Brave new world(s): Cosmos, blockchain and social evolution

Blockchain is a social technology. A close look at social evolution is also a close look at blockchain.

Cosmos is a system for connecting different blockchains and allowing interoperability between them. But if you're feeling grandiose, you could say the emergence of the Cosmos blockchain is analogous to the human race achieving globalisation though modern technology.

Sunny Aggarwal, research scientist at Tendermint and Cosmos and host of the Epicenter podcast, was feeling grandiose.

"I see it as this third - I know this is going to sound so cliche - third generation blockchain thing," he said.

"Here's how I see the generations. The first - I put it analogously to human development - first generation is kingdoms and villages. They couldn't really scale that big, there wasn't a lot of economic activity. Maybe they'd trade with direct neighbours. That's Bitcoin, Litecoin, all those independent things with no way for them to connect. Bitcoin was for payments, Namecoin was for DNS, Sia was for storage. There was no connectivity. You couldn't use your Bitoin to buy Namecoin names, you couldn't use your Namecoins DNS name to point to some file stored on Sia."

A growing population and advancing technology would then help humans consolidate into their next big step, Aggarwal explained.

"Then what we realised; large scale economic connectivity meets economic growth. For thousands of years humans have been empires. Ethereum came along and built the first blockchain empire. Here's one massive chain you can do everything on. If everyone's under the same control it can kind of force you to trade. That's what empires did. Ethereum does the kind of same thing... I can sell my CryptoKitty on 0x for a Maker DAI."

"Then you also get large scalability issues with empires. Technical scalability, but also social scalability. Like, you look at all the governance discussions happening with Ethereum... I think it's because it's too big - too many competing interests. There's no historically democratic empire, right? You just can't create a democratic empire like that. There's too many competing interests. Everyone who calls themselves a third generation blockchain, EOS, Tezos and all that, for the most part seem to be just trying to build a better empire."

"Everyone is going down this one chain to rule them all method... [But at] Cosmos, we're trying to create a world of nation states and city states," Aggarwal said.

Also watch

Money makes the world go around

If you follow the analogy down the line, the next generation of blockchains would naturally be a system of independent yet cooperative entities. To do otherwise would be to ignore some of the greatest lessons humanity has learned (the hard way) in preceding centuries.

"The greatest innovation of humanity in the last 100 years is we achieved mass scale economic integration without the political," Aggarwal explained.

It's hard to argue with that.

International trade, the lifeblood of the modern world, keeps pumping no matter what and despite all its diversity, the world has still managed to economically connect on a global scale. You can't drink the water in Afghanistan, but you can visit the Coca Cola bottling plant in Kabul which opened in 2006 even as the Taliban was gaining territory across the country. You'll find computers and smartphones in every country, and there's not a single country in the world where both the latest mass produced consumer technologies and boutique handmade goods from exotic regions can't be bought for the right price.

These are some of the results of "economic integration without the political", and it's profoundly shaping the world.

But this level of connectivity is no accident. It was a deliberate and gradual process, comprised of a great many inventions which we can now take for granted.

"A bunch of technologies enabled this in the last 100 years. Like free trade zones is one, where free trade became the default," Aggarwal noted. "Containerisation - what I mean by that is you go to a shipping port... all the containers are the same size and everything, right? Basically the entire world is standardised on the same container format, and this allows any port in the world to ship to any other port in the world. Institutions like the UN helped as well, where it's not a government but a discussion forum, and that's useful."

The Internet is another one of these connecting technologies, Aggarwal said, and it very much looks like blockchain is too. Global connectivity through the power of trade and economics, rather than politics, is a constant, ongoing process. The Cosmos platform is positioned as another one of those technologies.

"That’s why I got interested in [blockchain]," Aggarwal said. "We can do economic integrations without the political. That's what Cosmos is trying to do to the blockchain ecosystem. Let's try to get away from these massive one chain to rule them all systems. Let's create a system where there're specialised blockchains like in generation one, but they have the connectivity of generation two."

