Decentralised.co

Sunny Aggarwal from Osmosis

Decentralised.co

Send us a text

Sunny Aggarwal from Osmosis joins us in this episode to discuss the emergence of L2s, meme-assets, and the future of shared security. Sunny was an early member of the team behind Cosmos and is currently a co-founder of Osmosis, a leading dex. He joins us to share how his career has evolved over the past three cycles, what he has learned along the way, and a few contrarian stances on where the market is headed. 

His takes are backed with substance that comes from operating deep in the industry. Stream for lessons from one of the best in our industry. 

Speaker 1:

Hi, this is the Deco Podcast and I'm your host, saurav. Before we begin, views expressed here are personal opinions. None of this is any kind of advice, let alone financial or legal. It's a conversation about things we find interesting. I'm absolutely thrilled to talk to our guest, sunny Agarwal. Sunny is a co-founder of Osmosis. He was on a tandem team that helped create one of the earliest versions of the proof-of-stake consensus mechanism. He's helped build Cosmos, sdk, ibc and the Cosmos Hub. He's created a company named Sikka that runs validator services for the Cosmos Hub. He's also an advisor to GAWA and Akash Network. It's not very often that I get to talk to people who've worn so many hats, so I'm just going to try to sit back and learn as much as I can during this conversation, and I hope that you, too, learn a few things along the way, like did I miss out anything out of?

Speaker 2:

your introduction.

Speaker 1:

I mean, it's a long list. I've tried to cover as much as I can, but if I've missed out, no, I think that's right.

Speaker 2:

I also helped start this club called blockchain at berkeley. That's how I got my original start in crypto. We're still pretty proud of that one, just because I think that you know, I really like the educational resources we put out. But no, yeah, I think, I think I think you hit most of them you want to start that blockchain at ber.

Speaker 2:

Yeah, sure, this was like years ago. This was, like you know, when I was just first getting into the space, is like 2015, 2016. I get 2015 when I first got into Bitcoin, and then 2016,. Like I helped start this club at Berkeley. I was learning Bitcoin stuff myself.

Speaker 2:

But then it's like what happened was like me and two of my friends at berkeley we were like, why is no one else like learning about bitcoin stuff? This is so the stuff is so interesting. And then we're like, okay, maybe it's because no one understands how it works. And so we decided to teach a class. And at berkeley we had this like cool thing where, like, students could teach courses, and so we we taught a class about, like you know, explaining understanding bitcoin, and like all the lecture materials were recorded and everything.

Speaker 2:

And so I think, like there was a period of like two years, like 2016 to like 2018 uh, where that course was like the best way to learn crypto. I think today there's probably better ways, there's better materials out there, but there's a good two-year period for which, like, I think that was the place, like, if you want to learn crypto, like you go watch those lecture videos. So yeah, you know, and then I started you know from the students of that class I started this like club called blockchain at berkeley, which grew to be like the largest student org. Like we did a lot of like research work and then, like you know, a lot of uh, a lot of cool startups, actually like crypto startups you've probably heard of have like come out of like that blockchain berkeley organization well, and how many of you were running this?

Speaker 2:

the club actually at its peak, had like 60, 70 people, like every semester. The course itself was like run by. Like you know, like I said, it was started by three of us and then we would. What we did was the following I taught it for two semesters. The first semester was just three of us. The next semester it was still the three of us, but then we started having people, like people from the previous year, join as like tas, like helping with like some parts of the course, and then the following semester we had them take over the course and so that course is actually, you know, it's here like six, seven years later still running, because you have this like cycle of like student, past students taking over the instruction of the course now, did it sort of help you in any way?

Speaker 1:

I mean, I'm imagining that if it was berkeley, it was a bunch of smart people who were trying to learn from you.

Speaker 2:

Yeah.

Speaker 1:

Were there moments when you were like I don't know, scared or whatever to you know, try and keep up with them.

Speaker 2:

So like what I mean, like I said when I first. So what happened was when I first got to Berkeley, like the reason I joined the blockchain like community, there was out of all the people I hung out with, they were like I didn't. Oh, they were like talking about stuff and I just like didn't understand it and I was like, uh, definitely a bit overwhelmed because, like the first lecture, the first like club meeting I walked into, they were talking about like monero ring signatures or something, and I was like way over my head. But I was like man, these guys are so smart so I'm gonna like keep showing up because I'll just like learn, like by being around them. And then for me, like teaching the course is like I find the easiest.

Speaker 2:

For me, the best way to learn something is to like sign up to teach it. Uh, and that's why I like teaching the course. It kind of like forced me to learn how to learn this material very well, because it's like one. It gives you a deadline because if you're, if you have to teach it next week, you better learn it this week. But also it's like you know, sometimes you think you understand something and then until you get like a bunch of questions about it, and then you're like wait a second. Maybe I didn't actually understand this as well as I thought I did nice.

Speaker 1:

yeah mean, that's what happens with us when we write something so kind of to an extent I can understand where you come from. Can we start on the Cosmos side by giving sort of a light end to listeners of what Cosmos is like zones and hubs and things that make Cosmos a unique ecosystem?

