Tuesday, 28 March 2023

‘The Only Way to Truly Scale Blockchains Is to Parallelize Processing’ Says Piers Ridyard

by Earn Media

‘The Only Way to Truly Scale Blockchains Is to Parallelize Processing’ Says Piers Ridyard

Despite being touted as the possible panacea, decentralized finance (defi) still faces obstacles which greatly diminish the prospects of mainstream adoption, asserts serial entrepreneur and CEO of Radix DLT, Piers Ridyard. Ridyard added that while defi is seen as “a fantastic proof of concept,” widespread adoption of this alternative to traditional finance is only possible when the developer and user experience is improved.

Developer Incentives and Mass Adoption of Defi

Besides improving developer and user experience, the Radix CEO told Bitcoin.com News that the provision of ongoing and sustainable support to developers ensures “you [don’t] end up with a ghost chain.” Ridyard, a YC Alumni, also shared thoughts on how defi and Web3’s scaling woes can be overcome.

Ridyard further discussed Coinbase’s attempt to bolster developers with its recently launched layer 2 (L2) blockchain and why this is unlikely to result in the envisaged mass adoption of defi. Below are the CEO’s answers to questions which were sent to Bitcoin.com News via Whatsapp.

Bitcoin.com News (BCN): What do you think are the biggest obstacles facing defi today?

Piers Ridyard (PR): There are two major obstacles. Firstly, the user experience of Defi is completely unacceptable for the everyday person. Secondly, the developer experience is so difficult that very few developers actually get to the level of being able to create secure smart contracts.

That makes Defi today a fantastic proof of concept. As seen in Defi summer, there is no shortage of innovative ideas that provide real benefits to users and capital. It’s still very much a proof of concept though. Week after week, headlines of multi-million dollar exploits of Dapps hit the news.

A quick search on Twitter will show examples of experienced users having their wallets drained because they have to blind-sign transactions. And if you’ve ever tried to onboard a friend or family member to crypto/Defi, I don’t have to tell you that things like seed phrases are far from something the majority of humans will be comfortable using to secure their net worth.

Just with all good proofs-of-concept, we can see clearly how it can work, but it’s far from ready for mass adoption. The biggest obstacle for Defi is taking this proof-of-concept and creating an experience for the developers, entrepreneurs, and their users that gives them confidence when engaging with the Defi ecosystem. To do that, we need both a developer and user experience that is intuitive, secure, and scalable.

BCN: It has been said that developer incentives are important for driving the defi ecosystem’s growth. How do you incentivize developers to stimulate growth?

PR: Developers are the leading indicator of future ecosystem success. The more high-quality developers you have in your community, the more Dapps are eventually built on your platform. Many projects have tried to attract developers with big developer funds or grant programs. The idea is that if a successful Defi ecosystem needs many types of decentralized exchanges (DEX), lending, non-fungible tokens (NFT) or derivatives applications, you can create a fund to incentivize developers to build them.

What transpired however was that many L1 blockchains threw millions of dollars at developers who would build-to-specification, ticking all the boxes to get the funds. And the moment this was achieved, the developer would then stop work. The DEX would be there, but it wouldn’t be supported going forward. You end up with a “ghost chain.”

How is Radix different? We believe in sustainable incentives. That’s why we’re building an on-ledger automated royalties system that pays developers each time their code gets used by someone else. This incentivizes developers to build the primitives that they think will be the most useful over the long term, harnessing the power of market forces to guide what gets built on the network, instead of a central authority deciding this by handing out cash.

Having said this, developers and entrepreneurs do still need active support. That’s why the Radix grants program combines services, support, guidance as well as cash subsidies to founders and developers in the Radix ecosystem.

BCN: Coinbase recently announced a new layer 2 blockchain called Base to give developers an easy, low-cost way to build dapps. What impact will this have on defi adoption and how will it compete with/affect other layer-2s?

PR: So Base is an interesting development. It’s Coinbase leaning into centralized Defi, or “Cedefi” as some call it. But I would argue that it’s not an easy place to build Dapps. Nor will it be low cost in the long run. Why?

First, Dapps built on Base will run on the Ethereum Virtual Machine (EVM). While the EVM is undoubtedly the most popular environment for developers to build Dapps today, it has proven time and again that it is not safe, with billions of dollars worth of hacks over the last two years ($200m for Euler Finance in just the last week).

