Unlocking the App Layer: Decentralized Commerce & Beyond

Transcript

Good morning everyone. I'll be talking today about some of the research I've been doing at the Ethereum Foundation on basically how we can anticipate and unblock systemically important use cases and then specifically on the use case I've been focused on this year which is decentralized commerce. So I want to start kind of in the 90s when the smart contract concept was born because I think a lot of time we overlook the magic superpower that all these apps are actually u tapping into and a lot of the time it's easy to get caught up in the constructs that have developed you know over the last 10 years of Ethereum um NFTTS tokens Dows and really these abstractions that make

us forget the kind of fundamentals of what a smart contract is and how many people here actually read like Zabo's early blog posts on smart contracts. Okay, a few. Yeah, I definitely think this should be required reading like in addition to the Bitcoin white paper for anyone who works in the space because the the way in which it was thought about back then was I think quite different and interesting from what it is today. So kind of Zabo talked about smart contracts is these automatons uh similar to like a vending machine or a point of sale terminal that you can deter you can interact with and then you know the deterministic outputs that you're going to get. They have these bakedin rules that don't change for how you interact with them.

And then about 20 years later, Ethereum was born with this vision of being a general purpose platform for smart contracts with really some of the early use cases imagined being the internet of things and e-commerce. Basically, we would have these smart agreements baked into devices and systems that were either ownerless or maybe collectively owned. So you have these guarantees about how they're going to behave unlike the systems that we're used to where we rely on a trusted third party like a corporation to run the code in the way that we expect.

Yeah. And so okay some formatting issues but a couple years later uh 10 years later we can see kind of where we've gotten so far and how close we are to realizing this vision and also where we might have been off the mark in the early imaginings of what smart contracts will enable. And so of course we've started to have our first killer apps you know mostly you know the first two kind of mostly in the realm of um serving other crypto infrastructure. So capital formation for crypto projects collateralized borrowing of cryptonnative tokens. And then recently we've started to see the breakouts of things that go beyond just the crypto bubble, the self-reerential loop to enabling people around the world to use

stable coins for savings, remittance, and then now even payments and as well as bringing other real world assets on chain. But then beyond these purely financial use cases, um there's also this interesting space of new use cases that are emerging that um take things into the the real world in different ways that that more normal people would interact with that don't have a specifically financial motive. So basically utility, right? So physical infrastructure um that's cheaper and more reliable than the centralized providers, social networks, and of course everyone's favorite prediction markets.

But to really become the backbone of a global economy, you you're going to need a lot more than that, right? So there's far more uh use cases that we saw people experimenting with early on, especially like with the ICO boom when it was easy for anyone to uh raise capital and sell a vision of how you're going to decentralize some sort of industry. But then we pulled back and I think the space became a lot more conservative uh due to kind of infrastructure limitations um of the time when these were tried. And the process has been quite messy, I would say, of of figuring out these use cases that work and that don't work. And understandably so, um, because that's how innovation goes with creative

destruction. But I think that it's gone, you know, a bit too messy and that there's tiny kind of interventions that could be done to tweak things to bring them on course. And one way of summing this up is I see this mentality today of you know on one side you have throughout the course of crypto's history people who are um kind of idealistic and thinking of how we can like solve all these systems problems. But the problem with that mentality is that there's no one in particular that is going to adopt those solutions. You kind of hope that people realize it's the better option and then that's not a strong enough compelling pitch to get billions of regular people. And then there's capitulation of just okay well the first things that work all had to do with

speculation. So that's the only thing we can do. And I think the obvious third way here is to actually think about real pain points that smart contracts are uniquely positioned to solve. And so I think this starts with thinking about personas. So who are we solving problems for? And this is just super high level. But then within this you know different geographies, different industries to understand like the problems of these different parties and how could a smart contract help um you know a citizen with their involvement in the government or their community um an entrepreneur who's trying to sell goods and services or startups, smallmedium enterprises. maybe their financing

issues, but also their business processes, the coordination problems in these different industries. And it helps to zoom out and think about the the world economy, right? not just the kind of sort of random industries that crypto has started to um penetrate, but this broader world that has been affected by so many previous ways of technology and kind of taking it seriously of how can we actually build solutions for these different industries. And so you have, you know, all of them over time gradually upticking software and then the internet and now AI. And I think many people have given up on actually building smart contract solutions for real industries

and thinking about smart contracts as a solution that and a new paradigm that can be embedded into different facets of the world economy rather than what it's come become today which is largely like a a different sector. It's it's the crypto industry is its own sector. It's its own community. But I really think this is the wrong way of looking at it. And basically now I think is an important time to to reopen that box of the of the enthusiasm and the excitement of the early potential because a lot of the pieces that were missing then have come into place. So on the left we have the the parts that people already knew from the early days that we would need right we need uh to scale we've done so through L2s and now the L1 and L2 interrupt account abstraction and smart

accounts make it way easier to onboard regular people on and off ramps have proliferated and specifically um we have newer on and off ramps that are more composable with the rest of smart contracts. So for example an IBAN account that gets filled up um like a European bank account when you send a stable coin to it. We have brought real world data on chain and privacy tech has been advancing a lot in the past two years. And then we also had regulatory clarity starting to happen in in Europe, the US and many other places that makes it a lot scary less scary to build these different kinds of pieces for real world industries.

