economy, we have an American economy. But now I think with Cosmos, the vision that Cosmos is offering to the world is you can have a digital economy where the building blocks of your economy are preset. You can again choose your own adventure. You can leverage Tendermint. For consensus you can leverage the SDK to build this application-specific blockchain. Again, there’s going to be other use cases within Cosmos that are coming up very shortly like shared security to make sure that your economy is secured in however you want to define it. I think that’s incredibly powerful that Cosmos is I think a huge innovation on top of Ethereum and other chains from a sovereignty perspective. That’s what I wanted to add.
Yeah. Jelena, I think those are really great points. Towards the end you were talking about this comparison between Cosmos and Ethereum. While I think that Tom, you talked at the beginning about how all applications on Ethereum kind of share the same security level, at a base database and chain level yes they do, but at an application level, each application is generally governed by some kind of multisig. That level of security can vary quite widely across applications, whereas right now in the Cosmos ecosystem and even moving forward with shared security, all chains will have this validator set security assumption, which in some ways for a lot of applications is going to end up being better than multisig. That is one thing that we missed that I kind of wanted to point out.
Jack, that’s awesome. Jelena, great comments, and Ethan. Just to ask again on that second point, is what you’re saying that the validators for the specific application-specific blockchain, would that have a link to the token holders for that chain per se? What exactly is the link there?
That’s a great question. There is a direct link between the voting power on the chain and the token holders. It’s a staking-based system, so when you’re a user and you want to help secure the network and there are staking incentives for this, you would stake with one of the validators that’s helping run the chain. The power that each of those validators has in securing the chain is based on the amount of stake that’s sitting on top of them, and they also have the influence in governance as well.
There’s some proposal that I think Sunny has to change that which we could talk about, but that’s roughly how it work-
That’s awesome. Another dumb question, but hell, you guys are the experts. What do you get from-
No dumb questions.
I appreciate it. What do you get from having this tunable security level? How exactly does that benefit different applications? Let’s say I’m building a MakerDAO, but on the other side let’s say I’m building a high-performance perpetual app chain. What are the differences you get here on security, on speed, on not overpaying for security? How does that all shake out?
I’ll just make one quick comment about it and then I’ll let someone else take over. One thing that I really like about it is that it’s not one size fits all, and not everything meets the same settlement assurances. Base-layer blockchain obviously needs super-high settlement assurances, but if you’re trading NFTs or cryptokitties or something, maybe you don’t need that same level of settlement assurance. So you don’t have to pay as much for your security. You maybe let’s say, you only have to rent $5 million of security out, where if you’re settling massive financial transactions you might need to rent $5 billion of security. I like how it’s flexible in that sense. That was just my one comment.
No, I agree, Joe. Shoot, Jack.
It makes it easy to scale the security with your application. I think that that’s actually something that people want. This idea of sovereignty is something that comes up within the Cosmos ecosystem. When you’re running on top of a platform, you lose that in some ways. The reason people are okay with that on these large common platforms like Ethereum is that they’re really decidedly neutral. I know Sunny has some thoughts about this as well, but there’s a lot of applications and a huge application of the Cosmos SDK has ended up becoming exchange chains. Those exchanges want to run on their own infrastructure. They want to have their own security assumptions and run their own systems but still be able to operate with the rest of the crypto world, and that’s really what Cosmos offers.
I think Zaki and I were talking about this a couple weeks ago, but I think a lot of the times when it comes to security and throughput, we’re compared to actually layer two’s [inaudible 00:16:36] system like a Matic for example. Keen to hear what others on the call have to say about this, but from my understanding because of the sovereign nature of Cosmos, a chain within the ecosystem doesn’t have to worry that other chains will become super popular and that they’ll make their app or whatever unusable because there’s that sovereign inoperable nature between chains vs. a layer two, like let’s say a Matic, somewhat relies still on the base layer Ethereum in terms of gas prices and things like this.
Obviously I think the layer two ecosystem is highly complicated so I don’t want to make any generalizations necessarily, but there is definitely that distinction between Cosmos and let’s say the layer two ecosystem in Ethereum.
