Why Aragon Chain

Ethereum has been critical for Aragon's success so far. Without Ethereum this community wouldn't be where it is today. Aragon was warmly welcomed by the Ethereum community, allowing us to raise funds to continue building Aragon and the Aragon Network. We never forget our origins and will always be forever grateful to the Ethereum community.

We've also given back to the Ethereum community. We build and maintain infrastructure like radspec and aragonOS, which are common goods for the entire community. We started the Aragon Nest grants program, which has funded important work such as Prysmatic Labs (when they were the only team working on 'sharding') or Frame (the best desktop signing provider for Ethereum).

However, those of us working for Aragon have a duty to deliver maximum value to the Aragon community and expand the Aragon Manifesto in the world. While Ethereum was the best and only platform to build Aragon, we now see alternatives that are better suited for our needs and reaching our goals.

Even though the cost of making transactions on Ethereum has increased since we started, the real problem is the inability to predict future cost. 'How much does it cost to use Aragon' is a question that keeps on popping up when talking to users that want to use Aragon to run their organizations, and we aren't able to provide an good answer. Ethereum is almost at capacity at this point in time, if just one dApp manages to get traction and considerable growth, gas prices could increase by two or three orders of magnitude almost overnight. Using Aragon on Ethereum is already way too expensive.

Regardless of what you may have heard on Crypto Twitter, Ethereum is, first and foremost, a developer platform, and one that is still at a fairly early stage. Big changes are still needed (Ethereum 2.0 will change everything) and these changes can and do break things. As it is now well known, the Istanbul hard fork will break close to 5,000 Aragon smart contracts. While the situation has been mitigated and a migration path will be made available, I deeply believe it is not our fault but the implications of bad protocol design. We are not Ethereum's only users and we understand the importance of the changes even if it breaks our stuff. This has just confirmed our belief that relying exclusively on Ethereum poses a considerable platform risk that we should hedge against by having Aragon available in multiple chains. It's important to have a chain governed by the Aragon community as it will only have our best interests at heart.

My personal prediction is that Ethereum 2.0 won't reach feature parity with the current Ethereum mainnet until 2021 or 2022 (informed by conversations with some Ethereum 2.0 implementers). As a project, we must find product-market fit before 2022, and I'm absolutely certain that we will. When we reach PMF, we need a scalable platform to onboard as many users into Aragon as fast as we can.

Aragon is not leaving Ethereum. Incredibly important projects like Aragon Court and the Aragon Network v1 will still be deployed to mainnet Ethereum by the end of the year. I don't foresee Aragon dropping Ethereum support and we will likely make the needed changes for Aragon to work on Eth 2.0.

However, we must take control of our own platform from the bottom up. Owning the whole stack will be incredibly powerful, allowing us to capture value wherever in the stack it accrues (which is still to be seen in Web3) to reach project sustainability. I believe the research and plan are solid, and we can move into implementation now.

Polkadot at AraCon

Back in January, I mentioned in my AraCon talk that there was ongoing research on Aragon Chain. It had become clear that it was something that we needed to do to ensure Aragon could scale to serve the amount and type of users that need the product.

At that time, we were thinking that it made sense to develop an application-specific blockchain, building Aragon natively at the core of the chain, and only later add a general purpose VM to allow extensibility. With that direction in mind, building Aragon Chain using Substrate and relying on the Polkadot network for security was the most promising path forward.

When we had conversations with people working on Polkadot, we expressed our doubts on whether to use WebAssembly or EVM as the VM for smart contracts in Aragon Chain. We were recommended to use Wasm (which I too believe is a better VM than EVM by almost every metric), but we were told that building an EVM module for Substrate "could be done by a decent engineer in two weeks".

We had to put our research on Aragon Chain on hold during the first half of the year as we needed to direct all of Aragon One's R&D resources to building Aragon Court and launching the Aragon Network. During summer, as research work on the Aragon Network was mostly completed and it was just a matter of finishing the implementation, we started working on Aragon Chain again.