Blockchain interoperability: The missing link

Although no one's entirely sure what the blockchain world of the future will look like, that perception of a single glorious empire has always been one of the most popular. It also has an intuitive charm to it through the assumption that the best product will win out and get enough scale to drag everything else underneath its umbrella.

But it's probably not the best way to account for the messiness of the real world.

For example, the usefulness of an interbank blockchain settlement network depends entirely on the circumstances and is constantly changing.

Two banks might want to keep a lucrative correspondent banking relationship, a third wants a technological leg-up over larger competitors but a fourth can't spare the resources right now. A fifth bank in the USA wants blockchain settlement for corridors to Europe but not Asia (minus Japan) while keeping SWIFT for domestic transfers, even as its counterpart in Europe wants the exact opposite.

One size does not fit all, so even if it's objectively better than the incumbent system you're still going to have a tough time selling a one-size-fits-all empire-style blockchain to that industry.

Real blockchain solutions need to be flexible, and they need to be able to fit the contours of the real world. In this way, having a multitude of interoperable systems is much more sensible than trying to create one chain to rule them all.

But where does that leave Cosmos? How does it simultaneously be a public blockchain and the glue that holds multiple networks together, without also falling into the same trap of trying to impose a one-size-fits-all solution?

The answer is by designing its network to be just as flexible, modular and real-world-ready as individual blockchain applications need to be.

Building blocks

Blockchain design is bi-directional, Aggarwal noted. A blockchain changes the world, and the world changes the blockchain.

"At the end of the day, it's very much a social technology. It's very much about social coordination," Aggarwal said. "You can't separate [the technical] from the political side of the technology. Cosmos was looking at the current social structures of the world and saying 'how can we use that design of the current system to help?'"

"Trying to reaffirm the cycle of the technology of blockchain was trying to help solve the social issues of social coordination. But you also look at social coordination, and see how we can use it to inspire the blockchain ecosystem."

This mindful design is evident in Cosmos' technical vertebrae, which aims to allow the creation of modular, interconnected ecosystems, or "hubs". These hubs are highly customisable in their own right. Cosmos itself is one such hub.

The key elements behind this are:

• Inter-Blockchain Connectivity (IBC) Protocol: A set of standards for blockchains to talk to each other.
• Tendermint proof of stake: The Consensus mechanism that's used by the Cosmos hub.
• Cosmos Software Development Kit (SDK): A modular framework for developers to easily build blockchains and applications on Cosmos or in line with its interoperability standards, including hubs.

The point of these is to just get them out there and make them as widely available as possible for anyone to use them as they want.

"What we've really been focused on right now is building free and open source tools like Tendermint Core, and the Cosmos SDK and IBC, and all we're really trying to do is put the tooling out there to shape the blockchain ecosystem to how we think it should work," Aggarwal explained.

• Testing Cosmos' proof of stake

  Of course, there's a difference between how something should work and how it ends up working, and to a certain extent all you can do is make sure something holds up on paper and then try to test it under as close to real world conditions as possible.

  In the case of Cosmos' Tendermint proof of stake consensus mechanism, one of these tests was the Game of Stakes which ran prior to its mainnet launch. Game of Stakes was essentially a proof of stake game where nodes were encouraged to lie, cheat and steal their way to victory at any cost – within reason, with real money in the form of Cosmos' ATOM cryptocurrency on the line.

  It definitely succeeded in capturing some of that real world human unpredictability.

  "One of the validators, Bitfish, they Sybil attacked the registration," Aggarwal recalled. "So to start off in Game of Stakes you had to submit a KYC process - just for Game of Stakes, not mainnet - they just bought hundreds of passports."

  "Not going to lie, I did that too. I thought I was clever because I got three of my close friends to sign up, and I'm like 'hah, I have four identities here.' But Bitfish shows up with hundreds, so that was a kind of interesting experience there. What happened was the rest of the validators coordinated a hard fork to kick Bitfish out of the game. I was very impressed they managed to do that."

