Cardano SL versus Cardano CL (whiteboard)

Hi, this is Charles Hoskinson broadcasting live from warm, sunny Colorado. Today is January 28th, 2023, and I'm making a video to talk a little bit about the architecture of Cardano because some questions have come up. It's important that people understand who we are, why we did what we did, and the legacy behind things. So, I'm making a little bit of a whiteboard video. Let's share my screen.

The story begins back in 2016-2017 when I wrote the Cardano essay, "Why Cardano," basically outlining the thought process behind the design of Cardano. I introduced two concepts: Cardano's settlement layer (SL) and Cardano's computation layer (CL), which is closely related to our side chains concept. The idea is that the settlement layer is safe, deterministic, reliable, self-healing, very decentralized, and upgradable. There are many properties here that you care about. We introduced concepts the extended UTXO accounting model, Plutus, and Cardano's native asset standard.

These provide sufficient expressiveness to capture most of the DeFi, NFT, and custom user application space. For most things a person would want to do, Cardano's SL is either in the process of building that infrastructure or has enough built-ins to accommodate it. However, there will be circumstances where people want to do things that our design choices as an ecosystem are incompatible with. You can shed the need for determinism and gain benefits, but you reduce safety, reliability, and potentially your ability to reason about how the network behaves. You can remove some decentralization for more speed, sacrifice upgradability for bleeding-edge capabilities, and have different failure modes that give you different operating costs.

We knew this back in 2017; it’s not a surprise to anybody. So, we introduced the IDF Cardano computation layer. I know this because I wrote the entire thing myself and explained various different things we were thinking about. The idea of Cardano CL, which has become our side chain model, is that you would have different computation layers with different ledger network transaction consensus logic and design principles, all connected to Cardano. There are reasons to do this.

One is that you have different trade-offs. In our industry, some people believe there’s a "God protocol," a Holy Grail where someone has figured out perfection in every aspect. The reality is that this doesn’t exist. You have different trade-offs related to decentralization, reliability, safety, upgradability, and throughput, and you make decisions based on those trade-offs. Some trade-offs that are good for a stable, secure, long-term network may not be good for your business logic.

You can either say, as they do in Bitcoin land, "Get lost." That’s what the Maxis do; they say there is only one truth, and if you deviate from it, you’re the problem, not the product. They suggest using a centralized layer two to solve the problem and insist they will never change the core protocol. Another option is to accommodate all these things. Polkadot does this through parachains, and Cardano does it through side chains.

We wrote a blog post about it, published on July 5th, 2022, introducing the Cardano EVM sidechain. We discuss the side chain model and the kinds of things we’re doing with side chains. This is not an original idea; it’s something the industry has recognized as very important, so much so that people create blockchain toolkits. For example, we created one with Alex Chirpinoy called Scorex2, which is part of the Hyperledger Foundation's working group. Scorex2 is a modular blockchain framework in the Scala language that allows limitless experimentation with various designs.

It’s a complete rewrite of the original Scorex framework, focusing on compact functional code, modularity, asynchronous networking, and a nice JSON API. They pull out the network, transaction, and consensus layers and allow you to plug them in. Ergo, for example, has launched using Scorex, and several other projects have as well. IBM created the Hyperledger Fabric framework, which is used to build things like Marisk and Walmart. The most popular one right now appears to be Cosmos, which allows you to write a custom blockchain and iterate rapidly.

These different frameworks have trade-offs, viewpoints, and languages. I believe Cosmos is written in Go, and I can’t recall what Cosmos is written in. In any event, having a framework allows you to parameterize those chains to produce something. Did we build Cardano as a generic modular plugable Haskell framework? No, we built it as a safe, deterministic, reliable system with a huge emphasis on its ability to upgrade itself, preserve decentralization, heal itself, and enable an economy to be built on top of a financial operating system.

We wrote it in Haskell not for developer accessibility, but because we cared about these things. It would be a catastrophically stupid idea to take what we have and turn it into the next fabric, Cosmos, or Scorex. You can try, but spending 18 to 24 months refactoring and rewriting the code doesn’t make Cardano itself any safer or more reliable, nor does it make it faster. It just makes it more accessible for people to create side chains with that code base. The vast majority of developers are unlikely to opt for a Haskell side chain.

Some will, but most probably will not, especially when there are already existing options with years of track records, large development ecosystems, beautiful SDKs, and no token requirements. These SDKs do not require a token, meaning they are blank templates that a person can use and customize. For example, we have World Mobile, which has deployed some stuff on Cardano because they want the safety, determinism, reliability, self-healing, decentralization, and upgradability that we care about. However, telecommunications networks involve governments, trusted hardware, spectrum, licensing, high privacy requirements, and massive throughput requirements. Cardano can do many things well for them, but when you start looking at all these different needs, you realize that some custom infrastructure is required.

They could fork Cardano and build a side chain with Haskell, completely rewriting the ledger logic and violating design requirements, or they can use an open-source toolkit to deploy a side chain with the necessary trade-offs. Isn’t third-generation cryptocurrency about interoperability? Yes, it is. Wouldn’t it make sense for them to use Cosmos, as it’s a good toolkit, attach it to Cardano, and create a computation layer with a bridge for value to flow back and forth? Cardano can provide the quorum for that side chain.

These toolkits don’t come with built-in governance or security; they provide programmable infrastructure. They could do the same with Fabric or Scorex, and we encourage people to do that. This builds an ecosystem of these things—hopefully hundreds of them. Why is this good for Cardano? If you look at the staking reward curve over time, there’s a continuous decline.