In the meantime, we've also had the emergence of these new superpowers that we can layer into smart contracts to help them touch the world in more impactful ways. For example, AI agents. You can bring like immense amounts of data processing, subjectivity, human judgment, um machine judgment into smart contracts in a cheap and scalable way. We have social open social graphs that can create basically new forms of trust and reputation that you can't have with purely pseudonmous systems where everyone's in a atomic individual. Also with um with that you have the emergence of many apps as a distribution channel for new applications that use smart contracts.

ZKTLS is also a new way of basically bringing information from the real world onchain, specifically web proofs. And so you can prove that a server of some application like Airbnb, let's say, says that you're a superhost and claim kind of a an onchain identifier or rewards based on that. and broad more broadly programmable cryptography is giving us all these new gadgets that we can integrate with smart contracts to have privacy preserving uh mechanisms that um give you something more powerful than smart contracts which can be quite limited right and in addition to just basic regulatory clarity there's also these new sandboxes

and zones of experimentation where you can actually get support of local economies in rolling out these systems and lastly the L2s present like a compelling path for organizations, for institutions enterprises um governments to come on chain in a way that's plugged into the global onchain economy, but um, with more customizability. And so, quickly, we'll go through some techniques that we've been using to think about this design space of of use cases that could have been or will be, but have not yet kind of uh, manifested.

So, first idea mazes. This is this idea from from Biology, the the path that a startup goes through as they're trying to figure out uh what it is that their product market fit is. So, you start with an idea and of course that's not where you end up. You go through all these twists and turns, dead ends, you avoid traps that other companies have fallen into. And basically, we we need to think more in this way as an ecosystem about where smart contracts fit into the world. And then also for particular individual use cases and thinking specifically about what was tried maybe five years ago, eight years ago that failed but that we now have a solution in place to overcome.

This is from foresight institute the con concept of dependency graphs where you can understand the relations between different se sectors. This is a space tech tree and so you have different problems within space tech and you can understand how the different dependencies um interconnect. This is zooming in on one particular uh problem area which is debris mitigation. So you can see the capabilities um that are blocked by different challenges and then technologies that are needed that can be anticipated that come into place to open those up. And this effectively serves as a helpful tool for coordinating actors whether it's researchers, startups, funding organizations, investors to know where things need to go to have the

gigantic outcomes that people envision. And then we can use different kinds of um techniques like rice scoring to basically see the potential for these different use cases and the timing considerations of when these are ready. And then we can start to take steps on a use case by use case basis to envision the future that we we understand for that use case and then work backwards to understand the blockers that are still in place that we can take steps interventions to remove and to unlock new use cases that you know people have thought about for years but have been held back.

And I like this mental model instead of going forward iteratively which is you know what the market will naturally do to also step back and understand a more what's a more ambitious vision for a particular use case right so we have um things like stable coins today right which are a great leap from from fiat from financial rails that are traditional but what's a more ambitious vision of what you can do once you have stable coins and scalable platforms for smart contracts And the use case in particular that that I've started with is peer-to-peer commerce.

So real world markets for goods and services. So first understanding the problem so we can understand why our smart contracts can can solve those. So obviously smart contracts can disintermediate but they can also help scale different kinds of coordination and incentives to to have new networks gain adoption and smart contracts can also automate various business processes. Right? So we can see how the platforms of today that we we all use on a day-to-day basis we all know that they have these downsides and people have envisioned for years these alternatives but they yet haven't yet come to fruition, right? we've we've stuck with more financialized use cases.

So this is years ago, you know, Slocket with this vision of this um sharing economy that would be universal with smart objects that would control their own um access rights. Open bazaar is this e-commerce um substrate that would allow interconnected marketplaces on top. ride sharing uh ride hailing from trip and other projects that tried this on Ethereum and Dravel as a Airbnb replacement accommodation. Going back, this is the the concept of looking out to the future and p like trying to predict, anticipate and then create the conditions for the how we want to see this use case play out. So

we want interconnected markets, right? We don't want silos. We want inventory and reputation to be interoperable between different kinds of markets so that people can reuse their credibility and don't have to start from scratch and are at the whims of these algorithms and then we can start to create our own dependency graphs like we saw from foresight for use cases. So this is narrowing in on the base case for commerce which is just a simple merchant checkout flow. And the big question is why why haven't we seen this right? someone tried to buy uh pizza with Bitcoin years ago and and yet still today crypto isn't really being used for day-to-day payments.

