Waiting To Be Signed · interviews on generative art, on-chain
← Index
Interview // MAR 2024

Nicolas Streschinsky

Title: What Is Etherlink?
Role: Generative artist
Platform: Tezos
Duration: 1h 9m
Hosts: Will & Trinity
Listen on Spotify Guest on X Download MP3
#059 · What Is Etherlink?
Self-hosted audio // press play
1h 9m
MP3 ↓

Spot an error? Highlight the words or lines you want to fix — an “edit” button appears, and the panel opens with your selection ready to edit.

Will: All right, hello and welcome everyone to another episode of Waiting to Be Signed, a special interview episode. Today we're joined by Nicolas Streschinsky, the head of DeFi at Trillitech. Trinity is here too. We're really excited to talk to Nicolas about Etherlink, something we've been mentioning on and off on the show that we're trying to learn more about, which is why Nicolas is here. How's it going, everybody? Great to be recording this morning.

Nicolas Streschinsky: Glad to be here. Excited to get into it.

Trinity: Same. Etherlink is something we've heard about, know a little bit about, and read the documentation on — though I'm not quite sure how much of the documentation we've fully absorbed. High level, yes, but getting into the extra details around what Etherlink is, what the value proposition is, and what it means for Tezos and the generative art space on Tezos — really looking forward to digging into that.

Nicolas Streschinsky: It's good that you didn't dig too deep into the documentation. It's still a little technical, and we're working on it — it's an ongoing effort that's picked up in the last few weeks to make it more approachable and to include more details not directly related to the consensus mechanism or how it's linked to the Tezos L1. So if your viewers check back in a couple of weeks, they'll see a version of the documentation that hopefully resonates more and puts them on the right path before diving into videos from the various experts and apps discussing Etherlink integration in more detail.

Will: There's also a great video you did for Next Block Expo that's up on YouTube, which we'll include in the notes for anyone who wants to dig into that — about 17 minutes explaining at a high level what Etherlink is. But we're hoping to get more detailed than that in this episode. Before we jump into everything Etherlink, Nicolas, can you introduce yourself to our audience? Tell us about your background in crypto and how you came to join Trillitech and get involved in Tezos.

Nicolas Streschinsky: My background is in mathematics and finance — that's what I studied. I was planning on a trading career in traditional markets, so I started at European and US investment banks as a trader of various derivatives, mainly equity derivatives. I got interested in crypto when Ethereum came out, in early 2016 or so. I was never a Bitcoiner — I don't code well in C++, and more generally I was never a gold bug, so I never quite got Bitcoin, though I'm coming around to it more as I go along. But Ethereum itself is what got me into the space to begin with, first as a research guy on the side, then a bit in the markets, then a lot more involved in the markets.

I left my role at Merrill Lynch, an investment bank, in 2018, and went to the "dark side" of crypto for the first time as the first trading employee of the crypto effort at Brevan Howard, a London-based global macro hedge fund that's relatively well known in the crypto space. The timing was terrible — 2018, prices going down every single day, volumes going down, people making fun of me for being in the space, honestly. That was the bottom of the bottom in hindsight, but you never really know at the time. I went back to traditional finance for a bit but kept my interest in crypto on the side, and got pulled back in with DeFi summer in 2020, prompted by Compound on Ethereum. That token did really well and drove an enormous summer of on-chain activity — things became doable that I didn't think would be possible so soon after I'd left the space in 2019.

So I went back, got involved in protocols on Avalanche, then joined another firm to do crypto trading. And now I'm on the builder side, the base-layer side, having joined the Tezos ecosystem as part of one of our core entities — the one in London, called Trillitech. My role there is essentially to facilitate the use of Etherlink, Tezos, and any associated rollups we build in the future for financial applications, or applications where transfers of value are crucial to the functioning of the application. That could be SocialFi, could be NFT finance — anything related to transfers of value and finance on-chain.

Will: When you were on Avalanche, did you ever get involved with the rebasing tokens?

Nicolas Streschinsky: Yeah, I was. I wasn't at Avalanche Foundation or Ava Labs, but I was one of the builders and ecosystem participants. That was one of the very big narratives — all those Olympus forks. That part of the cycle was so crazy — there was an entire three-month period where anything people talked about in crypto was related to Olympus DAO forks. On Avalanche there was one; there was one on Fantom, which is still around, kind of; there are a couple on other ecosystems too. Even Olympus itself is still kicking, still innovating. It's come full circle.

Will: Could come back next cycle. We'll see.

Trinity: Still rebasing. Still there.

Will: Like we said in the intro, we're really excited to talk to you about Etherlink and get a better understanding of the project. Let's start with the Trillitech side of it and your role. When and why did Trillitech decide to start the Etherlink project? And in tandem with that, what are the goals of Trillitech as an organization? This isn't something we really discussed with Valérie, since that was a very art-focused episode. Is it generally just to grow the use cases of Tezos? Can you describe a bit more what Trillitech is trying to accomplish overall?

Nicolas Streschinsky: It's actually not so straightforward in the Tezos ecosystem, because it's slightly more decentralized than your typical ecosystem. Typically, how does it work? You have a foundation that holds money raised through a fundraiser at some point, and then you have something like Ava Labs or Aptos Labs, and those guys develop the tech and submit proposals to the community — the bakers, in the case of Tezos — which sometimes get rejected and sometimes accepted, moving the whole thing forward.