But during those 6 months, a lot has happened in terms of Aragon development. Significant work was done on two major app suites (Aragon Fundraising and Open Enterprise), and more than 15 useful Aragon apps were under development and almost ready to be released. The way in which these apps increase Aragon's utility and how fast they were getting built has definitely exceeded my expectations.

When coming back to Aragon Chain research again, it became really clear that we had to approach the problem from the opposite side: start with a chain with a general purpose VM which was compatible with everything that had been built to that point, and later on work on chain level optimizations to improve the performance of using Aragon.

With this in mind, an EVM-compatible chain became the only possible path forward and the only aspect that we couldn't compromise on. After receiving non-positive answers regarding EVM support in Substrate (incompatible with our target launch timeline) and discovering that Ethermint (with an EVM module for Cosmos SDK) was going to be finished before the end of the year, we started reconsidering whether Substrate was the best technology to build Aragon Chain.

Cosmos SDK and Ethermint

Back in August, we became aware that Ethermint and the EVM module for Cosmos SDK were being worked on by ChainSafe. The Ethermint project seemed stalled for a long time and they had now contracted with them to get it ready for production. After meeting with ChainSafe we learned that they were almost done and their target for completion was early Q4 2019.

After a productive meeting with the Cosmos team, Aragon One engaged ChainSafe to evaluate what the best technology for building Aragon Chain was. We structured the initial engagement around a feasibility study to compare both Polkadot and Cosmos in terms of how we could build a chain that satisfied Aragon's needs.

Since we started reviewing the first drafts of the report, we felt that it aligned with our own research that we were doing internally while the report was being worked on. Cosmos SDK with the EVM module that ChainSafe was working on for Ethermint was the best way to go for us to build Aragon Chain.

The report was made public yesterday as part of ChainSafe's AGP and can be read here. I encourage reading the full document in order to understand not only in what ways the two platforms are different, but also how some of Aragon Chain's features can be built.

Given the work that has already been done on both Cosmos SDK and Ethermint, launching Aragon Chain would require considerably less work than any alternatives. By having a chain with EVM support, aragonOS and all existing Aragon apps can be deployed to Aragon Chain without any work. The Tendermint proof of stake algorithm is the most battle-tested PoS consensus currently available, having secured over a billion dollars in value.

On launch, Aragon Chain, with just support for vanilla EVM will already have considerably more throughput than Ethereum mainnet (targeting between 2-10x) and it will probably be the only real proof of stake EVM chain available (unless someone else launches a chain faster).

Aragon Chain will deliver the promised proof of stake Ethereum, at least a year before Ethereum 2.0 is usable.

On top of that, Cosmos SDK's modularity will allow us to continue improving Aragon Chain beyond just being a vanilla EVM chain. Even though non-Aragon contracts and protocols will be able to use Aragon Chain, optimizations at the native level can be done to optimize for Aragon's needs, making it the best possible platform for running Aragon organizations.

In terms of ecosystem maturity, the Cosmos Hub, which is built on Cosmos SDK and uses the same consensus that Aragon Chain will, has been running without significant issues for more than 6 months and it's market cap is above $500m. Aragon Chain could eventually become a zone (a chain in the network in Cosmos' lingo) in Cosmos Hub when IBC is launched (targeted for 2020) and interoperate with other zones and even with Ethereum (via a peg zone), instead of using a direct bridge to Ethereum mainnet (as we will do for genesis).

ARA: Aragon Chain's staking token

One of the most interesting aspects of Polkadot is its shared security. Parachains don't need to worry about the security of their own chain (in terms of consensus) and every chain is as secure as the relay chain and all other parachains.

Polkadot parachains need to bid for a slot within the limited number of slots and lock up DOTs, effectively paying for security by the supply being inflated and not being able to earn inflation rewards.

Without that, Aragon Chain will need to take care of its own security. A new token is needed to act as the staking token that Aragon Chain validators will stake to secure the chain and come to consensus on the state of the chain.

