Hey, hello. Hey, my name is Makoto, and I work as, well, it says here, a protocol developer , but today my title is a little different because my ChatGPT said it was better, so today's role is the evolution of naming in Ethereum. And so, uh, I'm, um, "OG" as everyone calls it, but I just realized that Lisbon is full of "OGs." Uh, so, oh, and usually, if you've seen me before, I had long hair, and I had this, uh, little avatar that I drew by hand, back before , you know, the era of generative AI. And so , today's topic, since I'm kind of a self-proclaimed "OG self-proclaimed "OG ", I'm going to talk about the origin stories before ENS, the peak of ENS, the present, and the future.
Just a question: uh, how many of you were in crypto when, uh, there was, I think, a global event, the last time in 2023? Raise your hands. Okay, about half. Keep your hands up: how many of you were still in the crypt when we held the first community event in Lisbon in 2021? One.
Good. Stay confident. Good. What about 2016? Yes.
Only one. Two. Good. Three. O.
Oh, wow . Much more. Yes. Good. Class.
Uh, so in the pre-ENS era Class. Uh, so in the pre-ENS era , uh, 2016—I chose it because that's when I came into the crypt, it was about 10 years ago. And the major events of that time took place in March. There was a Homestead Ethereum release, which was kind of one of the first, uh, consumer-ready versions, although it wasn't really quite ready, but it was a big release. And we had the infamous, uh, DAO hack, where, uh, I think we created the DAO, and it got hacked and all that.
And that was the first time I actually bought crypto and invested it in this thing, and then I lost it for three months lost it for three months , and I just didn't know how to check the smart how to check the smart contract to get my money back. It was a very stressful moment. And in September there was the Devcon conference in Shanghai. Has anyone been to Devcon 2? No, just one.
Yes. And, uh, although ENS itself was launched later, in 2017, the seed was sown earlier. I asked founder Nick Johnson when he started writing this fast code, and he said, "Yeah, sometime in June or July." So, that was about 10 years ago; he came up with the idea, started writing code, and in September presented the first concept of a naming service. That was back then, and at that time, the biggest problem was scaling.
But I think scaling then and scaling now are different things; scaling then meant how to fully synchronize the entire blockchain on your laptop. When I first organized an Ethereum programming workshop, the first thing people did was download the software and just chat for 30 minutes or an hour until the blockchain synced. It was, but they considered it a commercial solution commercial solution , but it was a scaling problem. And one of the solutions he came up with was, in fact, MetaMask. They just celebrated 10 years, and when they came out, the reaction was mixed.
Then they also launched Infura, which is a centralized RPC provider. So from a purist perspective, "Oh, if you don't have a full blockchain on your laptop, you're not using blockchain." But at the same time, it was amazing, because now if you have a browser, you can interact with Ethereum. It was a real breakthrough and I was very excited at the time. And a year later, ENS appeared.
This is what the logo was like then. And the problem we were solving then was very simple. You have a name and want to find out which Ethereum address it is associated with, which is still called forward resolution . And if any of you are participating in a hackathon, this is the first level of integration, the simplest. We still consider this the most important thing.
How you use it: you open MetaMask, enter my token, see my address, and can send me funds. That's all. I remember in 2017 we held a seminar where there were people from MetaMask, Infura, MyEtherWallet and a few other companies, and if you integrated with them, then, in essence, you integrated with the entire world of the Ethereum blockchain. Those were simple times, and five years later, it was ENS' time. So, I think only half of the people now really remember what it was like at LisCon.
This was the last time I visited Lisbon and the dynamic was completely different. Everyone was excited, everyone was talking about, like, "Oh, let's get a 'golden visa' to move to Lisbon." I think those who rode that wave have probably stayed here ever since. And then, in November, ENS launched the DAO and token. The real peak of ENS came a year later came a year later .
We had the highest number of registrations number of registrations , but that year also became a kind of turning point for ENS. So I call it the peak of ENS, and then one of the most striking stories—does anyone know Dune Analytics? This is a kind of on-chain on-chain analytics, where anyone can write an SQL can write an SQL query for visualization. And as an OG, you essentially try every first product. And since I was a newbie, I think the previous Christmas they gave me this nice denim cap.