In the case of Tezos, there are actually a few of those entities — three, more or less. The largest one currently is Trillitech, mainly London-based, with employees here and there. Then there's a large one in Paris called Nomadic Labs, which is core-engineering focused — a historic entity that knows Tezos very well and also has an adoption arm. And then there's one in Singapore called TZ APAC, focused on adoption. So broadly, you could think of Trillitech as "Tezos Labs" — except there's more than one, and all of them cooperate in various ways.

Combined, they're Tezos Labs, and Trillitech, being the biggest, runs a number of specific verticals. You had Valérie on — she does arts and culture, which is one of three verticals. The second is gaming — you might have Jeremy on the podcast at some point to talk about that; he was at Riot Games, super interesting background. And then you have me, for the DeFi vertical.

Historically — and this leads naturally into Etherlink — the Tezos ecosystem has been less developed in DeFi than in the other two verticals. Arts and culture was a strength from quite early on, and still is. The gaming vertical has also been around and fleshed out for some time now at Trillitech, a bit further along, especially in terms of getting the word out about what gaming looks like within the Tezos ecosystem. DeFi is more recent in terms of the approach we've taken since I joined, and even a bit before.

DeFi specifically is arguably the vertical where most ecosystems have done well, and it comes with real requirements because there's real money involved, and therefore real risk in interacting with the chain. There are exploits all the time, hacks, and things that are undesirable without being outright hacks — like MEV, on most chains at this point. To give you an idea: over the last month and a half, more MEV was extracted on Solana than over the entire past year before that, combined. That's an important point, but it's talked about less.

For the DeFi vertical, one of our big needs was this: so many DeFi use cases and applications have been developed in the EVM/Solidity ecosystem — Ethereum, and in the previous cycle quite a bit on Avalanche and Fantom — and now with layer 2s taking over, it was important to give builders coming to the Tezos ecosystem the ability to draw on off-the-shelf codebases and tools instead of reinventing the wheel from scratch.

Etherlink, which initially was meant to be an example of what Tezos's smart rollup technology can do, evolved quite a bit over time and is now thought of as a community rollup. It's developed internally by those core entities — Trillitech, Nomadic Labs — but will be rolled out as a public tool. Once it's launched, our only ongoing link to it is that we'll continue doing work to improve and update that rollup, that layer 2 — subject to the bakers of the Tezos L1 agreeing that this is good for the ecosystem and the right way forward, exactly like what happens on the L1 right now.

So, to summarize: Etherlink is a layer 2 coming out soon — a blockchain in itself, inheriting its security from the layer 1, which in our case is Tezos, first of its kind — and it will be EVM-compatible in the same way Avalanche C-Chain is EVM-compatible, or Fantom, or Arbitrum One, or HOP Mainnet. It will hopefully have every piece you'd take for granted on EVM ecosystems that are doing well, both in terms of base-layer functionality — what's actually embedded in the kernel that makes Etherlink run — and what's built on top in terms of infrastructure. That includes oracle price feeds à la Chainlink or Pyth Network, subgraphs and off-the-shelf APIs that make frontend development easy, like The Graph, Subsquid, Chainstack, and others, and things like block explorers with the right functionality, exposing everything you'd expect on other EVM networks.

Will: A lot there — I think I understood most of it. To continue on Etherlink, a natural question would be how it works and what makes it different from other L2s. You rattled off a few L2s there, and as far as I understand, all of those inherit the security of Ethereum ultimately — if for whatever reason they fail, at least you collapse into the Ethereum L1 and retain the security of that decentralized network. Obviously one difference here with Etherlink is that it'll enable Ethereum applications to use it, but it'll be collapsing down to Tezos — everything transacting through Tez, which will be like the native gas token for this. But aside from Tezos being the underlying L1 that secures it, what else is different about Etherlink? We hear a lot about centralization, and you already brought up MEV — maybe you can give a basic explanation of that for people who aren't familiar. I can try to give my own understanding, but it's probably not going to be exactly right. So maybe you could talk a bit about what makes Etherlink different from other L2s.

Nicolas Streschinsky: You've basically answered your own question already. At a high level, L2s and the concept of rollups were invented on Ethereum, so when people think "L2" or "rollup," or even newer blockchains in general, they think Arbitrum One, or Optimism (now renamed OP Mainnet), or more recently Base, which uses OP Mainnet's technology.

What we're trying to do by going to L2s, and this is true for Etherlink as much as for Arbitrum One, is scale one particular part of blockchain: execution. Before rollups, the approach was called sharding; now it's rollups. Tezos has gone the rollup route too, but with a particular twist. Etherlink is indeed a rollup, similar to what Arbitrum One is to Ethereum, but the difference is that Etherlink is developed by the same core entities building the Tezos Layer 1 -- it's an "enshrined" rollup. That sounds like a fancy term, but it just means the code that makes the L2 run sits in the same codebase as the L1. To update Etherlink, you actually need to update Tezos. That has pros and cons. Interestingly, Ethereum itself, if you look at some of Vitalik's latest blog posts, seems to want to move in that direction too, using a different technology called ZK rollups.

Keeping it high level, there are essentially two ways to build rollups and provide the proofs needed to properly inherit security from the layer 1 you're rolling up to. One is ZK rollups: very complex, an outstanding piece of technology, but not ready for primetime in our opinion. Transactions are relatively slow and encumbered, the technology is complex, and as a result the cost of transactions tends to be quite high. Since our whole goal is to scale execution, we want the cheapest possible transactions while remaining secure -- so we went with the other option: optimistic rollups. Every time you see the Etherlink website or announcements from the Etherlink account, which I manage, you'll see references to optimistic rollups -- that's the technology we chose to make sure that whatever you do on the Etherlink blockchain is exactly as secure as if you'd done it on the Tezos layer one.