Right, but it’s generally true that by separating it out the way Cosmos does into independent validator sets, the apps aren’t competing for the same kind of resource. They are still competing for a kind of resource which is validators in the first place, but that’s much more flexible and elastic let’s say, than the block space of a single L1 like Ethereum. So, whereas all the apps on Ethereum are competing for the same block space and they’re limited by the gas cap and how much each app is willing to pay, if you’re a small app and you can’t justify paying hundreds of dollars in fees and the fees are hundreds of dollars in Ethereum, well there’s nothing you can do. Right? So the idea of having one size fits all, it kind of doesn’t. That’s really one of the key problems that Cosmos is trying to solve by saying well, you can avoid that. Yes there’s a cost to avoiding that which is you have to figure out your own validator set, but the benefit is then you’re not subject to the political-economic conditions of this thing that’s much larger than you that you can’t control. Maybe you want to be subject to that, maybe you want to be part of that, but unless you’re one of the top applications, it’s going to be hard to justify the economic costs and the fees your users are going to have to pay.
That’s awesome. Two followup questions for Ethan and Jelena, whoever wants to take this. If you get to not overpay for security as an app, is the main thing that users are going to see is just a lower transaction cost? Then, I guess my followup question is, from the validator perspective, if I start a new application is there a standby list of, “Hey, you know what? I need 10 validators,” or, “Hey, I need j100”? Or is it more opt in, like they have to convince the validators to use their chain? How does that side work?
There’s a lot of flexibility. One of the main things is they’re going to see lower fees, but they’re also going to see a validator set that’s more actually invested in their application. Like, Ethereum can make governance decisions. Ethereum miners can make governance decisions that don’t necessarily take into account the applications. So if you’re an application, the Ethereum system might upgrade in some way that screws you or breaks all your smart contracts. This has happened in the past to a few big Ethereum applications.
Whereas the validator set of an application-specific chain is specifically invested in that application, otherwise they wouldn’t do it because that’s the only application running there. So if they’re making upgrades, the upgrades are in the interest of the application and presumably the application’s users and valid user are trying to get more users to pay more fees and so on.
I’m sorry. I forgot the second question.
No [crosstalk 00:20:01].
I want to talk about the second question.
I have a lot of opinions about this. If you look at the historical life cycle of a Cosmos blockchain… This is sort of pre-IBC and pre-Gravity DEX, which is the last two years. Basically what would happen is someone decides they want to build an application-specific Cosmos chain, like let’s say Kava or Akash. They would basically recreate in micro the process that we did for ATOM when we launched the Cosmos chain. They would typically build a blockchain with the Cosmos SDK, recruit validators, sell tokens to validators, recruit delegation, bootstrap their chain, get exchanges to list them, get the token to circulate, and then the token would have value. We’ve seen more than a dozen chains go through the same process.
IBC really changes that, specifically the rise of DEXs within the Cosmos ecosystem really. When you connect your blockchain to the IBC network, no one knows or potentially really cares if all you’re doing is you and your friends are running a couple of nodes in your basements. The Cosmos network doesn’t care. IBC will connect to anything. IBC will connect to a single Tendermint chain that is running on your laptop. All of these things are, it’s the classic internet, no one knows you’re a dog model of just trustless interconnectivity.
This creates an entirely different way for chains to bootstrap themselves into existence, recruit validators, accrue value in the token. Staking tokens for new chains on Cosmos will just be available on DEXs for any validator to buy and become part of networks that are interested in them. This entire bootstrapping process is likely to change dramatically basically in the next month.
That’s awesome, Zaki. Great caller. I guess one of my last questions on the security side of things, are there any issues or attack vectors with having differing levels of security for each application? Let’s say I have two DeFi apps and I borrow money on one, I send it to another but the first app is easier to attack, the second app is harder to attack. I guess I’m a little lost there because I don’t know if the amount at stake was in the specific app scales with how many validators they have, is my question.
Kind of goes back to that comment I said earlier about settlement assurances. Not everybody needs the same settlement assurances but let’s just say you have a chain that all the tokens combined are worth $50 million. If someone tries to send in $100 million of ATOM, those validators if they were being rational actors and just using game theory, they’re better off stealing all of that ATOM than they are continuing the life of their chain. So the higher amount of value that is like skin in the game and staked, the higher the settlement assurances are for those transactions, because if those validators lie, cheat or steal, they have way more skin in the game to get slashed.