To provide an easy developer experience you need to look past the EVM to new environments that give developers the tools to create and manage assets, i.e. tokens, with security, validation, and accounting handled by the platform itself. If the platform is handling assets, not the developer’s smart contracts, many of the vulnerabilities that result in these hacks and exploits just aren’t possible.

Second, as a Layer 2, Base is ultimately just a new blockchain. That means it doesn’t add to Ethereum’s scalability, as none of the Dapps on Ethereum can be used directly on Base. And none of the Dapps on Base can be used directly on Ethereum. This is because you lose “atomic composability” (which we’ll talk more about later) between Ethereum and Base. As a result, Base will have its own instances of each Dapp, such as new DEXes with their own pools of liquidity, brand new lending Dapps, etc. Ultimately, if Base gets popular enough, it will reach its own scalability limits, and transaction fees will start creeping up again.

In terms of impact on Defi adoption, Base is definitely a good thing. With Coinbase’s brand and resources, it will encourage more users to “dip their toes” into Defi and get a feel for what it’s like. But with a limited set of permissioned validators, Base is not truly decentralized. It is useful mainly as a stepping stone to bring more users into the space. We won’t get mass adoption of Defi unless it is truly decentralized. The clue is in the name of that one.

BCN: On the topic of layer 2 chains, let’s talk about another critical growth problem for defi and Web3 — scalability. From layer 2s to sharding — most of today’s networks are in a race to scale. Do you foresee such solutions eventually working?

PR: So we touched upon this above, but to really delve in, let me paint a mental picture to help you understand why blockchains fundamentally don’t scale.

To begin, think of a block as a square that contains transactions. Once the block is complete, that’s it, all those transactions inside it are final. Any transaction inside a given block is able to be combined with any other transaction in that block. So for example, you could have a two-leg transaction buying and selling two houses: 1) Person A buys from Person B; and 2) Person B buys from Person C. In this scenario, the second leg cannot complete unless the first leg also completes.

For the transaction to work, you need to have a guarantee that both legs happen, or neither happens. And on a blockchain, you can only guarantee both legs completely when they’re both inside the same block. If leg 1 happens in one block, and leg 2 waits for another block, Person C could cancel the transaction and suddenly Person B doesn’t have a place to live.

Next, the only way to truly scale blockchains is to parallelize processing. There is a limit to how many transactions you can push down one pipe (think cars traveling down a single lane). With this limitation, the only way to truly scale is to build additional lanes. With an unlimited number of lanes or separate blockchains, there is in theory no limit.

But if you parallelize transactions across separate blockchains, you are by definition splitting your transactions across separate blocks. Our example two-leg house transaction cannot guarantee both legs if they are on two separate blockchains. So both legs of the transaction have to be on the same blockchain. But if they have to be together, what’s the point of parallelizing processing in the first place?

This is effectively what we have with Ethereum today. Everyone wants to be on the Ethereum main chain as everyone wants to be able to “atomically compose” with everyone else. If you’re on a shard or layer 2, you’re effectively on a lane that only a few people want to be on. You can’t complete important transactions in a single all-or-nothing transaction unless they so happen to be on your same shard or layer 2.

BCN: You’re launching smart contracts this year along with Radix’s Babylon mainnet upgrade, what’s that going to bring to the industry and in what ways will it improve today’s defi?

PR: The purpose of the Radix public network is to radically change what is possible for users and developers in Web3. The Radix asset-oriented programming language, Scrypto, has now been tested for a year, and over 9,500 developers have used it, helping Radix make it into the best possible programming language for building Web3 Dapps.

The Radix Wallet leverages all of the power of Scrypto and the Radix technology stack to create a mobile-first user experience that is hugely easier for a mainstream audience. It’s designed to provide all the benefits of decentralization, while also maintaining the convenience of the best Web2 apps.

For example, with the Radix wallet, smart accounts enable truly decentralized account recovery which eliminates the requirement for seed phrases. The transaction manifest gives users a truly human-readable view of the transaction they are about to sign. All of this is both intuitive and also secured by the underlying Radix network.

On the developer side, Scrypto and the Radix engine execution environment provide an intuitive and secure way to build powerful Defi and Web3 applications. With native assets at the core of the Radix engine, tokens on Radix behave like “physical” objects, as you would intuitively expect them to. This means that many of the hacks and exploits we see today on Solidity and the EVM are impossible on the Radix network.