That's the general explanation. Now, quickly, on what separates Etherlink from its competitors: partly it comes from arriving to the game relatively late -- Arbitrum One has been around for twenty months. We looked at how the others were doing things and what the pain points were. One of the biggest is that a number of actors sit on the chain when blocks are being crafted and can choose to reorder transactions in a way that benefits them. Since this is a zero-sum game, if it benefits them mechanically, it doesn't benefit you. That's what MEV really is: an unfair advantage that comes from validators and the whole complex ecosystem around reordering transactions after they've been submitted, in a way that hurts you and benefits them.

That matters most for DeFi-like use cases where a transfer of value is involved. But in reality, most MEV back when it was at its peak on Ethereum -- less than six months ago -- was linked to NFT mints, creating what people called MEV spikes. A lot of the MEV extracted on Ethereum through transaction reordering came from NFT mints, which I think your audience is quite familiar with.

We looked at that and thought it'd be great to address, even down the line once Etherlink's full roadmap is implemented and Etherlink has a healthy dose of decentralization. You don't need a layer 2 to be as decentralized as a generic layer 1 -- Tezos, Ethereum, Cardano, and others should be as decentralized and credibly neutral as possible, but for layer 2 it's less clear-cut, because the worst that can happen at the layer 2 level is that transactions get reordered to benefit a certain party. That's already happening on a lot of layer 1s as we speak. Since there's no real loss of money in most cases where decentralization matters at the layer 1 level, it's less critical for layer 2 -- though it still matters.

To make a quick comparison: if some Ethereum layer 2s decide to have a decentralized sequencer -- a set of computers ordering transactions rather than one centralized computer -- and that setup isn't done properly, accounting for the fact that some of those computers can misbehave, you can end up in a situation much like the one we had on Ethereum before Flashbots and the MEV-Boost process were properly implemented. Even now, a lot of people don't use that protection when interacting with the chain, and they're subjected to potential MEV loss as a result.

That's a very complex answer, but hopefully, broadly speaking, it made sense.

Trinity: Mostly.

Will: There's also, well, it's not even a Layer 2 yet, but some L2s like Blast -- as I understand it, right now it's really just a multisig wallet that they're asking you to send ETH to. That presents its own security issues: all it takes is the four signatories to that wallet deciding to take the ETH, and you're done. Their roadmap is obviously to launch their L2 and decentralize it somewhat, but starting from a decentralized place avoids that whole conversation about risk.

So how is this all going to work with wallets? Is it only going to be Tezos wallets? I'm thinking about this from an adoption standpoint -- for a lot of L2s, we've both been dabbling with Base and others to collect NFTs, and even though it's a bit of a pain, I can manage it all from my MetaMask wallet by just adding those networks. So is the eventual goal to have Etherlink in MetaMask?

Nicolas Streschinsky: That's a good question, and it's quite relevant to how users actually benefit from this in one way or another. One of the benefits of being EVM-compatible is that the chain essentially lets you forget about Tezos. Etherlink in itself, if you don't want to look under the hood and understand where the security comes from, is a blockchain in and of itself — just like Arbitrum One is a blockchain in and of itself. It does get its security from Ethereum, but if you're just a day-to-day user, you go on your Binance account, your Bybit account, your Coinbase account, and you withdraw to or deposit from Arbitrum One. You don't care about Ethereum at any point during that process. Similarly, once you're using Arbitrum One, you go on MetaMask, Rabby Wallet, Rainbow Wallet, Uniswap Wallet, and you interact with whatever application by adding the Arbitrum One blockchain to your MetaMask.

Everything I just described for Arbitrum One will be possible for Etherlink, and that will be the main way to interact with it. You're in the EVM ecosystem, where every tool designed for Ethereum, with Ethereum's execution layer in mind, is usable for Etherlink. That's massive, because a lot of those tools are extremely good quality — all the middleware providers for builders, like Alchemy, Third Web, Ethers.js, Web3.py. All of those are usable day one for Etherlink. You change the chain ID and you're up and running. The chain is compatible enough, with all the right bells and whistles, that you don't even have to think about whether you're using Etherlink, Arbitrum One, or Ethereum mainnet.

Now, from a user standpoint, the current meta isn't too different from what people are doing on Tezos currently. You spin up your Temple wallet, a browser extension, connect your Ledger physical wallet to it, or use a hot wallet directly. You interact with apps, and every time one prompts you to approve a transaction, you get a pop-up and click approve — hopefully having checked what's happening there. MetaMask pioneered that process and still works this way. But that only reaches a relatively small number of users: people familiar enough with the space, confident enough to create an account, take notes of a seed phrase on paper, and hope the cleaning lady doesn't find it and you don't lose your money.

If you think about how the space will look six months from now, it's a very different picture — and I think your audience will be keen on this, because it can attract a lot more people who have no interest in blockchain itself, just in what it can enable for them. That's something we're super focused on for Etherlink, and it's enabled by a standard created by Ethereum, an "Ethereum Request for Comments" called ERC-4337. Barbarian name, like many of those. But essentially, that standard establishes how smart accounts will work going forward — how we collectively agree to design them to make it less painful for a new user to start using an application.