  Beyond that, Game of Stakes served as a valuable experience for validators to learn how to harden themselves against attacks. It also seeded some ATOM among proven users so they could start staking and securing the Cosmos mainnet from day one. Plus, it was good for straight forward bug testing.

  And lastly, it was a way of testing the actual economics of proof of stake itself.

  "That's the one I kind of disagree whether it did it well or not," Aggarwal said, "because there was this idea that 'hey, let's test proof of stake at an accelerated rate' so the inflation in this hub was like ten thousand percent or something. We're mimicking four years in two months. The problem there really is the game theory of a short term game is different from the game theory of a long term game. And Game of Stakes was a short term game. You knew the end date and you knew the winning conditions... in a real proof of stake model, who knows how long the game lasts? There're no strict winning conditions, so the game theory completely changes."

• Building the IBC Protocol

  "We had IBC... we had that finished almost a year ago, a year and a half ago," Aggarwal recalled. "It was working with the Tendermint consensus and the Cosmos SDK, and two chains of those types talking to each other. But that's no fun. It's very limited to certain things that are both using our tooling, which kind of defeated the whole point."

  "So now what we've done is we've started a working group called the IBC working group. So a project called Loom, Moracle, Prysmatic Labs which is one of the Eth 2.0 implementers... I know Ripple has said they were interested in participating. The idea is we're open to anyone participating in this thing, and the goal is to make it so its like abstracted as such that like it works with any BFT consensus protocol."

  "We've charted a repo called the ICS repo. It stands for Interchains Standards. And so the idea there is we're trying to spec out this process, when two chains get to know each other for the first time, they've started up that initial connection, they need to do some sort of handshake where they kind of explain to each other 'hey this is my consensus protocol, this is all the technical stuff you need to know about me in order to communicate with me. This is the hash format I use, this is the merkel tree I use.' Things like this."

  "That's still a heavily ongoing process right now."

  With ICS still under construction, the Cosmos hub didn't launch with IBC enabled.

  "It just didn't make sense to launch with IBC enabled," Aggarwal said. "Honestly, proof of stake plus Tendermint is enough of an experiment. Let's let that run for a bit."

  Binance Chain is also being built with Cosmos SDK tools, Aggarwal said, and there are discussions around it being part of the Cosmos hub once IBC is implemented.

  In the long run, maybe it plans to create its own interoperable hub with the tools provided by Cosmos, he speculated.

Come for the code, stay for the governance

"Cosmos is designed with the intention of people creating multiple hubs," Aggarwal emphasised. "So you know we already have two hubs in the system. There's the Cosmos hub then another chain called Iris, which is more focused on the Chinese ecosystem, which is kind of the intention of how the Cosmos ecosystem is intended to work."

This also provides an excellent example of how Cosmos' modular design is adapting to real world conditions while still facilitating connectivity.

Cryptocurrency and blockchain regulations in China are very different to most other countries, even going as far as to actively prohibit supposedly essential elements like immutability. This naturally poses a problem for an empire-style blockchain trying to roll out a global one-size-fits-all solution. Those regulatory differences are an example of one of those messy real world issues that needs to be resolved somehow.

For clarity around what actually makes Cosmos different, it's worth looking at EOS for contrast. EOS is an empire-style blockchain that aims to be malleable enough to suit the real world. Its flexibility comes from being democratic, in that both its consensus and governance mechanisms are determined by vote. The idea is that its community can collectively modify the blockchain as needed, to keep it evolving as necessary and finding a globally acceptable sweet spot between mutability and immutability.

But as an empire, it's still trying to present a one-size-fits-all solution, which results in some serious problems. Its mutability means it could probably be manipulated by nation-level actors which makes it unsuitable for certain applications, its community of voters is largely divided by language and cultural barriers and it's subject to the same vote-buying problem as democracies everywhere, which gives lopsided influence to the most profit-oriented blocs of its community.

A jack of all trades is a master of none.