Also… Joe, that’s a great point. Going back to my comment earlier about how there’s a different level of security for apps on Ethereum based on the multisigs, I think we’ve seen over and over again what happens when you have interoperability between applications with different levels of security. There are issues where people walk away with money. We call it rug pulling in Ethereum. That’s the way it manifested [inaudible 00:23:56]. I think validators had security and then as we roll out shared security, it actually offers better assumptions for a lot of apps.
That’s awesome. All right, guys. Oops, sorry. I was muted there, guys. That’s awesome. Let’s switch over a bit to IBC. What exactly is the inter-blockchain communications protocol, and how does it further the ecosystem of application-specific blockchains built on Cosmos?
IBC is basically this generalized protocol that will allow any two blockchains to talk to each other. You can think of it, here’s the analogy we like to use. It’s like the TCP/IP of blockchains where TCP/IP was this stack that you implemented in all of these different OS’s and then they’re going to be able to talk to each other and communicate. IBC actually really refers to different stacks in a way. I felt like lower-level IBC and higher-level IBC. The lower level is what allows two chains to talk to each other. You can implement this just like you implement TCP/IP in different OS’s. You implement the IBC protocol in different blockchain development framework. Currently it’s implemented in the Cosmos SDK, but there’s work being done on implementing it in all sorts of frameworks like Substrate. I know people are working on adding it to Celo. I know there’s some work about how to add it to Solana. This allows any two chains that support this IBC protocol to be able to talk to each other.
Then once you have that, you need the higher-level IBC protocols, which is like, what are we talking about? This is like in TCP/IP. You need these higher-level protocols on top of it, such as FTP for file transfer protocol or SMTP for mail or HTTP for web. You need these protocols as well in the IBC world. Like Zaki mentioned, there’s the ICS20 which is for token transfers. There’s other ICS’s being developed as well. ICS stands for interchange standards. There’s things for NFT transfers. There’ll be things for cross-chain like data, like Oracle data, or cross-chain EVM calls. Over time we’re going to see development of more and more of these higher-level protocols built on top of IBC as well that will enable us to do much more complicated tasks throughout the interchain other than just token transfers.
That’s awesome, Sunny. Another naïve question, but just to frame this. If you’re on Ethereum, you don’t need something like IBC. Right? Because all the apps could speak to each other natively. But obviously you don’t get the trade-offs that we discussed here.
That’s right. On Ethereum you have the inherent ability for contracts to call other contracts, and then they’ve built standards on top of that. There’s the call opp code, which allows one contract to call any other contract that’s built into EVM. Then there are the conventions that are built on top like ERC20 and 721 and all the other conventions for how to structure calls between the contracts.
That’s awesome. I know IBC’s just launched, I know it’s been in the making for a while, but how do I get as a developer that automated ability to use IBC? If I build a chain on Cosmos, I’m assuming using the Cosmos SDK, do I automatically have compatibility with other chains? How does that all work?
Yeah, it’s out of the box.
Oh, baby. That’s easy.
I would encourage people to look at [inaudible 00:27:47] today to see all the beautiful connections being made in Cosmos with IBC. [inaudible 00:27:54] chains are already connected with IBC. There’s also the Cosmos Hub, Akash, IRIS. Many more to come.
Also, one thing I want to note about [inaudible 00:28:05], not a whole ton of traffic on there right now. I expect to see a lot more pickup with the DEX testnet. People will be testing the interchain swap capabilities of the DEX. Then again when the DEX is live, we should see a lot more activity on IBC.
I’m on here now. The list is awesome. Spoke to Akash’s CEO the other day so it’s cool to see them on there too. This is cool, guys.
Yeah, he’s awesome. Just to circle back, is IBC fully launched today and all its features and functionality? Or is it still early stages? Am I late here?
The version we have today which is version 1.0 has all of the core functionality and ICS20 which is the token transfer. So right now the only application-level thing you can do with IBC is transfer tokens from one chain to another unidirectionally. Which is basically the core side-chain functionality. Side chains effectively from 2014, that is at the heart of the IBC mechanism that allows you to send tokens from one chain on to another that may have a different security model.