Let me describe the workflow, since that'll sound more concrete. You've heard of an app, you Google it, you click on the website. Now you're there, and there's no "connect wallet" button assuming you've already installed Temple or MetaMask or Rabby. Instead, you have "log in with Google," "log in with Apple," and so on. You log in with one of those services, and a wallet gets created for you in the background — but it's self-custodied. No one can steal the funds you deposit there; you're the only one who can unlock that wallet. You even have a social recovery feature to recover it if you lose access to your Google account. That's still an ongoing process, but it's maturing fast, and applications like friend.tech, blackbird.xyz in New York, and builder.tech are already leveraging a first version of this, which will keep improving.

In practice, you'll think of that embedded wallet as something associated with that specific app. Say you wanted to use an Etherlink NFT marketplace — you didn't have a wallet, you went there, logged in with your Google account. Whenever you come back and log in, you have access to those funds and can spend them. But reusing that wallet for another application requires more skill and brings you back to the MetaMask paradigm, where you need to understand wallets and seed phrases. If you just want to use that wallet with that one application, though, you're happy to do it with your Google account alone.

What that might mean in practice — and I'm speculating here — is that wallets are going to progressively lose ownership of the relationship with the customer, and apps will go back to that Web2 paradigm where they control that relationship. The customer thinks: "I have my account with that NFT marketplace on Etherlink," or "I have my account with Uniswap." I think that's a game changer, because once that's possible, a lot of users who are crypto-aware but not crypto-native might think, "You know, I really don't care about blockchain. I don't even understand what Tez or BTC or ETH actually are, and that's fine — I just want to mint some NFTs and sell them to my friends, or collect them, because there's some utility attached to them. I want to prove to someone easily that I own that portfolio of NFTs without taking notes of a seed phrase or understanding how private keys and public keys work."

Then all those marginal use cases, which have been marginal until now, could become mainstream — ticketing, loyalty points for various brands represented as ERC-20s or NFTs. It becomes no more cumbersome for the brand or their end users to do it that way than to do it the Web2 way with a centralized database. We've spent a lot of time on Etherlink making sure these use cases — and the middleware developers need to build them — are there from day one and well documented. We're going to prepare videos to walk people through it. Third Web has done great work on this, and Alchemy as well.

Will: Trinity, I think you're the candidate for a follow-up here, since this is everything you think about with crypto all the time.

Trinity: It sounds like the end goal is making something scalable and foolproof from an enterprise perspective — whether it's the NBA, the NFL, Ticketmaster, that sort of thing. Where else do you see people in the Web3 ecosystem tackling this problem? Do you feel it's a differentiator, or are there competitors you'll have to compete with on this specific front — especially the Google-account, self-custodied wallet that doesn't involve MetaMask? It's super fascinating.

Nicolas Streschinsky: We're clearly not the only ones thinking about this — we didn't write that standard ourselves. But I think there's room for a lot of people here, because when it's a new thing and there's growth, you're not trying to take market share away from incumbents, you're capturing new business, which is typically much easier. So of course other people are thinking about what needs to be offered to the ecosystem, or to laymen, for this to be useful.

A lot of ecosystems are looking at this. If you look at what Ethereum itself has done, I think StarkNet was quite a pioneer — they have a wallet developing not just on StarkNet but elsewhere, called ArgentX. I'm not saying that because they're French, but I think they did good work there. Still, this has remained too complex so far — they were early, and the technology keeps improving. For someone like my wife, or my father, or anyone who's heard of crypto but couldn't care less, to actually use something useful to them with zero pain points — I think we're getting there.

Friend.tech is a great example. If the marginal effort of using a solution is tiny — for example, ERC-4337's account abstraction allows for something called a paymaster. One of the pain points of interacting with blockchains right now: you spin up a Temple wallet to interact with Tezos, and even if you're happy to take notes of the seed phrase, you get started, you've created the wallet, and guess what — you have zero Tez in it. You need a friend generous enough to send you 1 Tez for transaction fees, or you go to centralized exchanges, or you leverage one of a million on-ramps, some of which charge absurd fees.

What if, in order to claim your first few NFTs or do some low-value on-chain things, someone just pays for your first few transactions? That happens in Web2 every single day. If you order shoes on Amazon tonight, do you care whether Amazon is using AWS or Azure in the background? No one cares what's happening back there. You care that you click, the shoes get sent, a payment provider enables the payment, and they're delivered to your home or office.

That's one of the steps that need to be ironed out. We've spent a lot of work selecting the right on-ramp providers, making sure there are API calls that can happen, potentially involving a KYC provider if the amount is large enough in certain jurisdictions, so that after a few transactions covered by the application, you can easily buy NFTs on a marketplace, or sell whatever your local fiat currency is — say euros — and buy some Tez, or a piece of fractionalized real estate. But for that, you need your euros to be stablecoins denominated in euros, tracking 1:1, not fiat euros sitting in your bank account. Those have been pain points in the past, and there have been massive improvements made by various counterparties. Our added value is mainly about selecting the right partners, having them on board from the get-go on Etherlink and on Tezos, to facilitate those new forms of blockchain use cases.

Trinity: So it sounds like, on the business side, it's also about finding partners willing to eat particular fees because they can extract value from these end users elsewhere in the chain or cycle — like your Amazon example.

Nicolas Streschinsky: Yep.

Trinity: We often cite free shipping at Amazon as the classic way of getting people in — they don't pay for shipping, expectations shift, the game changes. But Amazon recoups those costs other ways: pushing competitors out, membership fees, sometimes higher prices. It depends. Is that front of mind for you too — thinking about how to make Etherlink a profitable entity in the long run and the short term?