Cosmos' goal isn't to replace EOS or Ethereum as a new empire. Rather, it's to present a way for people to use EOS, Ethereum or anything else for specific applications when they think it's the best one for their needs. They can keep their own consensus mechanisms and full functionality as hubs in the wider Cosmos ecosystem.

If another blockchain should decide that it doesn't want to be part of the wider Cosmos ecosystem, that's fine. You can just hack it in anyway, Aggarwal explained. A network that actively wants to be part of the Cosmos ecosystem can implement IBC in itself, or IBC can go to them.

"We call these things peg zones in the cosmos design, if something doesn't natively implement IBC," he said. "For example Bitcoin. I don't see it ever in the near term future natively implementing IBC. But the idea is that Bitcoin is a very valuable chain to the Cosmos ecosystem, so we kind of force IBC to be implemented in it using this concept called peg zones."

The full circle

Allowing for all these hubs, layers, layers within layers and so on, solves a lot of problems. But one of the best parts might be that this also brings modularity to governance, which starts solving one of society's most intractable problems; that people just don't care about politics.

"In the Cosmos hub, we at one point had this system where we said if people don't vote their money gets slashed," Aggarwal noted. "And what ended up happening was people just wrote scripts to auto vote abstain. You can't force people to care about governance, which I think is something a lot of people think you can do. Forcing them to do so doesn't work."

"So how I think you inspire people to participate in governance, you split things into communities they care about. Traditionally people seem to be much more involved in things that are like more relevant to them, right? The problem with something like Ethereum is there are so many competing things that don't really affect most people... too much stuff can filter through... I think if they feel a governance system is too big for them to have a meaningful say in, then they won't have the desire to participate in it."

What does this look like in the Cosmos ecosystem? Whatever you want. That's the point.

"If you have more specialised things, let's say you're the user of a certain application and you want to be more interested in voting on things that only have to do with that application, that's what that will help," Aggarwal explained.

"Ethan Buchman, who's head of the Interchain Foundation, he's very into local currencies and stuff. How can we create blockchains that are more for localised cities and whatnot? I think if you break things down into smaller communities you can get people to be interested more in participating in the governance of it. I think anyone who tries to create a single global system, you're going to inevitably hit this roadblock we've just talked about. I think what you really want is some local coordination, and then you have coordination tools for local systems to coordinate, and then maybe different layers."

"My favourite country in the world is Switzerland, because if you look at how their entire country is laid out, almost all power is delegated to local townships, then the next level of power is to the cantons, then the federal government has relatively weak power - it's only used for coordinating... when things can't be done by specific cantons and you need larger coordination, then it does that. But by default, power is delegated down. That seems to be one of the most stable democracies in the world."

Just like the real world, it ends up being layers within layers within layers to let people engage in the governance on the levels that affect them personally. But in a world taken over by an empire blockchain, the biggest and furthest outside layer would never be ideal. Even the most perfect governance system would be too vast and distant for most people to participate in.

Cosmos once again takes cues from the real world here. It's envisioned as the furthest outside layer, and there you don't actually find a governance mechanism per se. You instead find a discussion forum, and the tools to facilitate it.

In the real world this outermost layer is arguably the United Nations. And while it's often criticised for being a toothless tiger, that's also kind of the point. It's a discussion forum, not a dictator.

"When you come to higher level coordination of different communities, that's where you have some sort of higher level blockchain. That's where things at Cosmos hub come in," Aggarwal explained. "It's a spot for smaller blockchains to come coordinate on higher level things that need coordination amongst them."

"It's like the UN, right? The UN isn't a government. It's a discussion forum for people to try to coordinate to the best of their abilities. Hopefully we can try to help create one for the blockchain ecosystem as well. We kind of mapped what we see as how political systems form in the real world, and at the end of the day I think blockchains are just about human coordination and trying to map the natural formation of human systems to a blockchain system."

Maybe blockchain technology really is evolving from kingdoms to empires to a modern world?

Disclosure: The author holds BTC, BNB, ATOM at the time of writing.

Picture: Shutterstock