Speaker 2:

cosmos, so you would need people assistant. Yeah, so cosmos. Basically, the idea is it's unlike most ecosystems that you're familiar with like you know, solana or ethereum, or whatnot like they all have like one chain or one token at the center of it. Like you know, in ethereum, in some cases it's one chain right as a whole. Like solana, like there's no, it's all a single chain system. Ethereum has started to shift to being more of this like multi-chain kind of thing with l2s and whatnot, but at the end of the day, everything is still settling to ethereum and everything is like still trying. They're trying to use eth as, like money and the gas feast.

Speaker 2:

Cosmos, the idea is like no, like you know, there's no one token, one chain at the center of everything. What cosmos is is it's a set of open source tools. That's it, right. It's uh, here's the cosmos sdk, here's tenderman, consensus, ivc, and with these three tools, you can go build your own custom blockchain that's built using the cosmos open source stack. So you know, that's why it's because of the ability to like build using cosmos tooling without you know needing to pay.

Speaker 2:

Like you know, give up your sovereignty to anyone is why you get so many like large projects, uh, using it, you have, like you know, celestia and dydx and thor chain injective osmosis, obviously, um and so if you're trying to build a custom blockchain, like I'll say, most people, like most projects, maybe don't need a custom blockchain. Right, they could be a smart contract or something, but there's a set of applications that you could not build Celestia without a custom blockchain. You couldn't build Doortrain without a custom blockchain. You couldn't build the YDX without a custom blockchain, and so Cosmos SDK is basically the easiest framework for building custom blockchains.

Speaker 1:

Okay, and what about hubs and zones?

Speaker 2:

yeah, uh, so hubs and zones are sort of this like old concept it's not really actively used in in costos, like there's this idea that things are going to be very hub and spoke, but that ended up not like being what? What played out, you know for the better. In my opinion, right, I think, like I said, the point of cosmos was to be anti hub and spoke systems and it's meant to be this like very mesh network sort of thing where there is no you know, or like hubs. Are these like what? What is a hub? Right, a hub is a place where activity happens, right? So you could say cosmos has a lot of hubs, right? Osmosis is a hub of like, of defy activity, right? You have, you know, noble as a hub for, like, usdc routing.

Speaker 2:

There's this one chain called the cosmos hub, which it's unfortunately just like a bad name because, like it doesn't really mean anything, but like it, it it has its own token called Adam. It's trying to provide security, like opt-in security for different chains. So if you want, if you have your own token and you don't have enough security, you can be like hey, I want to get security from Adam, but there's there's a lot of people providing similar stuff as well. So, like you have now ethos, which is providing security for Cosmos chains by ETH restaking, you have Babylonos, which is providing security for cosmos chains by eth restaking. You have babylon, which is providing security by a bitcoin restaking, so you'll have like security.

Speaker 2:

Like you know, cosmos started so far with, like, every chain has its own staking token, but now so we developed a framework called mesh security which will basically enable it. So different, all the chains can sort of provide you can get a chain can have multiple tokens staked to secure that chain. So, you know, osmo says, for example, we currently only get security from Osmo, but if we want more economic security on our chain, we could say oh, we'll use Babylon to get Bitcoin restaking on top of our native Osmo security.

Speaker 1:

Okay, that doesn't mean you share the validator set right?

Speaker 2:

No, it doesn't mean you share the validator set. It means that you have your own validator set, but then people can. So how Cosmos works today is you have this delegation system into the protocol, where you have validators, and then you have people who take their coins and delegate it to the validators and give them voting power. So the idea here is you can have an independent set of validators, but you can still have them chosen via, or they get voting power via, osmo delegated to them or bitcoin delegated to them okay, and how does the security work there in that?

Speaker 2:

yeah. So what you can do is you would give like voting power based off of, like the price of each coin. But you can also build like caps and stuff so you can say, hey, osmosis wants okay, well, actually, here's, here's how it would. Here's how osmosis would probably really go about it. What we would say is we, you know, we'll look at what is the amount of bridged assets, value of all the bridged assets on our chain. So let's say it's $500 million, right. Then what we can do is say, okay, we want to target a 1.5 X security threshold, so we're looking for $750 million staked slashable, slashable stake. And so maybe Osmo's stake is $600 million right now. And so it's like okay, well, we have this $150 million deficit.

Speaker 2:

You could say, and that's where it's like okay, well, maybe that's where we start diverting some of our staking rewards towards restaking, diverting some of our staking rewards towards restaking. We say, hey, bitcoin. Like okay, you know we'll, we'll do like one percent of our staking rewards go to bitcoin and we'll see how much bitcoin's uh restake that brings in. Okay, if that's not enough, we'll increase it to like two percent and then you'll basically have this like controller. That like will adjust it until it gets to the 750 million dollars of stake that we're looking for.

Speaker 1:

Right, that makes sense. Now, since we are talking about mesh security and so on, what do you think of solutions like EigenLayer or Polygon's AgLayer in the context of mesh security?

Speaker 2:

Yeah, I mean I think they are one of the providers in the mesh security right.

Speaker 2:

Like mesh security is like a framework, that of many things all like doing restaking with each other, and eigen layer is just the way you get the eth to be part of the mesh right.