Nicolas Streschinsky: I'll just clarify that Etherlink is not an entity, but I see what you mean with the question. Etherlink, being this community rollup, can be thought of as an extension of Tezos — and if you look at the roadmap, Tezos and Etherlink will actually be one at some point. But "profitable" — what does that mean in this context? Essentially, it's about Tez holders, the holders of our token, being able to get some benefit in the long run from network activity. That materializes on Etherlink because whenever you do a transaction, there's a base fee, paid by the user, and it's paid in XTZ. Everything is paid in Tez, the native token for Etherlink, as Will mentioned. That base fee is burnt, which ultimately benefits all Tez holders collectively from more activity on Etherlink — and on Tezos to begin with, since more activity there means more need for Tez to pay for gas.

To get back to your actual question, which a lot of people are discussing on podcasts right now: customer acquisition in blockchain and crypto. It's been extremely costly since 2020 and DeFi summer. It's taken so many forms, and it's probably the most complex form of customer acquisition out there — though I might be biased, since I haven't worked in Web2. In the case of Amazon, there's free delivery, there's a Prime subscription that's a loss leader because so much comes out of it. It makes the customer more loyal — you're already paying $9 a month, so you might as well use the free delivery every time.

In crypto, customer acquisition costs have been through the roof, more expensive than anything else out there by an order of magnitude, at least. If you tell an app, "you might have to pay for the first 20 transactions of the user" — transactions on Etherlink are likely to average around a cent each, so 20 transactions is about 20 cents. For comparison, 20 cents is roughly the customer acquisition cost for games. A typical game launched on the App Store or Google Play would happily pay $1 to $5 per install. So 20 cents seems reasonable by that standard.

But in crypto, airdrops are considered by some — not all VCs — as part of customer acquisition, and that gets you to costs of hundreds of dollars per user, often for a user who doesn't even end up doing a single transaction in your ecosystem. There's a lot of room for improvement there. I'll get criticized for this, because a lot of people love airdrops — myself included, as a user — but going from hundreds of dollars per user in the current paradigm to a few dollars here and there to actually facilitate someone using your app is probably a no-brainer for most apps out there. We're definitely willing to help with that through various schemes and incentives we've refined over time. The Tezos Foundation has really stepped up under new leadership, and there's support available for the right kind of apps and builders.

Will: It's going to work the same way as any blockchain — fees accrue to the validators on Etherlink, and that's how you incentivize people to secure the network. But the fees in this case are incredibly low. They're already low on Tezos — you can get between 4% and 5% through liquid staking there. To provide any kind of reasonable yield, we're projecting forward to an extremely high transaction volume, a lot of people using this all the time, for the fees to accumulate — especially if they hit that 1 cent or lower.

How do we onboard people to the idea of "no one likes Tezos, but use this Tezos L2"? That seems like a big barrier. There's a lot of skepticism around Tezos in the wider crypto community, so securing Etherlink with Tezos might not be attractive to a lot of users out there. But Nicolas, a lot of what you're describing is about not necessarily going after those current users, but expanding the ecosystem and bringing new people in. What does that look like? Do you have any partners lined up, or anyone interesting you can hint at?

Nicolas Streschinsky: Let me take the first part. Let's address the elephant in the room — the latter part of your question — which is: for people already in crypto, specifically those already familiar with Tezos in one way or another, what's in it for them? What could bring some of them back, or make them more active again on Tezos, Etherlink, or other associated rollups in the future?

The answer is good apps. Apps that let you do something you actually want to do — not because you're paid or incentivized to do it, but because it's genuinely entertaining or useful for your personal finances, or because there's art worth collecting in a digital way that's traceable and can't be faked, whatever it is. Something good enough as an app or use case to drive people to it regardless of whether it's on Tezos, Etherlink, Ethereum, Arbitrum One, Solana, or anything else.

What I described before is about lowering the barriers to entry, in every way possible, so those use cases are incentivized to come to Tezos, or to build there first. One thing we're doing on the DeFi side specifically: we're running an incubator starting April 15th in Singapore. The first two weeks are in person, four weeks after that are remote. Then it goes to an investment committee, and some projects will be funded jointly by us and a partner of ours. After that, we help them go to VCs and raise substantial pre-seed money — that's the goal, at least. That should provide a solid pipeline of apps available on Tezos and/or Etherlink relatively quickly.

For those apps, we're thinking along the lines of what I described earlier — reaching not just people currently in crypto, but the wider audience who might have had some crypto-linked activity in the past. I'm thinking of users of Voyager, BlockFi, Celsius — a very low-touch approach to getting yield on crypto. Didn't pan out well for a lot of those users, I have to admit, but the idea itself was remarkably good — the execution was poor, but the idea had merit.

Similarly, if you look at centralized exchanges right now — Binance, Coinbase, Bybit — they're all doing one consistent thing: moving much closer to the on-chain side of things. Binance has seen themselves, from the very beginning, as ultimately a gateway to DeFi — that's what they're trying to enable more and more with their new Binance Web3 wallet. Same with Bybit — they've recently launched MANTLE, a layer 2 that's actually competing with us. And Coinbase has launched Base.

To touch on Base, for a related part of your question — you were asking, okay, fees are going to be low on Etherlink, and they're already low on Tezos, so how do you get to a point where those fees are substantial for billions of dollars in market cap, and hopefully tens of billions in the future? If you look at what's happening with others—take Base, for example, when they launched and friend.tech was hot—they had profits of something like $10 million. We're not talking billions; we're talking amounts that were already substantial a month or two after launch, with fees that were very small, at a time when the price for publishing data on Ethereum was low because there was little activity and base fees themselves were low. They still managed to make some money.

