If DeFi Is To Really Take Off, We Need Smarter Contracts

The state of the decentralized finance industry today can be compared to the dawn of the aviation industry, and like the earliest pioneers of flight, it's chasing a solution to the wrong problem.

At the end of the 19th century, pioneers of human flight such as Otto Lilienthal, the man who invented the first heavier-than-air glider, and the legendary Wright brothers Orville and Wilbur spent years devising ways for humans to master the skies. They were incredibly successful, with Lilienthal becoming known as the "Flying Man" of Germany after developing the concept of the modern wing and carrying out over 2,000 successful glides before his demise in 1896. Following in his wake, the Wright brothers wrote their names in the history books when they succeeded in launching the first sustained flight with a powered, controlled aircraft in December 1903.

Lilienthal and the Wright brothers successfully took to the skies, and in the process, changed the world as we know it. But they cannot be credited with creating the modern air transportation system that ferries people across the globe today. These trailblazers were focused on taking off and staying in the skies, but it was only with the invention of the modern jet engine several decades later that we were able to answer the question of how we can safely and reliably fly hundreds of people across the oceans.

It's a similar situation in DeFi today, where the vast majority of blockchain developers are trapped in an old way of thinking. Yes, decentralized finance is possible, but the way it works now will not enable DeFi to replace traditional finance.

Today, almost every blockchain relies on "messages" between smart contracts that update their balances. The core innovation of decentralized ledgers is to enable the free flow of assets, yet the current approach does not allow for these assets to flow efficiently or safely because most existing platforms don't inherently understand what an asset is. This means that they struggle to handle digital assets in much the same way as the Wright brothers' Kitty Hawk biplane would not be able to cope if fitted with a modern GE turbine engine.

Current Smart Contracts Can Only Fly So Far

Traditional DeFi is built using a "message-oriented approach" that has no concept of what cryptocurrencies actually are. Because of this, it means that your crypto wallet doesn't actually "hold" your assets. "Wallet" is really the wrong word, for what it actually does is act as a gateway that provides access to digital assets that are located somewhere else entirely. The majority of crypto owners don't even know where they are located.

Digital tokens are, in fact, just a creation of smart contracts. To create a token on Ethereum, developers deploy an ERC-20 contract that creates and maintains an internal list of balances for account addresses, which define who "owns" each token. So, this list might look like:

Alice owns 50 USDC, Bob owns 23.4 USDC and John owns 6 USDC. When Alice decides to "send" 5 USDC to Bob, she doesn't send those assets from her wallet to his. Rather, she sends a message to the smart contract, essentially telling it to delete 5 USDC from her balance and add it to Bob's.

In other words, users don't actually hold or directly control their tokens. They're completely reliant on the smart contract and its ability to maintain its list of balances correctly.

This kind of smart contract architecture might work, but it's never going to be enough to power DeFi on a global scale, just as Lilienthal's first glider and the Wright brothers' biplane were never going to form the basis of a global aviation industry.

Traditional DeFi with its message-oriented approach is too complex and too insecure with a convoluted transaction process that's entirely reliant on the function of smart contracts. Unfortunately, the current crypto ecosystem is one that's rife with hacks, where attackers invariably exploit some kind of bug in a smart contract.

Recently, the cryptosphere has seen a big jump in so-called "spend approval hacks" that drain users' accounts when they're interacting with a dApp. Due to its inefficiencies, the message-oriented approach used in traditional DeFi requires users to give "approval" to any smart contract they interact with to withdraw any number of tokens from their wallet without their explicit say-so. When transacting on a dApp such as Uniswap, for example, users are allowing it to spend their assets on their behalf and trusting that it won't do anything naughty.

So when someone tells a Uniswap smart contract to send tokens to another address, they're really giving Uniswap permission to withdraw as many tokens as it wants — and simply trusting that it will only take out the ones it's authorized to. But if that smart contract contains a bug that can be exploited, as happens way too frequently throughout the DeFi ecosystem, hackers can withdraw more tokens than the user has authorized, resulting in their balances being fully drained.

It's all caused by the fact that DeFi doesn't know what an asset is and that people do not "own" their tokens the way they think they do.

It's an alarming problem, with Rekt.news stating that the top 100 hacks in the crypto ecosystem have cost users a combined $6 billion. This is clearly not the foundation of an alternative, global financial system. There's simply too much money that goes missing, and it's all due to the fundamental architectural design mistakes of crypto's earliest innovators.

DeFi Needs a Jet Engine

Just as with aviation, where the advent of the jet engine took flight in a whole new direction, DeFi can be transformed with an entirely new architecture. It requires an "asset-oriented" approach that treats digital tokens as native assets that are built into the platform itself, rather than something that's implemented thousands of times in smart contracts.

With an asset-oriented approach to DeFi, users will actually hold their tokens within their own smart contract account on-ledger. It will work just the way most people imagine it does. Just as you don't need to give your bank permission to spend a physical coin in your pocket, with asset-oriented DeFi, you no longer need to approve other smart contracts to spend your tokens. Spend approval becomes a thing of the past.

Furthermore, with the platform natively understanding what an asset is, the user can define exactly how many tokens they intend to pass or receive back from smart contracts rather than simply "asking" and hoping that it does as it's told.

It's a revolutionary approach to DeFi that's not just safer but easier to work with, too. Developers can create dApps with exactly the same kind of functionality as if they were building on Ethereum, only the process is much simpler. Using components and blueprints, developers can implement powerful logic for financial systems, games, payments apps and more by using pre-written "resources" that can be configured for their specific needs around tokens, stablecoins, NFTs and other assets.

It's a vision that makes DeFi simpler and safer with a more sensible, logical approach that enables wallets, tokens and smart contracts to function as the user would imagine. By making dApps easier to build and eliminating the possibility of smart contract hacks, asset-oriented DeFi can serve as the foundation of a platform that's able to scale globally to support potentially billions of users.

Orville and Wilbur Wright made a historic contribution to the world by inventing the first powered airplane. But as impressive as it was at the time, the one-seater Kitty Hawk Flyer was really just a stepping stone toward the modern airliner. In the same way, the original smart contract platform devised by Vitalik Buterin is really just a proof of concept. Ethereum might have helped to get DeFi off the ground, but we need a jet engine to take it into the stratosphere.

(Jeremy Epstein is the chief marketing officer of Radix.)