Over the coming months and years, there’ll be a variety of additional application, IBC-based protocol applications that allow you to do more between two chains.
For example, one that will probably come around this summer is what we call interchain accounts, which effectively allows you rather than having to transfer tokens to another chain to be able to use the functionality on that chain, you can just use the functionality on that chain from the chain you’re already on. So you can basically have a wallet on a chain B that is controlled by chain A. So that’ll reduce the number of transfers and make all of the functionality on other chains available on one chain.
Further down the road we’ll also have interchain staking, which is the Cosmos version of shared security, which is also built on IBC. This idea for IBC was to have this very general-purpose communication protocol above which we could build all the more advanced functionality. The different kinds of sharding, integrations with rollups and other kinds of applications for more advanced communication between chains, essentially.
That makes sense. My other question there is, there’s obviously a huge swirl of events going on at Ethereum right now with layers twos and then other chains which really didn’t solve much. Just centralized, like financed more chain. A lot of the stuff going on in layer twos and Ethereum comes into throughput for security. A lot of the side chains obviously are to secure. A lot of the stuff with ZK rollups is really secure because you’re linked to the base chain. How do you guys handle supreme scale in that sense? Is it just a matter of choosing less validators for more scale for your app chain? How do you get that scale to run a decentralized [PRX 00:31:03] exchange or a decentralized exchange on Cosmos?
Oh. There we go. Could you go into that a little bit, Jack?
Yeah. Absolutely. I feel like I’m going to butcher this, so Zaki or Ethan please step in here. What LazyLedger basically is is it’s going to be, they’re making some improvements to the Cosmos SDK to allow this type of shared security base chain. It splits the consensus-critical stuff and then the data availability over into a separate system. That allows for arbitrarily scalable Cosmos chains.
How did I do on that?
Reasonably. LazyLedger’s an optimistic rollup solution, so those are also being used to scale Ethereum. LazyLedger is building their optimistic rollup solution using Tendermint and the Cosmos SDK so that rather than just being for EVM applications it would be for Cosmos SDK applications. There are trade-offs between optimistic rollups and ZK rollups, but the optimistic rollups are coming to Cosmos.
For scalability right now, the core Tendermint consensus software can already process. It’s already much more scalable than what you get out of Ethereum, partially because the FT consensus is faster than the proof-of-work consensus, but also because you’re writing native go applications rather than things that are compiled down to a virtual machine. A software virtual machine like the EVM. So there’s inherently a lot more vertical scaling already.
The current Cosmos hub has a latency of about seven seconds or so, which is somewhat artificial. That could be improved depending on how aggressive you want to be or how you want to tweak things. We could get into why that is, but in principle Tendermint has been tested to be able to do sub-second latency with tens of validators, so if you want to have 64 validators or something like that and still have a couple-second latency that’s still possible. With that, you should be able to do hundreds if not thousands of transactions per second.
There’s also a lot of improvements that are coming in terms of things like aggregate signatures that will allow us to reduce the amount of data that has to be transferred between validators, and that will also help reduce the latency. So there’s a bunch of scalability improvements that will come over the next couple years that will allow validator sets to grow to hundreds if not thousands without too much impact.
That’s awesome, Ethan. I won’t take credit for this next question. It’s obviously crowdsourced from the group here, but I’ll throw this one to Sunny. When you look at Cosmos, what is the why for all this? Why is Cosmos designed the way it is vs. your peers?
I think one of the main differences or the reasons why it’s designed this way is, I think there’s a pretty common design philosophy that’s throughout the Cosmos ecosystem which is a heavy focus on sovereign or [heterogenous 00:34:23] systems. Like, really believing in the ability for communities to be the atomic unit in a lot of these things.