Speaker 2:

In my opinion I don't think people will want to use or at this I'll explain how osmosis thinks about it we would not want to use just eigen layer right like I. I think at the end of the day, you know who do you. You start with keeping all of your staking rewards and revenues for your native token right for osmo. If you're building a military right like which is kind of what building a staking like economic securities is is, you start with your citizens and then you only use mercenaries when you need to right, when you're trying to hit some security target. So that's why it's like, okay, we would use something like eigen layer or babylon and whatnot, but like, just only at the point that we need that final security top up, we would not like ever switch all of our staking like all of our rewards to go to e3 stakers right like what would be the point of that?

Speaker 1:

right makes sense, okay. And when you say that, I I think initially you touched upon the fact that not every application needs to be a chain right. So how do you think of that in context of growth, of cost loss in general?

Speaker 2:

Yeah, I mean, I think what happens is not every project needs to be a chain, but I think the most high value applications need to be a chain. So, especially if you're building like a perp stack or something like that, which I think is one of the most high value applications in all of crypto, to get to where you like the final state of your architecture, I think that's where you need to sort of be an app chain and that's where you've seen a lot of the growth in Cosmos stuff. Right, you have like DYDX and Injective building, perpstexes. Hyperliquid is built using like Cosmos technology as well.

Speaker 2:

And then fun fact actually Polygon is actually also built using Cosmos SDK. They don't actually say it but, like, the Polygon POS chain is actually also a Cosmos SDK chain. But then you have something like Celestia. Right, they needed to make their validators do all this like data availability, sampling and all this kind of stuff that you need to have a custom blockchain for thor chain. They make their validators run like bridges for all of these other chains. So there's like, I think what like? One way you can see it is like, yes, cosmos, it's definitely harder to build applications, but the applications that are built are like the more unique ones, because they're the ones that are making use of that custom technology, the framework.

Speaker 1:

Okay, just quickly. Can you explain why high value applications? I mean it makes sense for them to be a chain and not just an application.

Speaker 2:

Yeah, like I said, you would not be able to build thor chain as a contract because, right, thor chain to work, you need the application, the validators, to run like nodes of bitcoin and of ethereum and all this stuff, so they can do the bridging. But another reason is that, like, you don't want like your application, you don't want it to like have to deal with the throughput like of other applications, right, with your own blockchain, you get all your dedicated throughput for yourself. While you're on a generalized blockchain, some other thing becomes super popular, right, and it takes up all of the block space and now your application suffers because of it.

Speaker 1:

Now your application suffers because of it. So that's why DYDX and Hyperliquid would want to have their own change, because for them, throughput matters. It's one of the most important things for them.

Speaker 2:

Yeah, for them, throughput matters Also. Dydx does stuff like they make uses on the custom functionality as well. So, for example, two things they do is one they make the oracles, they make their validator set, provide the oracle data for their chain, and on every single block. So that way they're getting, like you know, a lot of PerfStacks is on, like Arbitrum and stuff. One of the big challenges they have to deal with how do you deal with chain link oracle upgrade delays? In UIDX this is never a problem because you are from a chains architecture, you're guaranteed that you have an oracle update every single block so that they can do liquidations much more safely. Another example is this has to do with throughput and performance but they actually have their validators all run sort of like you can think of it like an intent matching system where they're matching orders in the mempool and then submitting matched orders on the chain to be settled.

Speaker 1:

So this is like an architecture that like only really works in if you have your own validator set and your own mempool and all of this Okay yeah, that makes sense, like in in this vein, can you talk us through like journey of osmosis from being a dex to a like cross chain, cross roll up communication protocol? Yeah, so we, you know, I mean, I suppose what was it for in the first place that you know that made you go, let's, let's build the dex and then, you know, make it a little bit more general purpose of sorts.

Speaker 2:

So originally our plan was we wanted to work on, like, privacy stuff. That was the goal. And then we were like, ok, what is the product here? Right, I think that people don't pay for privacy on its own. It's like not a compelling privacy is a feature, not a product. And so we're like we're looking at, like, what we can do. And then we thought like, okay, well, dexes seem to be the thing in crypto that has the most product market fit, or exchanges in general, right, like and so like okay, well, let's, let's build a dex and then let's figure out what are the privacy features we want to build.

Speaker 2:

So you know, we started with two things. One is this like concept of encrypted mempools so this is so people can't like front run you by reading your transactions in the mempool. So that's one thing we were working on. And then the other is this like on-chain shielded trading. So you know, today, when you trade on a dext, everyone can see every single trade every person is making, which is like not okay, cause for a lot of traders, they don't want to leak their trading strategies, right, and so being able to, you know, on a centralized exchange, you see all the trades that are happening, but you don't know which trade is coming, came from which person, right, yeah, and so that's kind of the other thing that we're working on no-transcript.