But if we understand this way, we can see that's because a lot of these pieces only recently came into place to make this possible. So you need on and off ramps that are integrated into people's regular financial flows. You need regulatory clarity so that those um can operate uh without fear of being shut down or debanked. And with when these pieces come together, when you have stable coins everywhere, you have easy wallet UX, then it's it's time to kind of actually put the pieces together, you know, have an escrow that handles the kind of flows that you need for merchant payments.

It's a very just a few weeks ago, we saw this with Shopify and Coinbase uh releasing their commerce payments protocol that takes the basic uh flow of escrow for purchasing a good or service from a merchant online and builds it into smart contracts. So this can like reduce uh issues with uh that Shopify has as a payment processor basically reducing their cost uh reducing the chance of something going wrong because they're outsourcing to Ethereum the process of um validating that all the steps are being taken and then the superpower you have on top of this is because it's an open protocol because they're smart contracts now other systems can tap into the same flow and obviously with Shopify you can't right

it's a black And then we can zoom out to a more ambitious picture of what it would look like to do the entire peer-to-p peer commerce flow through onchain and adjacent technologies. So once we have merchant payments, uh you need other pieces of the stack to enable people to discover one another for listings to happen and for inventory to be matched. You need arbitration and different risk management solutions if something goes wrong. So, insurance and just dispute refunds and identity and credentials, reputation for understanding uh who you're interacting with and then having a feedback loop that updates that. And so, while this is definitely a bit further

off than merchant payments, which we see are accelerating, we can start to see the other enabling technologies and institutional changes that are going to enable this. So, the ones kind of highlighted here about identity and credentials. So specifically the combination of privacy tech um letting us now do selective disclosure so that you can build reputation algorithms on top of people's uh revealed credentials. You can use social graphs to build those reputation algorithms. And like we said earlier about ZKTLS, you can pull in people's um history on up from Upwork or from Airbnb or Uber or eBay or anywhere to then bootstrap the solve the cold start problem of of having a network effect

around the reputation data. And then social graphs also give us this new substrate uh and many apps where we can distribute the apps. We can already have integrated wallets and social graphs that these different apps tap into. And this makes it a lot easier to go from 0ero to one with marketplace applications as we see with these examples like slice which is for purchasing uh goods. So kind of like an on fully onchain Shopify and labor markets like icebreaker which also serves up as a professional market professional network but then uses zkls to import information from LinkedIn verifiably and the social graph of caster.

And not only can this help with reputation, but then this also can serve as the way that listings and um discovery happens as well. Other examples built on Farcaster of marketplaces both for labor and Farcaster marketplace which is a experiment through Farcaster channel to build a Facebook marketplace type experience. And with agents, they can help with a lot of different parts of the commercial life cycle. So you can imagine client side, I can have my agent running to figure out what um goods and services I need, right? Wait, if they're being hosted in a decentralized way, let's say through uh forecaster posts, then you

want to have a robust discovery engine because you don't have a platform in the middle that's algorithmically figuring out, you know, who should interact with who. the process of pricing, negotiation, reaching an agreement, all this can happen offchain and then be resolved onchain as an example from virtual's agent commerce protocol. And then payments, right? So agents can also help with um actually uh affecting the payment. So making the payment happen through APIs that are optimized for um for agents so that you don't need to fill in credit card information or um even like if it's an onchain solution that's fully onchain the maybe the

shipping information your agent already knows you know where the goods should go and then you can give the agent rules more examples that um give the agent limits on how much it can spend on your behalf And lately we start to see new protocols that are kind of bringing a lot of these new pieces together to realize that original vision of peer-to-peer commerce. And so I talked to a lot of these projects, many others that that we saw with the primitives to see how we can help with bridging the missing gaps.

For example, uh we collaborated with an artist Tessa to build a um installation at ETH Dublin recently that enables people to basically build an attestation graph so that transactions that you do um in this case offline can then be represented online and build a network effect and essentially a reviews and rating system. And this is a piece that a lot of these different u marketplaces aren't thinking about yet, but we anticipate it's going to be a blocker, right? Because for example with eBay um it only took off in 97 once they added the review system and so I think we're still before that point and this is just my call for

participants. So if you want to engage in any way on use case exploration specifically pitching um new use cases, use cases that haven't gotten enough attention that we can collaborate on um bridging the gaps, taking steps to realize them and then looking across use cases to see the the keystone blockers that are intersecting between many different use cases and basic technical and institutional steps that can be taken to unlock them. Thank you.