We believe that communities have the ability to build their own chain and be able to secure their own chain and not need some globalized system to force it. This comes through the design of the Cosmos ecosystem as a whole with all these sovereign application-specific chains, but I think the same design paradigm flows through all sorts of different things within the Cosmos ecosystem. Even the design for shared security. Our design is way more bottom up than the more top-down scaling solutions such as… or sharding solutions. If you look at things like Polkadot or E 2.0, they’re very top down in the sense where they’re like, “Okay. Here’s this global system. We want to make sure everything is homogenous security-wise across the entire thing, and so we’re going to build this system to assign validators to different places,” where it’s like, no. Maybe if we’re okay with more heterogeneous things, it’s more bottom up, more Hayekian where it’s like, let the individual incentives drive the system towards allocative efficiency.
Then you see there’s some things like, I think the design of for example how the Akash network works, with how you choose which sort of nodes you want hosting your applications. This is something I’ve been looking. I am working on a project called Osmosis where we’re applying a lot of that design philosophy towards how to build AMMs.
Sunny, I had one followup question for you. Just to take the devil’s advocate side of this, let’s say… I’m brushing over a lot of technical stuff that I don’t begin to claim an understanding of, but let’s say ZK rollups and all that stuff on Ethereum works. You inherit all the security of the base chain, you have all the scale of using an L2. Does it really matter if people or app chains have to choose their security to pay less if they could just pay such a small amount on Ethereum?
If they can get their full security from the higher chain that’s great. Obviously, there are going to be costs to that. If you are getting into security there inevitably are going to be costs. I think what’s more important than the validator security is more about the community governnace and the ability for a community to do their own upgrades, decide when to do irregular state changes, when to do hard forks. The ability for a community to not be subject to the whims of a larger [inaudible 00:37:24].
An example of this would be in the last Ethereum upgrade, not the one that just happened last week but the one from a couple months ago, there was some changes in the pricing of certain opp codes and it broke a bunch of contracts. Most notably the Aragon contracts. The Aragon team was reasonably pretty upset about that because it’s like, “Hey, there was this higher-level chain that we didn’t consent to that was pushed at the Ethereum-wide level that broke our entire system.” We’re going to see this happen especially when the storage refunds are going to get removed. Who even knows how many contracts that’s going to end up breaking?
That’s I think what it’s more about. It’s the ability to not have things be changed from under you.
Sunny, that’s a great point. This idea of sovereignty is kind of core to what Cosmos does. When Sunny is talking about bottom up vs. top down, that is really deep in our DNA, is the ecosystem. If you listened, Bucky has this amazing talk on sustainability and sustainable biofeedback loops within different communities and how to bootstrap a community with this kind of technology. I’ve always thought of putting a validator set together as a barn raising. It’s a group of people getting together to build something on top of all this cryptography and it’s really cool. Anyway.
That was a slightly unfocused thought, but Sunny you’re absolutely spot on there.
To your point, yeah. The ZK rollups, I think if they inherit the security of a more secure chain like Ethereum or the Cosmos hub that’s great, but I still believe that application-specific ZK rollups are going to end up winning out over these [inaudible 00:39:25].
One misnomer I see in the L2 community for Ethereum is I get this notion that everyone thinks L2 scaling is going to come by having one optimism chain that’s going to run 10X the speed of Ethereum and ta-dah, we solved Ethereum scaling and we’re just going to shove all of DeFi on that one optimism chain. I think that’s pretty misguided. I do think that that’s not going to really solve the fundamental problems and we really do need to move towards a more application-specific L2 world in order to actually solve real scaling problems.
There isn’t one overarching government in the world. There is not going to be one chain to rule them all. This is one fallacy that I think we’ve run into in crypto over and over again, and for a long time that’s been Bitcoin. In this compute world people thought it was Ethereum. What Cosmos does is just acknowledges the reality that there is not going to be one chain to run everything. There’s going to be a lot of different chains and they need to talk.
I can jump in here. I think Sunny’s way of describing this kind of bottom design process is also mirrored on the human side. On the very human, organizational governance side of things. A lot of the times we’re talking about highly technical concepts. Layer twos, scaling solutions and so forth. But we forget that technical systems often mirror human systems and vice versa. If I can speak a little bit about the human system behind Cosmos, it’s an ecosystem of many entities. I don’t even want to give a number. All these entities are constantly trying to coordinate. Constantly trying to come together, set agendas, set priorities, come up with a road map. Oftentimes it’s hard because it is a bottom-up grassroots effort a lot of the times. These are anchors of the ecosystems that do drive funding and initiatives like the interchain foundation naturally, but at the same time the success is reliant on this self-organizing principles that is integral to the ethos of Cosmos.