Over time, as more users come on-chain and the activity becomes more repeated, things will shift. Right now it's very specific: on one hand people trading NFTs, on the other people trading meme coins. That's about 90% of the activity at the moment. That's not the endgame—that's just where we are right now, because there's nothing else easy or sexy enough to do on-chain yet. You have Rollbit on Solana and things like that, but those are piecemeal, the very beginning of what we could get to.

I think what's really going to drive this cycle is finding the next use case or set of use cases. My hunch is that could be unlocked by lowering the barrier to entry enough to get tens of millions of users, as opposed to the roughly 1 million active users currently doing things on-chain—whether they realize it or not, whether they're familiar with what "on-chain" even means or not.

This connects to my answer to the previous question: no blockchain has product-market fit at the moment, in my opinion, in the grand scheme of things. If you're talking about big-tech or big-finance or big-oil type profits, no one is making that kind of money in crypto right now apart from Binance—and Binance is off-chain. So how do we get some of their users on-chain, which I think they're keen to do themselves? And how do we scale some of these use cases to be widely used enough to generate billions in profits, as opposed to tens of millions or even single-digit millions?

At the end of the day, we're an infrastructure piece—software that lets others build cool B2C apps that actually work and do the right things. To attract those builders, we need well-structured grant programs, investment programs, incubators, you name it, that create a genuine sense of community. The CEO of the Tezos Foundation calls it "closeness." I wasn't super familiar with Tezos until I joined recently—I'd always been aware of it, but in the background—and one of the things that surprised me most is how tight-knit that community is. I'd like to scale that while keeping the warm, in-person feel. The Dev Success Team does that really well, and I'd like to see it at a much bigger scale. That's what I'm trying to do with the incubator we've called the DeFi Catalyst Accelerator, or DCA. Right now it's a one-off, but I'd like to run it once a quarter in different cities, and have builders coalesce, build use cases that increasingly compose with and rely on one another, and trade tips on how they've overcome struggles—whether linked to the Tezos tech stack, more general Ethereum execution-layer hurdles, or regulatory hurdles in a given jurisdiction, since DeFi is getting increasingly regulated. That's partly what we're marketing towards.

The Etherlink Discord has a builder section that's quite active—Anthony has been working with two other guys, David Ryan and Sasha Aldrich, to keep it super alive and vibrant. If you go there after the show, you'll be surprised how many people are suggesting things. They can be angry, they can be super enthusiastic—anyone is welcome. We want civilized conversation, but at the end of the day, every idea can have merit if the arguments are put right.

Builders are the first goal. For the wider community, it's about making them understand that Etherlink is worth a try and very easy to get started with—especially once we have some of the first apps using that embedded-wallet setup I mentioned earlier. If you look at friend.tech users, when they started, they were like, "What actually is Base? I transferred some coins from my Coinbase account to Base, that's it." They use friend.tech—they're not into Base as a blockchain itself, they're into friend.tech. That, I think, could be a huge source of users and awareness in the wider crypto ecosystem: having the right apps built either ready from day one or close to launch, and co-marketing and running events with them. The secret sauce is engaging with the right people in the right way. It's about the builders in the end.

Will: I'm in the Discord, by the way—I got in before the gates opened two weeks ago. I'd color my impression a little differently. I think it's very chaotic in there, with a lot of people just begging for an OG role. I'm genuinely confounded by what's going on, because it feels like people expect an airdrop, given how they're behaving with all the different levels and cycling—but there isn't going to be an airdrop. So I don't really understand what the goals of the Discord are as it currently exists.

Nicolas Streschinsky: Criticism and constructive engagement are super important, and what you said is 100% true—it's shared by a lot of us. There are a few aspects to this, and it relates to my earlier answer. The current community still in crypto—remember 2022, Terra, FTX, Celsius, we've had a bunch of that this cycle—the people who are still around are there for a specific angle. Not all of them, but a lot of them. I'll be honest, I do it too: I go into various ecosystems based on the incentives on offer. I think a lot of us are like that, and it's helpful in the sense that it means people are still engaging with the tech being built. But it's obviously not sustainable long-term.

Jupiter released a $500 million airdrop a couple of days ago. In what other industry do you spend $500 million on customer acquisition? That's because we're early, because American VCs are extremely risk-taking, and because we're hoping there's a next step where people are actually using something that makes sense for them beyond the pure incentive. You watch videos these days of guys using Vertex Protocol—a great protocol on Arbitrum One—or Hyperliquid or some other trading platform, saying, "I'm going to trade once in a while because I want my point allocation to go up, and I think the airdrop is going to more than make up for the fees I'm paying." But by definition, an app like that doesn't yet have product-market fit, because the only users are people who think that by using it, they're going to be paid rather than pay. The VCs financing this—many of whom are really on top of their game—are betting that eventually the "plus-one" user, the one from a more mainstream audience, will find something valuable there aside from the potential airdrops, the point systems, the incentives, the liquidity mining. That's where we are at the moment.

And that gets reflected in the Etherlink Discord too, because we're part of the wider ecosystem—not just us. Blast has some of that, Manta Pacific has some of that, and upcoming ones like Monad and Shardeum will have to deal with it too. I take your feedback on board, and I think there are ways to refocus the conversation in a more productive and engaging manner than purely asking "who is the OG."

