Compiled by Kaori, BlockBeats
Editor's note: Base TVL has surpassed OP Mainnet to become the second largest Layer2 in the past two days, and it has also launched the Onchain Summer series of activities as scheduled. In this conversation, Brian Armstrong, co-founder and CEO of Coinbase, and Jesse Pollack, head of the Base protocol, talked about how to promote innovation and development in the blockchain field. Brian shared his unique perspective as an engineer and CEO, discussed Coinbase's efforts to help developers, and his outlook on on-chain development in the next few years.
Jesse Pollack: Hello everyone! Welcome back to Onchain Stories, a special episode hosted by me, Jesse Pollack, founder and head of Base. Today we are honored to have Brian Armstrong, co-founder and CEO of Coinbase. Welcome, Brian.
Brian Armstrong: Hi, Jesse, great to meet you.
Jesse Pollack: Thanks for joining us for this conversation and inspiring developers, let's start from the beginning. Ten years ago, you and Coinbase were on a mission to make crypto more accessible. Can you talk about your thoughts? How far have we come? What is the user experience on the chain today?
Brian Armstrong: Yeah, we've made a lot of progress, but I think we still have a long way to go. At the beginning, we were thinking about simple problems like how to send Bitcoin to an email address.
I remember in 2012, when we were working on an early prototype of Coinbase, we were inspired by PayPal — sending money via an email address, and maybe we could apply that system to Bitcoin.
At the time, we were just trying to solve some very basic problems, like storing bitcoin with a username and password without having to run a node on a laptop, which were very basic things at the time.
But we've made some progress over time. For example, now instead of sending those long addresses that look like passwords, we can send ENS IDs, which are human-readable rather than machine-readable, which is great. Layer 2 is also a big step forward, and Base is an important part of it, but there are a lot of Layer 2 solutions like Lightning Network that are working to make this process easier.
We need faster confirmation times and lower fees, but I think we still have a long way to go. One of the big problems is that cross-chain asset transfers are still difficult. People don't know how to transfer assets between L1 and L2, or between different L2s.
There are a lot of jokes about this on social media, like, is my USDC on Ethereum or some L2? There are so many icons, and I feel like we need to find a way to make this seamless in the background so people don’t even need to know which chain they are on.
Maybe there could be some kind of handshake between the sender and the receiver that says "here's what I have, what do you accept" and then the exchange is automatically done on the DEX. Something like that might help. My standard is something like Apple Pay where you can tap your phone to pay at a point of sale device.
How can we get to this point? Like paying between two phones, or scanning a code to pay in an e-commerce environment. At Coinbase Commerce, we’ve been working hard to achieve a world of “one-click magical checkout” where cryptocurrency is the easiest way to pay.
We are not quite there yet on L2, we need to debug some of the APIs built into the hardware and phones, and there are probably a lot of details involved like NFC, etc. But I think we are getting closer, and it requires hard work from a lot of smart people.
Jesse Pollack: Yeah, that's right. There are a lot of people who are building on-chain for the first time this summer, and some who have been in this space for a long time. What are your expectations for this summer's on-chain build-a-thon and some of the projects that are currently being built on-chain?
Brian Armstrong: Yes, I have a few ideas in mind. Of course, I am not the judge of these good ideas, I am often blown away by the creativity that the community comes up with. We often talk about these Lego-like things starting to fit together. For example, we have ENS, stablecoins, L2, identity, NFTs, and social.
Another piece I think might be missing is a reputation system on the chain. You can imagine a version that uses the graph structure of the chain to represent trust relationships, such as I trust this node, they send funds to another node, which implies a certain degree of trust, similar to the Google PageRank algorithm, which would be cool to build on the chain.
Another thing I've been thinking about is that in Web2, the biggest business model was advertising. We don't want to repeat all the mistakes of Web2, but how distribution is achieved in Web3 is an interesting question. You can imagine an ad network or the equivalent of Google Adwords, but instead of just CPC, you can pay for an action.
You could imagine a world where smart contracts or dApps expose some metadata like, if you refer a user to this function call, passing your referral ID or address, you will get a 1% reward. You could advertise the different rates people are willing to offer, and those rates could be displayed in various wallets or distribution areas.