A lot of the times, again, in crypto and blockchains we talk a lot about highly technical computer science topics and theories, but we should also really pay attention to the human governance behind all of this because it will to some degree dictate the way that this technology is adopted and by whom and for what purpose.
That was awesome, Jelena. Just to round out the example, can you guys give me an example of… You don’t have to name names or projects. That’s not why we’re here. I’m just trying to get a little more illustrative with it. Can you give me an example of maybe an issue that could happen in another chain that wouldn’t here? Like the buy-in of the community? I just think of Ethereum with all the GPUs and the miners vs. the actual community and it’s all stupid, but just any specific example would be really helpful there.
I think the example that Sunny gave and that I mentioned earlier of Aragon, their application breaking without their permission wouldn’t happen on an application-specific Cosmos chain because it would be the thing coordinating the upgrade. When Ethereum upgrades and it breaks one of the biggest apps and that app had no say in the matter, that’s lack of sovereignty. That’s one of the key things that Cosmos is prioritizing for.
I just want to go a little bit more finer grained on that issue. Ethereum from their point of view was solving a denial of service vector from mispriced opp codes. They were like, “We have mispriced opp codes. If somebody wants to they can cheaply slow down the entire chain. This is bad for everyone.” But a specific application which was like [inaudible 00:43:34] DAO contracts was like, “Well, our application depends on the gas price of certain oppp codes being within a certain range.” The Ethereum [core devs 00:43:49] rightly for Ethereum prioritized the consensus security of the entire chain over a specific application, but in the Cosmos model people are allowed to make a lot more subtle and interesting trade-offs where they’re like, “Well, we are not writing a general-purpose application. Maybe a gas mispricing issue or how a transaction is structured.”
Another example of this is on Terra. In Terra, all validators also run Oracles. They run price Oracles for the stablecoins that are on the Terra network. The fees for the Terra Oracle messages are like, there’s one fee level for validator messages because you have to be a validator in the validator set. You have to have a huge amount of stake staked in order to be a Terra validator. You don’t have to worry about spam really from the vaildator set. So you can charge specific application-specific fees so that the stablecoin remains stable and the Terra network can prioritize that.
If the people who are using Terraswap are paying more fees but the validators for it take less fees to run the consensus messages, that makes entirely sense. So again, the Terra dev team is able to make application-specific trade-offs across the entire economy of the system.
Another thing that would also happen is if there was a security vulnerability again related to a gas mispricing issue on Terra and that didn’t affect any other chain on Cosmos, it all just kept trucking along, this is again default isolation that we want on the Cosmos ecosystem. A single application developer who has a problem on a single application chain doesn’t affect the entire system.
Related to that is at the end of the day I think everything comes with trade-offs. When you’re building a chain, you’re making a trade-off between scalability and decentralization, to an extent. Or oftentimes you make a trade-off between security and ease of use. Trying to find a one size fits them all is not a great strategy. It just doesn’t work at the end of the day. The point of the Cosmos model is that says every application has different needs for where they want to be on different trade-off [inaudible 00:46:26] and so we should allow them all to explore and choose the best trade-offs for their application rather than being subject to trying to find one size fits them all.
That was awesome, guys. I want to spend the last 10 minutes to talk about value accrual. I’ve been talking to Joe a lot on the side about this after I saw the Messari graphic. Value accrual is something we spend a lot of time on on our venture side, so it’s something we’re keenly interested in and we want to make sure the value created, there’s a way to capture that. How do you guys think through that with the ATOM token itself? Obviously I don’t want any investment figures thrown around or anything like that, that’s not why we’re here, but how do you think through the value accrual side of things with Cosmos and ATOM?