And six months later, when I joined LisCon and went to a side event, someone offered me four ETH for that denim cap, and at that time one ETH was worth $4,000. So, that was $16,000, which my wife liked so much. She wore it on the beach, and I said, "Don't do that, because they'll just rip your head off and steal everything." But I don't think anyone cares about that right now. But it was one of those peak stories that half of the people here still remember— here still remember— that excitement.
And at that time, we began to solve another problem that we had no idea existed . When we launched in 2016, we did a direct resolution that mimics the DNS domain name model on the internet, i.e. name-to- address translation. But if you can link a name to an address, you can do the opposite.
So Nick Johnson just added it without thinking too much. But then, it's actually the reverse resolution that we have, the namespace called the address reverse namespace, which has a reference to your name, that's all. And that's become the same usage pattern where people connect to many websites, and you know, like in Portugal, like Pedro with WalletConnect, he did a critical part of connecting the wallet to the site, which is becoming kind of the standard way of interacting with the smart interacting with the smart contract and the Ethereum system. And when you connect, or the site actually knows the address, you can just use a reverse lookup. Here, please, you have a name.
So suddenly your DNS name becomes your login, and we've added ways to link it to your avatar. So you had a JPEG and it's not CryptoPunk. This is the so-called CryptoPunk "bust" that someone gave me. I didn't have the guts to buy CryptoPunk, so I had nothing, but had nothing, but anyway. So then it suddenly became something else.
We reinvented ourselves as a Web3 identity without changing anything in the protocol because people discovered the way to use it themselves. And I don't know how many people used POAP; you know, Patricio, the creator of POAP, is here. And every time we go to an event, we also create something called an ENS card, where you can tap it and get this awesome POAP, and these are all animations that I created. It was a wonderful time. But this great success brought a new problem of scaling, as we know it now.
When people say " people say " scaling," that's what it is —the price of gas. It seems like it was 2021 or 2022; registering an ENS name for $5 cost over $100 in gas fees. So how did we decide this? We used, you know, layer 2, sidechains, all that to actually distribute the workload. So, this is the latest transaction counter transaction counter .
I think the bottom part . I think the bottom part is Ethereum. It is not quite stable, but it is growing a little. Then purple is Polygon, blue is Base, and yellow is now the Robinhood chain. So overall we have significantly more transactions.
It says 25 million transactions per month, which is much more than in 2021. And then that brought a new problem. You know, when I showed that one line of name, sir, one line of name, sir, address, that was the decision for 2016. But now you have to think about which, uh, coin to choose on which chain. These are known types of bridge protocols, uh bridge protocols, uh , I think, Across and Jumper, uh.
They, they do the same thing . The only thing they do is swap the coin and the network. One person first chooses a coin, the other chooses a coin, the other a network, but there are still too many options. So yeah, this is kind of a new problem, a UX problem, that we're trying to solve. And we, uh, we, since I'm on the ENS protocol team, we're trying to solve this in a different way.
Multicoin this in a different way. Multicoin is a way to specify many different types of addresses for the name ENS, uh addresses for the name ENS, uh . CCIP Read is a way to allow you to issue subnames on different chains so that you spend less gas on name generation. And I must say, name generation. And I must say, these two are already in use.
For example, I would call Base ENS one of the successful examples, and it has, I think, over a million names, er, it seems like seven million, much more than .eth names. So it's successful. But the de So it's successful. But the de facto standard that we ultimately chose is a certain approach from the application side.
What does this mean? Here is the Wallet Connect integration with ENS that I made back in 2020. There's, you know, MetaMask, Wallet Connect, or Yale. Then in 2026, three sites appeared: Pump fun, Market, and Network School, uh. It seems there was news that they were shut down in Malaysia, but now they are headed to Kazakhstan.
I was there in May; It was a very interesting experience, and they still use a similar model. The priority has changed: they are no longer promoting this universal wallet for connecting anywhere, but simply offering to log in via Gmail. And Network School—their app is great; they actually came up with a wallet, but they don't even offer to connect an external one to connect an external one . They just use Privy as standard, they just plug in, and they have wallet functionality. So, it's a bit of a mindset shift.