Speaker 2:

I've been like a Bitcoin maxi for like forever. You got the class back, yeah, exactly. And so I always knew that like one day, like the Bitcoin, everything will move to Bitcoin, like that's why that's kind of why I started working on Cosmos initially was I was working at ConsenSys for a little bit on Ethereum stuff and I was like this stuff is all all this stuff is. Ideas are cool, but like what is this ETH shit coin? Like I thought we're building for Bitcoin. And so that's kind of why I joined Cosmos was to build the app layer for Bitcoin, because Bitcoin is an app chain, and my idea was like okay, bitcoin, btc, the asset will flow off of the Bitcoin blockchain onto all of the Cosmos chains and be the money of all these Cosmos chains.

Speaker 2:

Then I think it just took a while for the Bitcoin ecosystem to really wake up. The vitality wasn't there. Then, you know, ordinals came and changed everything, right, and so now it's been like you know there's different terms you bitcoin, renaissance, bitcoin season two, whatever terminology you want to use but like the energy is there and so that's why now osmosis is like starting to make a move towards like okay, with all these bitcoin, l2s and stuff we want to be the dex for for this ecosystem right, how close do you think bitcoin is to actually shipping l2s l2s that we know of, I mean, where you can do general purpose stuff, like yeah, I mean, depends on like what the security level you care about, right, like so, okay, is polygon pos an l2?

Speaker 2:

if yes, then pretty soon, right, but it's like polygon pos doesn't actually get. It's not a real yeah, it's not an L2?. If yes, then pretty soon, right, but it's like Polygon POS doesn't actually get any. It's not an L2. It's not an L2, right, okay, in that case it'll be a while, I think. So there's this thing called BitBM, which is like someone figured out how you can almost achieve like very similar to optimistic roll-up style guarantees on Bitcoin. So, but that I think it'll be a few years before we see those in production.

Speaker 2:

You have things like Babylon, which will give you, like Bitcoin restaked stuff, and then I think they'll actually get to finally start to see some new soft forks in Bitcoin with like op 119, ctv, which is like covenants or cat, and I think these things will like okay. So, basically, like, if you, if your question is like, when will we see a like optimistic roll up level guarantees on Bitcoin? I would say probably like a year or two, but when will you see things starting to be marketed as L2s? I mean, keep in mind, today, even optimism, all the OP stack chains don't even run fraud proofs, either, right, it's all just sort of run fraud proofs, either right, it's all just sort of. So I think, like you know, I think you'll start to see the launch of these things claiming to be l2s in the next like week to three months.

Speaker 1:

There's ones launching like every every few few weeks exactly and you think runes sort of acts as a catalyst for like all these things claiming as l2s, I think runes is just like a new token standard that's bringing excitement, like basically the brc20 token standard.

Speaker 2:

It was just like a very bad token standard and so runes is actually created by the guy who created ordinals and it's just a much better token standard, easier to use, and so you know, I think just easing anything to make development easier will increase it like develop development velocity.

Speaker 1:

Okay, you touched upon ETH Bitcoin, so I just want to get your general view on tokens Like what do you think of that? You know?

Speaker 2:

I think tokens have different purposes, right, I think, like you know, you have money tokens like Bitcoin and probably ETH. At this point, I think ETH has earned its place as a money money-ness asset probably ETH at this point, I think ETH has earned its place as a money moneyness asset. But then you have like tokens like Osmo and DYDX, which are, like you know, they're more like ownership in a like a DAO that, like then you know, has, like it's a business, right, it's right. So there's that. Then you have, like you know, meme coins and stuff like Dogecoin and you all these current meme coins.

Speaker 2:

So, I don't know, I think that, like, the room for money assets is not that big. I don't think there's going to be like hundreds of those, right, right, I, I think there will be like two or three. I think, like, I think it's a very small set the company, like us ones. I think there'll be like hundreds thousands, just like there's thousands of you know, companies and products out there in the world, right, right. And then the meme coins I also think is, actually I think this is an over proliferation of them. I think that, like most, like the staying power, lasting power, will only be in a few of them. So I I like, really like dogecoin and stuff, but, like you know, I don't think there'll be more than like hundreds, thousands of meme coins. Like I don't think this is sustainable. I think, like you know, yeah, like two or three, that's what that will work right.

Speaker 1:

So I mean that's pretty clear that as far as money goes, there may be just two or three, as you say right, and the rest we don't know yet you think we have achieved, like, in terms of what tokens can do?

Speaker 2:

we have exhausted use cases as an industry, yet oh no, I mean once again, it's like what, what a token is just a standard of representation like? Have we exhausted the scope of possibilities, of what we can do with defy or with like? No, definitely not really right. Yeah, I think there's like. So much you, I still think that there's a lot of things left to be solved. Right, decentralized, stable coins I think that's not really been solved properly yet. There's things with like you know, new, like game mechanics. Like you know, a lot of these things are sort of like casino games. Honestly, like a lot of the you know, people like playing those kinds of games. So I think there's a always like new game mechanics that can be created absolutely. So maybe that's another category to say there's like okay, there's ones that are like pure meme coins, whereas like has no game associated with it, but then there's like.

Speaker 2:

I think there's like things that have like game mechanics associated with it, right, this ponzi game but you know, I thought, I think if, if people are aware that it's a ponzi game, then it's like okay, then it's, maybe it's okay, right, it's uh what?