I think the main value accrual is going to be through the shared security. I think that the Cosmos hub is going to be the best place within the Cosmos ecosystem to be able to lease security from. Having ATOMs is what gives you the right to be part of this… You can think of it almost like a union of all these validators and stake. Like okay, here’s $5 billion that’s available for leasing this much amount of security, and you can earn. When a chain wants to lease that security it will be paying the ATOM of all those. It could be paying them in ATOMs. It could be paying them in the chain’s own native tokens, but you need ATOMs to be able to earn all those rewards and get the opportunity to validate all these chains throughout the Cosmos ecosystem.
Makes a lot of sense. Is there any way for other holders in the future? Obviously a lot of projects have straight-up staking and things like that. Do you guys envision any type of fee flow from those transfers or anything there to be able to flow back to the [crosstalk 00:48:29]?
Definitely. It’s already the case that fees get paid to ATOM holders. I’ve always found the value-capture obsession of people in the space a little bit backwards because it’s like hey, build the thing and before anyone uses it make sure you’re capturing value. It’s like, make a useful thing and you’ll figure out value capture as you go. I don’t know if any of us are paying for Clubhouse but Clubhouse is certainly taking that model of build something useful, get users, and you’ll figure out the value story. That’s always the approach we took with Cosmos. It was like, let’s build the most useful thing at the time and when value capture, quote unquote, is something “useful,” that’s when we’ll focus on it. But in the meantime we’ll focus on adoption and providing value to the world.
I think that’s what we’ve done. Cosmos at this point I think is pretty clearly, if not just arguably, the most popular blockchain development platform after Ethereum. That’s because I think the values we injected into this technology and because we didn’t obsess over capturing value into a single coin and trying to wrap everyone into a single monolithic political economic scheme.
As for the Cosmos Hub and ATOM, the Cosmos Hub hasn’t really had too much of a use case yet. It’s been like just another blockchain. Or a lot of people thought it was, “Oh, it’s some experiment in on-chain governance.” For some time it was. It was really proving out. You’ve got to keep in mind we’re in a different world now than we were two years ago. Two years ago there were hardly any proof-of-stake chains. Now there’s dozens of them live or launching. A lot of ha was because of the proof in the pudding of the Cosmos Hub.
In the world of IBC where now you have all of these different proof-of-stake chains and they’re all going to want to start talking to each other and you’re going to have all these problems service discovery and identity and verification and different kinds of security you’re going to want over certain operations, the Cosmos Hub is the only one that’s specifically targeting the use case of serving all of these other chains, of really being a hub for them. So the expectation is that there is going to be a lot of value flowing through the Cosmos Hub for a variety or reasons. Whether it’s because two chains just want to connect, whether it’s because of shared security, whether it’s because there’s a DEX there and it’s the fastest way to do interchain exchange or something like that. We’re going to see a ton more use cases emerge both for the ATOM token itself in terms of new features on the Cosmos Hub that will require you to have ATOMs to use for instance to open a new pool on the new decks, you’d have to pay a certain number of ATOMs. There’ll be more cases like that.
Also, we’ll expect to see larger fee flow in many different currencies. Not just ATOMs but in other currencies as well through features like shared security and other kinds of interchain routing and so on. It’s very much a, build something valuable and then value capture will come rather than, figure out all the value-capture story up front and then you can delay actually delivering value. Again, we’re that much more bottom-up approach.
One thing I do want to quickly touch on here. Sunny gave us the vision in the far future. Bucky gave us where we came from. Let me talk about where we’re at now. Zaki put out something called ATOM 2021 last year and in that he identified a few key things that we need to do as the hub to capture some value and to move into this next phase of the interchain. One of those is the DEX. We are going to be launching a DEX here on the Cosmos Hub very soon. It’s called the Gravity DEX. It’s been in development for a little while. We’ve got about 15,000 people signed up for the test competition that’s starting in a couple weeks. There’s about 10 hours left to register for that, so if you haven’t done so, go ahead and get out there.
Obviously, being the hub, having a place to exchange value is pretty critical. This DEX is going to be critical to the value capture for hub. Another piece is staking derivatives. Right now all of the capital that’s locked up in staking on the Cosmos, how it can’t be traded, those shares of the staking can’t be moved around. Staking derivatives is a feature to allow you to sell those, move them around, securitize them and offer more financial products on top of those staked ATOMs and offer actually better security for the underlying network.