Trinity: In summary, despite the name, Etherlink operates much more like an Avalanche or other EVM-compatible chain—just on a Tezos L2. It's not directly connected to Ethereum beyond being EVM-compatible. We've already talked about the relationship between Tezos L1 security and Etherlink. Is there anything else people should expect from Etherlink in terms of—not airdropped tokens, but its own level of security? Will it have its own native token separate from Tezos? Anything else people should be thinking about?

Will: Are we going to have to wrap our Tezos for it to work on Etherlink? Is it going to have to be wrapped?

Nicolas Streschinsky: There are a few ways. Let's say Etherlink launches tomorrow. How do you actually use the chain? How do you bridge to it, deposit to it, whatever you want to call it? I'll start with what I think will be the most-used method day one and work down from there in terms of expected user numbers — though this is just my personal take, not something we've discussed as a team.

The biggest one, I think, is our partner LayerZero, a cross-chain messaging technology that lets you bridge value, particularly tokens. They have a standard called OFT and a wrapped-asset bridge. Without getting too technical, you'd use a UI — linked directly on the Etherlink portal, or findable on DeFi Llama — where you can bridge from any LayerZero-supported chain (Ethereum mainnet, Arbitrum One, OP mainnet, whatever) to Etherlink at low or no cost. That could be ETH, USDC, USDT, and so on.

There's also a handy feature: say you're bridging 100 USDC, you can tick a box saying "I also want some gas tokens," and LayerZero will convert a bit of that in the background and give you some Tez to get started on Etherlink. From there you can use a decentralized exchange to convert more USDC into Tez if you like. That's the well-known path — familiar to anyone already in the space — for moving money from other ecosystems into Etherlink.

The second way is for users already on Tezos. Say you hold some Tez on Tezos, the L1 in this context. You can use the L1-L2 bridge, which we designed ourselves and get essentially for free by virtue of Etherlink being a rollup. That bridge lets you move native Tez between Tezos and Etherlink, and it also lets you bridge any FA1.2, FA2, or FA2.1 token from Tezos onto Etherlink in ERC-20 format. So a token like USDT on Tezos goes through the bridge and comes out almost instantly on Etherlink as an Ethereum-type contract — not a Tezos contract — given to you one-to-one.

The final option, which I very much hope will be live by March 13th, when Etherlink mainnet launches, is direct deposit or withdrawal from a centralized exchange. Right now you can withdraw Tez from Binance directly onto Tezos L1 for about 0.1 Tez — very cheap and fast. I'm hoping you'll be able to do the same directly to Etherlink from day one, skipping the intermediary step of going through Tezos first, which admittedly isn't very user-friendly as an extra hoop.

So those are the three main ways to get value and native gas tokens onto Etherlink from the start.

Will: If I have a stablecoin on Tezos L1, I'll be able to bridge it up to Etherlink fairly cheaply because it's Tezos.

Nicolas Streschinsky: Very cheap.

Will: But once it's on Etherlink through LayerZero, that stablecoin could travel on to Arbitrum or elsewhere — so you could bridge something entirely off Tezos into Ethereum if you wanted. Does that open the door for NFTs to move the same way? Could you take an Ethereum NFT all the way over to Tezos L1, or vice versa?

Nicolas Streschinsky: Not necessarily. It depends on whether the token in question — fungible like USDT, or an NFT like a Bored Ape — has the necessary function built into its contract. Those are LayerZero-specific functions that create a standard called ONFT, for Omni-Chain NFTs. If a token has that function, the LayerZero infrastructure understands what it is, and the bridge can move it from Etherlink to Ethereum mainnet, to Arbitrum One, and so on.

It's genuinely cool technology — we work closely with the LayerZero team — because effectively it creates a single pool of tokens rather than siloed versions on each chain. Say you and I create a collection called the Nico Will Penguin, 10,000 of them, all built on a codebase with that ONFT function. We could burn them on Etherlink and mint them on BSC or Arbitrum One, and vice versa. That's the omnichain NFT standard — less widely used so far than OFT, the fungible-token equivalent behind USDT, USDC, and so on. But I think it's interesting going forward, since it means NFTs aren't siloed to one chain — useful for NFTFi applications or multi-chain games. Right now it's still behind the pure OFT side of things LayerZero is known for, but who knows what happens next.

Will: I had to ask, because it comes up constantly with collectors — especially given the FUD around Tezos — what happens if, for whatever reason, Tezos goes down? Is that the last failure point? If the chain fails, does $50,000 of NFTs just go to zero? Is there a way to rescue them? I know it's not as simple as just moving them — there's provenance and everything else you'd want to preserve. I don't think Trinity or I believe this is going to happen, but we've heard other collectors who straddle both chains raise this fear. It sounds like it'd be more applicable to new NFTs created in the future than to existing ones.

Nicolas Streschinsky: This was a live topic when, for example — I can't remember exactly who it was, I think it was DeGods — moved from Solana to Ethereum via Polygon and ended up giving back a grant because they didn't think Polygon was a safe enough chain, given it was having block reorgs of hundreds of blocks at a time. They saw it as great for other use cases, but not necessarily for NFTs.

That's a good final point to end on, because it's a generic concern. A few things: I might be biased, since I work for one of the core entities in the Tezos ecosystem. But having come largely from the Bitcoin, Ethereum, and Solana ecosystems before this, I'd say Tezos, despite its ups and downs, has a setup that's genuinely very decentralized and safe from an incentive standpoint for the bakers. You're talking about the roughly 5% issuance that goes to bakers right now, which is being reworked in various ways — there's a lot of community discussion around something called adaptive issuance. But the core focus of Tezos's founders and core entities, Trillitech included, has always been keeping Tezos secure, decentralized, and reliably validated. To my knowledge, that chain has never really gone down. That's not true of Solana. It's not even true of Arbitrum One, which had issues recently — and I really like Arbitrum One, so that's not a biased jab. Tezos remains robust from a first-principles standpoint: how simple and sound its consensus and validation are.