Jesse Pollack: So I think we might see ad networks develop in that dimension. One more thing I've been thinking about is on-chain P2P deposits and withdrawals. A lot of emerging markets need to transfer money in and out of platforms, and these centralized versions are often shut down or under pressure. If we can do all of this on-chain, such as dispute resolution, mediation, rating and review systems, and build more of these on-chain systems, these markets will become more resilient and global. These are some ideas I've been thinking about, but I'm excited to see what ideas people come up with.
Brian Armstrong: What excites me is that the technology was always not ready before, we didn't have the right components, the chain fees were too high, we didn't have something like ENS, and the wallets were too difficult to use. But now it feels like we finally have the components, which will enable the next wave of use cases. For example, now you can really build reputation systems, advertising networks, or deposit and withdrawal systems, and they are low enough and easy enough for everyone to use.
I have two ideas, the first is that the technology is finally ready to support reputation systems, ad networks, and deposit and withdrawal systems. This will make these use cases cheap and accessible for everyone.
The second is that it's really interesting to think about how these things really fit together. To have a good ad network, you need to know if the people who are engaging and seeing the ads are real people, which is reputation. To have a good P2P network, you need to know if the people in the network are good actors, which is also reputation.
So I think we're at a stage now where the on-chain infrastructure is already built, but we're starting to see opportunities for this application-level identity infrastructure that will enable P2P, advertising, and better commerce, which will help the world move on-chain.
Jesse Pollack: I know you're getting more hands-on at Coinbase, making it easier for people to get on-chain, not just from a user perspective, but also from a developer perspective. I know you're an engineer yourself, can you talk to us about what Coinbase is doing right now to help developers and what drives you to invest your time and energy?
Brian Armstrong: Well, if you guys haven't checked it out yet, you should check out CDP, the Coinbase Developer Platform. It's at cdp.coinbase.com. We're trying to make a lot of the tools that we've built internally at Coinbase available to third parties to use to build their applications.
We've built a lot of things over the years, whether it's a simple way to store crypto, or staking, or trading, or even a way to index a lot of NFTs or handle ENS and different assets. Hopefully, we can reduce the workload for developers and provide these features through a very easy to use API.
Jesse, you posted a really great video about a lot of the on-chain builder tools. We've got Base nodes and Paymaster in there, where you can make fees completely invisible to your customers. Obviously, the launch of the Smart Wallet yesterday was also a big deal. I think now we're seeing that instead of having to onboard with a 12-word seed phrase, you can just tap in with biometrics or a YubiKey, which makes it a lot easier to use and onboard with the wallet.
I'm very passionate about this and I think there's a huge opportunity for Coinbase to help thousands of developers build applications, which is one of the reasons why I've been deeply involved in the Coinbase Developer Platform CDP product, to try to see how we can make it easier to use.
As a developer, I always love those magic moments, like the first Ruby on Rails demo where I could create a blog in five minutes with features like live comments and RSS feeds. These tools increase productivity. We want to be able to provide developers with a range of these tools to make it easier for them to get started and lower the barrier to innovation.
Jesse Pollack: Absolutely agree, and if people don't know, Brian was an engineer when he started Coinbase, and he's still an engineer today. We recently had a three-day weekend for Memorial Day, and I got a message from Brian on Sunday, and he asked, "What's the best way to learn how to write smart contracts?" I recommended some resources to him, and later we talked about it in messages, and he said, "Can we code together?" And then he and Wilson Cusack spent their vacation writing smart contracts.
I think it's very inspiring for me that Brian, as the CEO of a public company, is personally involved in writing smart contracts and improving developer tools. In the past year, almost no one thought about what Coinbase was doing for developers, but now Base is the leading L2 for developers, and the Coinbase Developer Platform is a very powerful developer tool where you can find everything you need to start building on-chain applications. We are just getting started, and I think it all comes from having a leader like Brian who understands the technology and can personally write code to make everything possible.
Brian Armstrong: Thanks for saying that, Jesse. I feel like my programming skills are getting worse every year, but hopefully my CEO skills are getting better every year. I'm trying to lead by example, and I had to learn the basics myself. Everyone is a little intimidated when they learn a new language or way to build an application.
This is complex stuff, so you have to learn it with a beginner's mind, experience some frustration, and ask a lot of seemingly stupid questions. Thanks to Wilson Cusack for helping me with some pair programming, which was very cool. In Singapore, some officials, like the former Prime Minister, have written some things on GitHub. This attempt to dig deep into the basics shows good leadership within the company, and I think we need every engineer at Coinbase to build on-chain. If I don't lead by example, I can't ask others to do it.