Speaker 1:

it's a game that people are playing correct musical chairs or like whatever you want to call them, but yeah, that's like revenue accruing or fee accruing tokens. I, I understand that oslo is one. What was the rationale? How did you sort of implement it? Because I understand that Osmo is one. What was the rationale? How did you sort of implement it? Because I understand that there might be regulatory concerns and so on, because with all this YoloSwap drama, right, how did you like achieve that?

Speaker 2:

Yeah. So I mean, honestly, we didn't really do much. It's like the revenue, like Osmosis, is controlled by the DAO. We're not the, we're just developers. We don't make decisions on this. And so you know, someone from the dow made a governance proposal to like activate the revenue share and you know it passed, unlike uniswap, which, like it's failed multiple times. This time on osmosis, that actually passed. So, um, yeah, so you know the rev share kind of, and like the dow has pursued like different revenue streams as well, like things like you know, along with the trading fees they have, like top of the block auctions, which is like mev, like revenue, there's different things. And you know we don't as the dev team, we just sort of whatever the dow says to build is what we go build okay and yeah, I mean as far as tokens go.

Speaker 1:

I wanted to ask you about, like now, imagine that there are already several app chains that are attached to the Cosmos ecosystem. Let's assume that this number grows to hundreds or maybe thousands in the future. You know how does Cosmos navigate the gas and like those quirks with sort of impact user experience to a great deal.

Speaker 2:

Yeah, so one of the things Osmosis did was we made it so that on our chain we can support any asset as a fee token to pay your gas fees in. Because what it does is it looks at the DEX itself, gets the TWAP price, uses that to determine the fees and then it swaps all the gas fees to osmo at the and distributes it. So this is like super good for ux, because that means someone can come into osmosis with usdc or eth or bitcoin or atom or tia and they can all they can trade without needing to have our native token. And so then we take this module that we built and now we build a modified version of it and open source it and are like sort of supporting. We're getting integrated into like many more Cosmos chains, so that way throughout Cosmos you'll be able to use any any token for fees.

Speaker 1:

Just shout out to you guys. I mean, a couple of days ago I was trying to mint this slot NFT and it was on Stargaze and for that I needed a stars token to list it and so on, and at that time, uh, I suppose this was like the user experience was really really good, lightning fast transactions and, as you said right, I didn't need to own osmo token, like although I had an address which had some, but like I didn't have to and as as a user, I didn't even realize that oh, that was not the same address I was using and I mean it was just fantastic. So, yeah, like kudos to you guys for that user experience.

Speaker 2:

Yeah, of course, thank you, and it'll keep getting better and better. So, like you know, on Stargaze, you still have to pay your fees in stars. Yeah, they're actually the first chain that we're going to work with to activate the fee abstraction, so on target days you'll be able to pay your fees and everything.

Speaker 1:

And do all of those get routed to Osmo, and because I understand that end of the day you would want to convert those fees to your native token right.

Speaker 2:

Yes, yes, so that way it still provides some value capture to the transaction fee. But honestly, my thing is I like transaction fees is like not a real source of revenue for us, like it's meant there as spam resistance, but our bread and butter, like is the, the trading fees on the decks and stuff. So I like transaction fees are there just to minimize the, the spam, and so whether we swap it to os, we do that right now, but we could also just we just swap it to Osmo. So that way, you know, we collect transaction fees and so many different tokens and I think like sometimes all the like, all the stakers don't want to just get this dust of like 50 different tokens Right, so that's why we swap it all to Osmo. Or, you know, in the future maybe we swap it to.

Speaker 1:

Bitcoin or something like that, and distribute it right makes sense. I mean, I was looking at tbl metrics and users and so on. I think the ecosystem has sort of recovered from the luna crash, but I don't think we are there. Yes, right, I mean pre-luna activity, like for the cosmos ecosystem in general, to match the activity on, let's say, chains like ethereum and solana. What do you think needs to happen?

Speaker 2:

well, I think actually, if you like I don't know if you calculate like the total activity on all the cosmos chains together, so that includes, like osmosis, dydx, you know, neutron injective, yeah, thor chain, akash, like if you actually add kaba, I think, if you add all these together, like I think it is probably like pretty competitive with like something maybe not ethereum, but like I think it's pretty close to something like solana, right, and so I think the problem is mostly, like you know the way the model that people use to compare chains, is this different, right? They're like oh, why does one chain causemos not as big as Solana? But it's like you know, cosmos was not designed that way. It was designed to be like all the activity is split across these hundreds of chains.

Speaker 1:

Right, that's fair. I think you touched upon like meme coins. I don't think that has happened yet. On like in the Cosmos ecosystem yeah, maybe I'm wrong. How do on like in the cost-force ecosystem yeah, maybe I'm wrong. How do you deal with this as a developer or a builder? I mean, does it affect you that you are paying attention to the solid problems that the industry faces and someone else comes along, creates a random token and that goes to I don't know, 500 million market cap or whatever? You have any advice for developers who are probably irritated or frustrated with this?

Speaker 2:

yeah, it definitely is like a little kind of disappointing in the industry in some ways, right, but I think it's just a matter of. I think a lot of these things are flashes in the pants, right. For every like successful meme coin, for every whiff, there's like thousands and thousands of probably more than like of just like random coins. I think a lot of these things just don't have that that much staying power. I think the meme coin craze it like it'll last for like a little bit, but like I don't think it's going to be like you know, the thing that lasts. Like you know, I don't think people are gonna be trading meme coins like this like a year from now, like once again.