Finally, another big piece of it is the Gravity bridge, which is what Zaki and I are working on, which is a way to allow Cosmos chains to talk to Ethereum-based chains in a very generalized way. So, think if you’re in main chain, think optimism. Think finance [inaudible 00:53:30] chain. That’ll be coming later this year as well.
That’s awesome. I’m super excited for that. Guys, this was awesome. You packed so much info into 55 minutes with multiple people. This is incredible. Let’s just go again. Jack, we’ll start with you and end with Ethan. Just say your name again, because it’s audio. There’s a lot of people and I want to make sure everyone knows who everyone is. Just chill again real quickly what you’re working on so people can check it out.
For sure. My name’s Jack Zampolin and what I’m working on right now is something called Sommelier Finance. We’re building tooling for liquidity providers over on Ethereum using the Cosmos technology. I’m working on that with Zaki and I’d encourage you guys to go check it out. Go to app.sommelier.finance today. We’re working on something special for you to [inaudible 00:54:17] for you as well, so be on the lookout for that.
That’s awesome. [inaudible 00:54:21]?
I’m Ethan Buchman. I’m at Informal Systems. You can find us, Informal.systems. We run the Cephalopod equipment validator, that’s the little octopus validator if you’re looking for a place to delegate. We built Hermes which is a relayer for IBC. It’s one of the ways that the actual packets get transferred between chains, and we do a lot of protocol design and formal verification work and we’re building tools to make it easier for instance for developers. Not just blockchain developers, any developers to write advanced tests with much smaller amounts of code. Things like that. So we’re trying to commercialize some of the research in correctness assurance and in testing and so on to make software safer overall, both for the blockchain world and beyond.
I love it. We’re a little out of order, but Jelena? You want to go?
Well, Ethan covered it. I work with Ethan every day at Informal, so I think we’re good on that front.
Hell yeah. Zaki? Who doesn’t know Zaki at this point? Sorry to keep [crosstalk 00:55:22] yourself.
Yeah. Jack and I are working together on Sommelier Finance. Jack was highlighting the stuff that we are doing for Ethereum liquidity providers. We are also bringing about both the Gravity DEX to maturity with a bunch of partners like Althea to bring the Gravity DEX to the hub.
One of the other things I’m really excited about is the work we’re doing on the distribution model which is a subcomponent of Cosmos that was one of our core innovations to enable proof of stake, to scale it to millions of delegators. That whole system can be upgraded to enable a whole new generation of DeFIirewards systems and farming and liquidity rewards. All the innovations that we’ve started to see on Ethereum to be really scaled up on Cosmos. So that’s another area that we’re particularly excited to be working on.
I love it. Sunny, last but not least.
Hey, my name’s Sunny and I’m working on a project called Osmosis. It’s going to be the most powerful DEX within the Cosmos ecosystem. Really focusing a lot on especially the [MEB 00:56:29] resistance and preventing front running and things like that. You should check it out. Launching soon. If you had ATOMs on [inaudible 00:56:39] you already have some Osmo token, so get ready to stake and be an LP provider.
I love that. Just to shout out Joe, thanks so much for getting everyone here and organizing this. Really enjoyed everyone’s time today.
Oh, well Joe’s working on something interesting too.
Yeah, Joe shoot, man. Don’t hold back.
I’d just like to thank everyone for coming and joining us here. I want everyone to check out cosmos.network. There’s a sweet new site. You’ll have fun exploring around there.
I think it’s GravityDEX.io where you can get signed up for the trading competition. There’s going to be $200,000 plus in prizes given away. It’s free to enter. You’ll get to play around with the DEX/AMM and not have to pay insane gas fees for it. That’s about it.
Oh yeah. I’m running [crosstalk 00:57:32]. Yeah. I’m running a marketing campaign that the funds were given to us through on chain governance. The ATOM holders voted yes on getting a groundroots organic marketing campaign going. I’m just the plebe that decided to help run it. It’s amazing and I love the Cosmos community. Check it out.
Don’t sell yourself short. You’re crushing it.
Joe, you are. Well guys, thanks everyone for your time. Impactful 60 minutes and I want to respect everyone’s time, so we’ll all talk soon. Thanks again.
Thank you, Tom.