Jesse Pollack: Absolutely agree, your example is being seen and felt, and it's changing the company. About a year ago, we didn't have many developer products, and we didn't have many teams building on-chain within the company. But today, there are many teams building on-chain within Coinbase, and we say we want to build the best smart contract on-chain engineering team in the world.
If you had asked us a year ago if we could do this, we would have thought it was crazy, but now we are getting there, pushing the boundaries of what is possible, and setting an example for ourselves, our internal teams, and the entire ecosystem on how large companies can improve their on-chain building capabilities.
I'm curious how you see the future of on-chain in the next few years and how quickly do you think this will all happen?
Brian Armstrong: I think these things always feel very incremental while they're happening, but looking back on them they seem sudden. People think of the iPhone launch as a sudden turning point, which it was historically, but it didn't feel that way to people who were in the thick of it at the time. The iPhone didn't even have an App Store for a year or two after it launched, and it only ran on AT&T's 3G network. So it felt very incremental at the time, and only in hindsight do we see how dramatic these changes were.
I think we are at such an inflection point, but it feels very incremental when you are in it. L2 is a huge step forward, it reduces the friction to build different applications. What I have been thinking about recently is how all these decentralized protocols can come together in on-chain super applications. We need a moment like Netscape or iPhone, but not just for payments.
We are seeing the rise of decentralized social media like Farcaster, and interesting messaging apps like XMTP coming online. How can these be embedded into people's daily lives, such as their livelihoods, paying for basic necessities, daily operations, etc. Content creators, social media, journalism, and even the way organizations are created, voted on, and governed may be affected. I think this is very exciting.
If we can achieve this, there could be an on-chain super app that brings these protocols together, which would be really interesting and I’m excited to see what you guys do. It really feels like all the infrastructure is in place, but the user experience hasn’t really brought it all together yet, and there’s a big opportunity here. I’m looking forward to continuing to build out the developer and builder side of things and seeing what Coinbase does on the consumer side.
Jesse Pollack: Last question, we're doing a hackathon right now, thousands of developers signed up to build over the summer. You're a very successful founder who's been doing this for a long time, and I'm wondering what advice you have for these developers who are building at a buildathon?
Brian Armstrong: The early stages of innovation aren’t glamorous, it’s just a few people tinkering in their apartment, which is where Coinbase started, and some of the ideas I’ve tried in the past that didn’t work out, which is the norm.
I remember my friends would call me and say, hey, come out to this party. I would always say, I don't know, I would just be at home tinkering with my laptop. But I didn't feel like I was changing the world, in fact, most of the time I was struggling to fix a bug, probably because of a misplaced semicolon.
So a lot of innovation is like this: you need to persevere and allow yourself to be a beginner, allow yourself to slow down and not understand how it works. If you're on the cutting edge, there aren't a lot of manuals to refer to. You can learn to make an LLM in five minutes, but if you really want to build something new, those resources don't exist. So you have to overcome these difficulties by persisting. Perseverance is probably the biggest determinant of success in starting a business and building something.
I think a lot of startups, they get a few people together and try something, and three months later, the first product that comes out doesn't work, or there may be some internal co-founder disputes, and they crash and burn. In my opinion, that approach almost never works, almost nothing works the first time it's launched, it's very rare.
Coinbase’s first product certainly wasn’t a success either, you have to launch a very embarrassing version 1 of something that isn’t actually ready. If you’re not embarrassed about it, you’ve waited too long. A lot of people just can’t get over the hurdle of launching a slightly embarrassing product and never launch anything.
Once you release version 1, nobody cares, release it on Farcaster, get 10 people to come and look at it. Then you have to keep looping: talk to customers, improve the product, talk to customers, improve the product, keep doing this for two, three, four, ten years, and eventually you might have a breakthrough. That's the game of entrepreneurship, you don't know if it's going to work. You have to be willing to go into the unknown and keep trying enough ideas and eventually something will work. And 99% of people give up, or they never start. That's the way.
Jesse Pollack: I like this summary, it's a great way to end. If you're building, don't give up, you're only on day one. The most important thing is to stay determined, stay focused, keep your eyes on the goal, and stay based.