Speaker 2:

When I say it by like meme coins, like obviously I know, like I've been a doge person for a long time. I I went to doge con in like in like 20 what is it? 2019 or something. Like I went to like the dogecoin conference, like um, but like it's these like the nth meme coin of, like haha, I'm gonna come up with a funny word launch a ticker, trade it for 24 hours and then dump it. I think that's not going to stick around, right? Okay, that's good.

Speaker 1:

But, yeah, a few of them might stick around. Yeah, yeah, okay, that's okay, because I think, even on Bitcoin right, I mean, that's what is happening currently with these BRC20s I don't think those tokens do anything.

Speaker 2:

Okay, tokens to it? Okay, yeah, exactly. I think right now they don't, but I think what's going to happen is like people are going to launch tokens via brc20 or, probably, more likely, runes, and then I think these things will eventually use things like babylon to be like restaked in order to create like bitcoin. Like you know, your token was issued on bitcoin, but then it actually is a staking token for your own chain. Okay, similar to how polygon? Right, the matic token was issued on Bitcoin, but then it actually is a staking token for your own chain.

Speaker 2:

Okay, Similar to how Polygon right, the Matic token was minted on Ethereum, but then it's used as a staking token for Polygon. I think you might end up having a similar situation with Runes and Babylon.

Speaker 1:

Okay, so in general, like governance aspect, right, I understand that there have been changes on Atom tokens, value accrual and inflation mechanics in general, and largely governance has played a role there. I mean, we often think of governance or governance tokens as worthless tokens that don't do anything, but contrary to that, there's actually some evidence here. Can you walk us through? Like what happened?

Speaker 2:

with Adam. Yeah, you know the real answer. What happened with Adam was the dev team blew up, right, and I was part of that original dev Right, right, and like the YouTube drama with the CEO kind of not doing, you know, not leading properly, the dev team kind of fell apart. We all wanted to keep working on cosmos but no one wanted to work for the company anymore, aib. And then when we all started new things, like no one has any like upside in atom, no one has any attachment to atom or some people do, right, but like it's like well, you know, atom was distributed to this stakeholder set and it's like we could start a new token and like with a fresh distribution and build products around that right, and so that's kind of what, honestly, the biggest problem with atom is it doesn't have a dev team and so without a dev team it doesn't have.

Speaker 2:

It's like it's always just like struggled to find a purpose. There's so many easy products for at, ideas for adam to do, but like just unless there's a dev team to execute on them, they're just never, you know, they're never going to that. That's sort of the problem with adam. It's kind of this like I don't know, a little bit of a ghost token in in way, but like it's the thing is, it's Lindy, right, it's old, it has like staying power and it's like. So it's like now the community is trying to figure out OK, what are things that we can do with it that are just like don't easy products that don't require a large amount of development effort but can capitalize on Atom's market cap or governance capabilities, to what are products that you can do with that?

Speaker 1:

Okay. So it's like everybody's problem is nobody's problem? Yeah, kind of yeah, exactly Okay, I want to just go back to the UX conversation. I understand that there is something related to mobile that's happening at Osmosis. Can you share more about that?

Speaker 2:

Yeah, so you know, for a long time, like Osmosis, we've always focused us on our main website, apposmosiszone, and it works on mobile, but honestly, it's not great. Like the UX, like if you access it through the mobile browser, but, honestly, it's not great Like the UX, like if you access it through the mobile browser. Like, admittedly, like whenever we're developing a new feature, making the components reactive to work in mobile is usually our like afterthought. It's like the last thing we do. It's like, okay, but you know, so much of our usage will come from mobile. If you look at something like Coinbase, over half of their user base comes from Coinbase mobile, and so we're like coinbase, over half of their user base comes from coinbase mobile, and so we're like, okay, now it's time to find it like let's build a proper mobile app, not this like web-based thing, it's like a native app and make use of like a lot of the new stuff that we're building. So, uh, we started building a new mobile app and one of the so a thing that we're working on in parallel that's worth mentioning is like this thing called smart accounts, where it's, it's basically our name for like account abstraction. Yeah, yeah, so we don't use four, three, three, seven or anything like that, because we're not an EVM. And so what we've done is we've built like a native account abstraction stack, native account abstraction stack. We built a native account abstraction stack and this allows it, so, like your account can have like multiple keys with like different cryptography, you can add and remove keys from your account, and so what we're doing is for our mobile.

Speaker 2:

The cool feature that we're like I'm really excited about is when you download the mobile app, it you know normally you have to like take your private key and copy it in If you ever download a new mobile wallet or something, and I think that is just terrible, absurd UX. And so what instead we're doing is you download the mobile app. When you do it, what it does is it generates a new key using the secure enclave of your mobile device, so this requires slightly different cryptography. But then it takes the public key, sends it to your computer. You just like scan a QR code. It sends the public key to your computer and then you sign a transaction on your computer adding that public key to your device. So basically, now your account has like two private keys associated with it One that's on your Kepler wallet on your computer and then one that's on your osmosis app on your phone, and then what's nice is if you lose your phone, someone steals your phone, you can just delete that key from your account.