It's also proof-of-stake, so we want overall staked market cap to remain high enough that the stake is meaningful enough to make attacks difficult. I believe that's still the case, and it's something hard for new entrants to bootstrap. If you're worried about that kind of thing, it means you understand what you're doing, which is good news.

Net-net, the chain works well as is. We're doing everything we can with Etherlink specifically, and there are other initiatives underway to make validation as safe as possible while keeping it profitable for bakers in a way that doesn't disincentivize holding Tez or using the ecosystem's platforms, Etherlink included. Striking that balance among stakeholders is something the Tezos ecosystem has done relatively well. So my honest take is that this isn't a real concern — though take that with a grain of salt, since I do work for an ecosystem entity. If anything, compared to other chains that have been used for NFTs, without naming names, there's arguably more risk outside the top two or three chains than there is with Tezos, or with Etherlink, which offers the same security properties as Tezos from day one.

Will: From our standpoint, as people invested in the art NFT side of Tezos — we collect cross-chain now, but we started as a Tezos-only podcast — the idea of Etherlink driving more activity and TVL to Tezos, and just pushing up the price of the coin so people stop posting charts on Twitter making fun of the chain, would go a long way toward curing a lot of that FUD. Even people like us, who don't fully understand the technical side, get that it only takes twenty people in Europe willing to run nodes for Tezos to stay up. Every generative artist in Amsterdam could keep this thing afloat off some Raspberry Pis if they wanted to. It's not going anywhere. Hearing you talk through it with your experience is genuinely reassuring — thank you for that.

Nicolas Streschinsky: I think ultimately those things are part of a whole. A lot of the pieces in crypto might not seem connected from a 10,000-foot view, but they actually feed into one another. In the last bull run, in 2021, the buyers of the Bored Apes, the Punks, and all the expensive NFTs were largely guys who'd just made money from DeFi-related tokens, or from AVAX and Fantom going up. There was a very clear correlation between the prices of NFTs and the prices of those tokens, which shows you that the activity underpinning all of this is actually quite linked.

Things are siloed to some extent — you have people who prefer profile picture NFTs, and others focused on utility NFTs for trading on decentralized exchanges, perps, and so on. But it's all intertwined, at least to some degree. So if something attracts activity to Etherlink, a rising tide could lift all boats, and everything else gets traction from it. You saw that recently with Avalanche, where inscriptions became a big thing — I still don't fully understand why — and as a result the entire Avalanche ecosystem got lifted for a number of months just from that.

Will: I'm still trying to wrap my head around Bitcoin inscriptions myself. I keep getting into heated debates with artists who are considering going over there, and I still don't fully understand it, but I'm trying to stay open to it. We've been going for a while here, so let's wrap up with a rapid fire. Trinity had to drop for a work call, so Nicolas, two quick questions: first, the simple one — when's the next bull run? And second, the real one — who would you like to hear us interview on the show? You've already mentioned the gaming person at Trillitech, and we've had Valerie on. Is there anyone else in the Tezos ecosystem you think would be good for us to talk to?

Nicolas Streschinsky: I'll answer the first one quickly. We're already in the next bull run. If you go by the definition a technical analyst would use, you're already seeing the structures and volume patterns changing in a way that's consistent with what we saw at the end of 2020, at the end of 2016, and so on. I have no crystal ball, but for some reason it does seem to coincide with the wider economic cycle and with the Bitcoin halving. So arguably we're in the midst of the bull cycle — though that doesn't mean it always goes up even during a bull run, and it doesn't mean we're going to go up and up for the next few months. But I think we're already in there. What's encouraging is the private market trend: what VCs are funding, and the covenants and criteria they're applying, are starting to relax a bit.

For the second question — these days I spend a lot of time with builders who were in Tezos long before I arrived. One name is Dmitry Kovalevsky, founder of CordFi. He's been in Tezos for a while, has looked at a bunch of other ecosystems, and does great comparisons. He's working on a very cool project we're trying to support in any way we can — an order book decentralized exchange that works much like the NYSE or the Amsterdam Stock Exchange. That's something fairly new to blockchain, and he's a very technical guy with a great understanding of it. I think he'd be one of the good ones to talk to.

Will: I'll have to get his contact from you later — definitely worth thinking about. I think that's a good place to wrap it. Thank you so much, Nicolas, for your time, that was great. Hopefully our audience understands a lot more about what Etherlink is now. I'd hope this episode gets circulated — it's pretty bullish for Tezos and what's going on in the ecosystem here, and it shows people are actually building. There's a plan. Even if we're not necessarily coming for all of Ethereum, we're coming for the broader audience that hasn't adopted crypto yet, and we're going to try to bring them in in a way that's cheap, fun, and easy. Seems pretty simple from there.

Nicolas Streschinsky: Let's hope so — fingers crossed. Thanks for having us, it was a pleasure to speak with you.

Will: That's it for this episode. Hope you all enjoyed it. Check out the links below, and consider joining the Discord — you're not going to get an airdrop, but you might learn something. Definitely check out the YouTube video too for more on Etherlink. All right, that's it. Bye everyone. Cheers.

Change log

  • Initial transcript — auto-transcribed (AssemblyAI) and readability-edited.