And I think we lost a little bit along the way. You still have the block as a blockchain; we have layout, ownership, identity, but it's all hidden in the name of simpler UX. I think that's something we've lost to some extent. I mean, there are still people who say "good morning," but now I feel awkward, so I stopped saying it. But you know, that's what we were losing, uh.
And I, I do n't know if it will come back; Maybe he'll come back, maybe he won't. So, the last part is the present and the future. What we have now, in 2026 have now, in 2026 , is a bear market, and we have agents. Instead of moving here, I decided to move to Dubai last year. So, you could say I was hit directly.
I don't think blockchain can solve this, but right now, everyone seems to be obsessed with AI, AI obsessed with AI, AI agents. As I said, I created this avatar myself in 2021, and these are two of my agents. Let me introduce Makito, created by Gemini Nano, and it is based on my apparently open source agent code. And this is my, uh, har-agent. I think everyone has one or two agents, and the role of identity in agents—it's a strange world where you can have an identity per model, per AI model, per AI framework, or if you use something like Cloud Code, every directory where you clone a repository becomes an identity.
I started doing this, I got a bunch of terminals, multiple sessions in the same repository, and I have a messaging system for them to communicate with each other. Then it's not even a directory anymore, it's just a role; When I do something, I create, for example, a code researcher or a reviewer, and that role also becomes an identity. So yeah an identity. So yeah , crypto before AI—it was a pretty straightforward experience, but now things are becoming very disparate. There are so many different agents between a human and a service, but honestly, are we using AI in the crypto world yet?
Not quite. For my precious Makito and Mashido, I'm a little worried that they'll be broken right away. I don't know how many people started using them; if you use something like Poly Market, you might already have billions of agents doing everything in a row. But I'm still worried: how do they trust each other, how do they know which chain to use, and how do they deal with paying for all of this? And for this solution in Ethereum, there is not only ENS, we have several options.
I think the first one is X42, introduced, I think, by the guys from Coinbase. They essentially use DNS, which we modeled as an internet domain, as a kind of identifier, and they started using the old protocol 42, which was never used, to map payment data. And that's good; this is more geared towards users not related to cryptocurrency. But this is somewhat limited for each URL, so this is just one way to go. They have 75 million transactions in the last 30 days.
And we also have ERC8004, which is Ethereum's answer to agent management, which allows you to indicate that there is a registry on the network. Anyone can create such a registry on the network where metadata can be stored, such as where your MCP endpoint is located, or reputation, but how will agents be able to find the registry if there are many of them? So the little addition that ENS has added is a little extension to how we already configure text entries in ENS names. You simply use this to link the X8004 registry information to your agent. And we also have 26 agent.
And we also have 26 , which is an easy alternative to X8004 alternative to X8004 . So we have some, and also the basic things— also the basic things— now contracts have their own names, you can name the contract itself. This way you can let the agent detect them. I think it's important to understand this. In the previous era, the problem was excessive complexity.
So let's simplify it by hiding all the benefits of Web 3. But now agents can handle it, agents are smarter than you. You have all these possibilities; they can handle the complexity, but you need to name your contracts or roles in these registries or naming systems to allow agents to discover them. I think this is where we are now and where we will be in the future. Still, I think this is Still, I think this is probably the last barrier to getting a feel for what it's going to be like.
I think we are at a turning point in whether cryptocurrency will become relevant in the Asian region, not yet, not yet. But I think things like this hackathon are designed to solve this problem, so if you can offer a service offer a service that uses existing standards to fill this gap, I think we still have a chance to become significant in the field of AI agents. Well, I think so, just to sum it up. So, 2016 , uh, yeah, the naming role was very simple, it was just, uh, yeah, defining an address, and that was 10 years ago, five years ago, it was about kind of representing ownership. But now I think the role of the address or the name itself is becoming simply a focal point to allow agents and people to discover new ways to use AI and the internet.
That's all I wanted to say, and thank you very much .
Automatic transcript — names and jargon may be misspelled.