Speaker 2:

You don't have to like migrate all your stuff to a new account altogether. Migrate all your stuff to a new account altogether. Um, so yeah, building proper multi-device support into into your blockchain account is what I'm like. The mobile app is, like you know, honestly, every all the other features, the mvp, mvp we have we have like a lot of like really cool features. We have planned down the road for the mobile app, like how we imagine the ux. But, honestly, the launch version it's very simple. You know nothing fancy. You'll see your balances, you'll be able to trade everything you'll expect. But I think this is like the feature that I think will be like very novel in the launch version is this new multi-device support.

Speaker 1:

That's really cool. Does it mean that you will get support from Google, pay and Apple?

Speaker 2:

To get into the App Store. Yes, you need to be approved by Apple, which has been challenging. But you know we have. You know we have done some strategies. You know we've been talking, we talk a lot with the DYDX team. They have an app approved in the App Store already and so you know they've given us tips and advice on how to do that. And you know they've given us tips and advice on how to do that.

Speaker 2:

And you know, honestly, people always ask like, oh, are you wasting your time if this doesn't get approved? No, because we look at our stats today, actually majority of our usage is from android, not from apple, not from ios devices, and so, even if the apple app store rejects us, if we get an app onto android, that's like that's still a big win because that's honestly where most of our users are today. And I think you know yeah, I think a lot of people, like in the us, like kind of think that ios is like the mobile platform, but globally, android is actually where most of the usage is yeah, that's definitely not the case in india.

Speaker 1:

I've been living here, I know that, uh, it's not as dominant dominant in this part of the world. I want to talk a little bit more about app chains. And so, okay, here's this right. I mean, I think it was John Shabanu who tweeted or wrote an article on it that typically, when an application starts becoming successful and starts capturing value, they realize that they can capture more value and whatnot by being an app chain. And gradually they realize that they can capture more value by being a general purpose chain. So I think couple of examples dy, dx, avo have gone down that path. So I mean, what end of the day, we'll be left with a number of general purpose change that everyone sort of starts to follow this route, but again, I mean, liquidity gets fragmented, which means that user experience is fragmented, and so on. So what do you think of this?

Speaker 2:

Yeah, I mean, I think so. You know, even, actually, even Osmosis has smart contracts as well, but what we've done is made them permissioned smart contracts where governance has to approve the deployment of smart contracts, and this is you know. Really, what happens is like, even though app chains will add smart contracts and they'll become a little bit more generalized, but at the end of the day, it's mostly for the purpose of driving volume and integrations into the core app. So something like osmosis yes, we needed smart contracts, but that's because we need to integrate bridges and fiat on ramps and all these sorts of things, right, and because the goal here is we just need to make it. So, you know, these things are all things that we need for the core osmosis product. So I think that, like you know, you'll see even let'sdydx at smart contracts, right, it's not going to be like trying to compete with like ethereum and all these generalized like things. It's gonna. All of the applications built on it are going to be like auxiliary applications or functionality that add to the core product.

Speaker 1:

Okay and what about the liquidity that sort of gets fragmented across these things? And I know I mean the bridging experience was great with cosmos when I used it, but it's a lot clunkier with the rest of the like crypto ecosystem in general yeah.

Speaker 2:

So I mean, I think this is uh, I think the answer is just that we have to make it better throughout the rest of the ecosystem and, like the rest of the ecosystem has to like adopt things like ibc and like fast bridging, and like I think there's a lot of room and I think that just means there's an opportunity there, right, like I think right now bridging hasn't been solved correctly or like it's far from complete and I think there's yeah, there's, I do, and I think it will get there.

Speaker 2:

Right, like you know, the whole world it runs off of. Like you know, if you think the example I give is like web development, right, nothing on web development is synchronous, everything is async. Right like, and for a long time it sucked right the ux of like doing it. Like you know, you had like callback after callback and it was hell right, but then, like you know, over time the tooling just gets better and better. You have async await and you have, you know, now, like doing async development is like in javascript, is like much, much simpler and doable, and I think that's just kind of where you know we're just in that pre-async await era right now and like eventually, the ux and speed of these like cross-chain interactions will just get like completely abstracted out, in the same way it did for web development makes sense.

Speaker 1:

When you look at some of these other ecosystems, do you think that there were a few things or one thing that cosmos should have done earlier, or osmo should osmosis should have done earlier? Or you can borrow something from others or anything that you wish. Other systems like borrowed borrowed from here.

Speaker 2:

Yeah, that's a great question. I think Cosmos in general has always been very good at coming up with ideas and then they get like Eigenlayer, like all this restaking stuff was an idea that originated from the Cosmos ecosystem and then I think it was just like you know, like I said, that was the original purpose of the Cosmos Hub was to then do this like security via restaking. I think it just was like slow to execute and then, like you know, eigenlayer and stuff comes along and like executes faster and kind of leaves Atom, like kind of holding its thumbs. So I think just like, yeah, sometimes there's like things where like there's a lot of ideas that originate within the causes ecosystem but then are just like slower to execute.