It takes over 100 years for it to approach zero asymptotically, but after a decade, most rewards decline considerably. The area under the curve in the first decade is significantly larger than in later years. What augments stake pool revenue and makes the system sustainable beyond transaction volume is that when you have a computation layer, these also pay staking rewards to stake pool operators and delegators. The security of the SL, powered by Ouroboros, extends to these layers. They don’t come from another network or token; they come from Cardano.

These chains are in partnership with Cardano, so they have to pay, and they pay in tokens. Over time, while the ADA rewards decline, stake pool operators receive ADA plus rewards from all those other side chains in the system. Isn’t it in our best interest to have a diverse ecosystem of different technology custom parameterized to the needs of each? They keep all their traffic on the side chain while benefiting from the security, reliability, liquidity, and infrastructure of the main system. Their concerns are parameterized.

We recognized that business model five years ago; it’s written down in "Why Cardano." You can look it up in the Wayback Machine. It’s nothing new; we’ve been discussing this consistently since before Cardano launched. We built a side chain—two of them, actually—one in Scala, with the Mamba side chain as a case study, and we also built a generic framework with Alex Chirpinoy called Scorex. We’re a member of the Hyperledger Foundation and regularly communicate with the Fabric team.

Now we’re getting deep into Cosmos because there’s a desire to attach that to Cardano. What’s extraordinary to me is that some people online say this is evidence that Cardano is a failure because there are dApps and ecosystems that require doing things off-chain or in a different ecosystem. The future is multi-chain and multi-infrastructure. You cannot escape that. Why?

Because there are trade-offs in design, and the people who will win the future are the interoperable ones, those who support everything and run infrastructure with many different things. The more of this we have, the easier it becomes, and eventually, you wake up with 200 of these things, which increases the network value and utility of the system. That’s the point; that’s the machine being constructed here. It allows people to move from trustless to trusted infrastructure and back, especially when you look at traditional finance, which we call TradFi. They’re regulated actors, meaning they’re centralized.

Until we can find a way to decentralize regulation, that’s the reality. They have their own infrastructure requirements and can’t move everything to a blockchain. What are we going to do? Be like Bitcoin and tell them to go away? Or do we build a system where they can operate with their own logic and benefit from our system in partnership?

Telecommunications is somewhat centralized due to nation-states and monopolies, including government-run corporations that own and operate the spectrum. We’d love to live in a world that’s entirely libertarian and decentralized, but that’s the reality of telecommunications today. This means you need bespoke infrastructure to accommodate that reality, even if your intention is to fully decentralize. You can’t just go to a government and say, "Change your laws." If the goal is to connect the unconnected, you have to build bridges between the traditional and non-traditional worlds.

If someone wants to deal with traffic and do security tokens or traditional finance, they are regulated actors. What are you going to do? Just tell them to change all their laws to accommodate us? That’s the Bitcoin way. The compromised solution is to allow people to build side chains with bespoke logic that corresponds to the needs of those regulated domains.

You then have decentralized infrastructure in partnership that preserves people’s rights, like self-sovereign identity and privacy, while accommodating other requirements. That’s what a partnership is about; that’s what an adult ecosystem is about. It’s about real trade-offs. A lack of understanding leads to misinformation. It never ceases to amaze me how much disinformation flows through.

I figured we’d make a video to talk about this because it’s nothing new. There’s six years of discussion and history, and we’ve been saying the same thing repeatedly. Introducing these concepts again and again, with projects like Midnight and World Mobile, I see comments saying that because someone’s building a side chain, it means Cardano is a failed project. No, it just means there are different design choices. Instead of fundamentally changing Cardano to accommodate the needs of an individual actor, which would lead to centralization, we build extensibility so people can have their own domains to succeed within.

Then we create an economic model where everyone benefits collectively from that. It’s that simple. That’s how you build things properly, and that’s the Cardano way. I hope in the future that people mature in this industry. It’s hard every day when you say the same thing, show all your work, be completely honest, publish your results for third-party review, and have open-source code.

You go to great lengths to communicate, and then the opposite is perceived by mainstream news. You just have to show up every day and keep pushing through. It doesn’t hurt us, but it does hurt the industry as a whole. It discourages professionals and those with good intentions from getting involved. They wonder why they would want to engage in something where they do good work and get criticized.

Telling them to grow a thicker skin isn’t the answer. It’s not about the skin of any individual; it’s about whether we want this to be the technology of the future. If we do, we shouldn’t set up landmines that burn people down. If we want to succeed, we shouldn’t lie. When we do those things, people say, "I don’t want to be involved in that," and they leave or give power to those currently involved, who may not be good actors.

We’re at a crossroads. I was excited about the announcement of Cosmos and World Mobile because we’re bringing two great ecosystems together. These ecosystems will be mutually beneficial. They’ve spent years building great technology in a 200-plus blockchain ecosystem. It’s open infrastructure, which means we can absorb it.

That’s the Darwinian and interoperable nature of cryptocurrencies—they make us better, and we make them better. This is the first side chain that will have an awesome security model because of Ouroboros. I got excited about that, but to see the reception of, "This just means Cardano is irreparably broken," is a mindset issue. Maybe it’s generational, with a pervasive cynicism and dishonesty. I hope this elucidates more about what we’ve been doing and trying to accomplish.

I really believe this is the way. We said there were three pillars to third-generation scalability: interoperability and sustainability/governance. The second is interoperability. How do you achieve interoperability unless you build bridges and side chains? Do you just put everyone else out of business and force your standards on them?

I guess that’s the Bitcoin way. I’m not in the mood to embrace that. We’ll keep building, and we’re excited to bring this technology into Cardano. We can’t wait to see the wonderful things that come from it. Overall, I think we’ve remained remarkably consistent in our vision and approach.

That’s auditable; you can go back and look at documents written six years ago and compare them to where we are today. The only difference is that we were saying we would do it, and now you’re seeing us do it. Welcome to crypto.