What’s critical is that both the user experience and developer experience work together to enable a radically better platform. Developers benefit from the improvement to the user experience as it means that onboarding users is far easier, and users benefit from the improvements to the developer experience as it means they can confidently use Dapps knowing that the Radix engine drastically reduces smart contract risks.

BCN: It is often said that a strong ecosystem is key to a strong network. Can you share a bit about the progress that you have made?

PR: Over the last year, the Radix programming language, Scrypto (based on Rust), and execution environment, Radix engine, have been in early access with developers. Over 9,500 developers have already tried Scrypto in that time, and already there are 50+ projects actively getting ready to deploy on the mainnet.

The Radix Olympia mainnet has now been operating for almost two years, has done more than a million transactions, and has had no stoppages or outages.

Not only has the programming language for the Radix network been shown to be incredibly effective, but the network has also already gone through a significant amount of robustness testing before smart contracts get added to the running public network.

(BCN): Radix is said to be focusing on an asset-oriented paradigm. Can you explain this and share your thoughts on why you think this is better than what’s already out there?

PR: On nearly all smart contract platforms today, such as with the EVM, developers have to create assets from scratch inside their own smart contracts (e.g. ERC20). Developers do this by creating a list of accounts and their respective balances and then defining the logic around how those balances can be updated, including validations to make sure there aren’t issues such as double accounting or re-entrance.

But if you think about it, this is madness. Practically every Defi or Web3 Dapp interacts with tokens in some form. Why are the common bits of functionality for tokens rebuilt by each developer every time they need one?

So what is an asset-oriented paradigm? It’s where the platform natively understands assets such as tokens or NFTs as they are native features of the platform. Tokens are represented as physical resources held in accounts. With this, if a developer needs a new token, they just ask the platform to create it for them, parameterizing it with things like type: fungible, supply: 1,000, or divisibility: 18. All the accounting and security are handled by the platform, not by arbitrary logic created by the developer.

More importantly, the developer’s smart contracts are no longer responsible for doing things like maintaining balances – the ledger itself does that. This removes huge numbers of checks and boilerplate code that developers today have to slog through, just to make a token interact with another smart contract. This not only massively improves security, it frees up developer time to focus almost purely on business logic.

This is not the first time we have seen such massive productivity improvements in history. In the 1990s, game developers had to build their own engine from scratch every time they built a game, defining how gravity, physics, and graphics would be rendered. Then in the late 90s, game engines were born such as Unreal Engine. Now to build a game you just ask the engine to parameterize the things you want, such as setting gravity to 1. Any game imaginable can still be built, but now developers have the tools to do the standard things they need to do every day safely, intuitively, and quickly.

That’s what the asset-oriented paradigm means for Web3 and DeFi.

BCN: Can you explain in very simple terms what atomic composability is all about?

PR: This is a perfect segue. So when a transaction is “atomic” it means that either every leg of it happens, or none of them happens. It’s all or nothing. Just like the house example above. “composability” means the ability to combine things together. So for example, lego bricks are composable with one another as they have been designed to snap together.

So atomic composability just means that you can join things together (such as the two legs of that house transaction) and you can guarantee that it all completes or it doesn’t complete.

BCN: People in the crypto and blockchain space often talk about the blockchain trilemma — or quadrilemma. Radix has said its consensus layer Cerberus will solve this. How does it work, and how will it manage unlimited scalability without breaking the so-called atomic composability?

PR: How long do we have? This is quite a deep topic but let’s revisit that mental model from earlier. On a blockchain, transactions live inside blocks. Once a block finalizes, that’s it. So what a block does is it stops you from having “atomicity” across two or more blocks.

Cerberus instead gets rid of blocks entirely. Instead of chaining blocks, Cerberus chains transactions, transaction to transaction to transaction. This means that if you ever need to interact with any part of the Radix ledger, such as for example leg 1 of the house transaction needing to interact with leg 2 of the house transaction, it doesn’t matter where that data is stored, you can combine both transactions together atomically whenever you need to. Transactions are freed from the confines of a block.

The result of this is that you can massively parallelize transaction processing across many trillions of shards (2^256 to be exact). But when you need to, you can snap anything together — with atomic composability — whenever you need it. A DEX on Radix, no matter where it is stored, will always have atomic composability with every other Dapp on the Radix ledger no matter how many transactions are being processed.

This particular insight took 7 years of research (from 2013 to 2020). With truly linear scalability without compromising atomic composability, and that’s why Radix will always have low transaction fees forever.

What are your thoughts about this interview? Let us know what you think in the comments section below.