Speaker 1:

Okay, what do you think of the hierarchy of these tokens Like now, since you brought up Eigen layer, like LRTs and LSTs, and native tokens like, like, does it make sense for Eigen layer using all these stake tokens in general?

Speaker 2:

I think the architecture of Eigenlayer is wrong. I think so. This is why we did it. Mesh security is architecture kind of differently. Yeah, that's right. What I mean by that is like Eigenlayer takes LSDs and you stake them. But how we designed the mesh security is you stake them, but what are? How we design the mesh security is you stake your native token and then you add restaking into it in the native system.

Speaker 2:

Right, like you say, okay, I staked my osmo. I want to say, oh, I also want it to be slashable in case of axelar. I'm going to delegate to axelar. Um, you basically, like we've built this concept of restaking more natively. And I think the Aguilera approaches the problem slightly wrong, where they put LSDs as the base primitive and then they put restaking on top of that. Well, I think the right way to do it is you do restaking as the native primitive and then LSDs get built on top of that. Like I don't think restaking should be taking on. There's no reason that, like you, should be able to restake without taking on lsd risk, or you know, I think like, yeah, that, that I think uh, oh, and it just drives centralization towards like specific lsds and validator sets and stuff. So I I think that that's just like a bad design from eigen layer and it's just going to lead to a lot of like centralization do you think it was to achieve that devial growth and then it can be changed later on.

Speaker 1:

Is it possible? Potentially okay, yeah, I mean these are some of the questions that I had thought of. If you think that we, we can talk about something else, yeah, let me know.

Speaker 2:

Well, I mean, I think I touched on a lot of you know, I think you know, in the three things in like I'm most excited about this year for Cosmos, for Osmosis are probably like Bitcoin, privacy and account abstraction, and so I think we actually touched on all three of them pretty well. So I'm pretty happy.

Speaker 1:

Yeah, I mean, apart from that, apart from those three, is there something else that you're looking forward? One thing related to hospices and one thing as an industry? I mean.

Speaker 2:

Yeah, so related to okay, so not Bitcoin or privacy yeah, yeah, yeah, I think related to the industry. Okay, I'll say, like, the thing that I like I hope will happen is like more I've seen more like web of trust stuff begin to play out. This isn't more for the industry as a whole, and so I think that, like that's, I think part of the end goal of crypto should be to build like web of trust systems. I think that eventually we we might even be able to replace proof of stake one day with a web of trust based consensus protocol, and I think that, like that's what I hope, that we start to see more of. And I've seen like a one or two projects like start to pop up that are working on like web of trust systems and all of these like social layers that are being built, like firecaster and lens, and all this, I think, will just help make that data to build web of trust more available.

Speaker 2:

So that I'm really excited about osmosis specifically, besides the three things I talked about. I mean, I talked about a bit about mesh security as well, like this, uh, so I think very excited to see that roll out and I think maybe like native cross-chain swaps. So today, like the ux of osmosis is very much meant, it's meant to replicate like a centralized exchange and, like you know, you deposit assets, you trade, you withdraw. But we also want to then build like the more of the cross chain swap UX of like hey, I have Bitcoin on the BTC, on the Bitcoin chain, I want Sol on Solana like one click, do the whole trade for me and I don't have to even have an osmosis account or anything. So I think that's the other thing that we're the new UX flow that we're excited for.

Speaker 1:

Yeah, that sounds great. What do you think Solana did right and what do you think they are doing wrong? And do you think that, like having your own mobile and whatnot does that actually help?

Speaker 2:

I think are things they did right. I think they performance engineered very well, like they have great engineers that like squeezed out a lot of performance from a, from the blockchain. Now they made a lot of like centralization, like more centralized, in my opinion, like things that like lead to centralization trade-offs when doing that, like they don't have merkle trees, they don't have light clients, they don't those are trade-offs. I kind of like I I guess we were not willing to make with osmosis, but right, I think they actually did like a lot of like really good performance engineering. Like they have good engineers.

Speaker 2:

I don't think the mobile is like I'm skeptical. I don't think it's a the best usage of time. I understand from the ideological perspective why it's right. Right, like you know, if we want to fix decentralization you know you asked me earlier what if apple and google don't allow our app into the hour. Yeah, that's a problem. We have to figure out how to like solve this as an issue. So I think it's a very ambitious goal. I don't think that at the stage that solana is in, that like that's the best thing to focus on, but I don't know that maybe that's they have some like plan around it. So, okay, there's that. What else did they do? Well, you know, I think they built a good community, right? No, I thought Solana was dead after the FTX crash. I'm sorry, like I thought it dead and I think massive kudos to them, for, like you know, I think that's the most like amazing, like revival story that I've seen in, so that that's pretty cool I hope that we go above and beyond what cosmos was before, before luna and whatnot.

Speaker 1:

Yeah, all right, sir, thanks a lot for spending this time. I mean I, for one, learned a great deal and, as I said earlier, I mean I didn't, I meant it that you know, it's not often that I get to talk to OGs, so yeah, really glad that you spent time with us awesome, thank you. Thanks a lot.

People on this episode