We are tentatively calling it the ARA token and its distribution mechanism should be very similar to Aragon Court's ANJ. ARA can be minted and burned by depositing or withdrawing ANT in an Aragon Fundraising bonding curve.

Aragon Chain validators will earn transaction fees from activity on the chain. Given that most activity on Aragon Chain will likely be Aragon activity, through ARA, ANT holders will have a way to directly benefit from Aragon's open source software usage. Aragon Chain usage will lead to validators earning more fees, which will lead to ARA becoming more valuable, which will create an incentive to lock up more ANT to issue more ARA.

Although still not fully specified, ARA validators and delegators could also earn ARA as a reward for staking and protecting the network.

One DAO, multiple chains

Even though Aragon Chain should be a superior platform for running most Aragon organizations, there will still be reasons for Aragon organizations to prefer using Ethereum mainnet, security and interoperability with other Ethereum protocols being the most important.

In terms of security, given that it is a proof of stake chain, Aragon Chain's security will be a function of its market cap. An increased use of Aragon Chain should lead to a higher market cap for the chain, so one can expect that it will become more secure as time goes on and many organizations start running on Aragon Chain instead of Ethereum. Organizations holding a significant amount of assets will probably be better off on Ethereum mainnet until the security of the chain is sufficient. The long tail of relatively capitalized organizations ($10k-100k in assets) should be safe to run on Aragon Chain.

Interoperability wise, one of the features that should be included for the initial version is a bridge to mainnet Ethereum. This bridge should allow to transfer tokens from one chain to another. The bridge should unlock use-cases such as using Dai as a fee token to pay for fees on Aragon Chain or trading tokens of an Aragon Chain organization on Uniswap.

Apart from value transfers, the bridge should be able to pass arbitrary messages between both chains. Arbitrary message passing will unlock exciting setups like a DAO running all its heavy governance process on Aragon Chain (which could be prohibitively expensive on Ethereum), but having an Agent on Ethereum that performs actions if and only if a message is sent by the organization on Aragon Chain via the bridge. This will allow organizations running for very little cost on Aragon Chain to interact with any protocol on Ethereum.

Making Aragon Chain happen

Aragon One worked with ChainSafe to come up with a proposal for them to build Aragon Chain as soon as possible. ChainSafe is now wrapping up work on Ethermint, which should be finished in the next few weeks. Because they are the ones building Ethermint, they are the best positioned team in the world to work on Aragon Chain. We planned the next steps for Aragon Chain during Devcon, coming to the conclusion that the development of Aragon Chain and the bridge to Ethereum mainnet can be worked on in parallel and also can be started right now without having to wait for Ethermint to be completed.

Fostering a strong validator community is critical to Aragon Chain's success. This is uncharted waters for the Aragon community and we should start working on it now. I imagine someone or some team from the community stepping up and leading the creation of this validator community.

Infrastructure wise, thanks to Ethermint implementing most of Ethereum's JSON RPC methods, adapting existing Ethereum infrastructure (wallets, block explorers, exchange intergrations) to work with Aragon Chain shouldn't be too hard. We are incredibly well positioned having Frame as part of the Aragon community. Other infrastructure will need to be adapted and I expect the community to step up and prepare the infrastructure ahead of the launch.

As mentioned before, Aragon apps shouldn't be impacted by whether they are running on Ethereum or Aragon Chain. Yet some changes might be needed on the Aragon client for it to work with Aragon Chain. As owners of the Aragon client, Aragon One would take on the Aragon Chain support tasks for the client.

The next step to make Aragon Chain happen is for the community to approve AGP-106 in the upcoming Aragon Network Vote. I'm biased as I have been pushing for it, but I personally vouch for AGP-106 and the ChainSafe team to build and launch Aragon Chain. The rough timeline that ChainSafe has given us is so exciting that I prefer not sharing it right now, but Aragon Chain could be live sooner than most would think.

It's exciting to see an ambitious idea entertained a few months ago crystallize from just an idea to early research, then serious research, and now implementation.

It's going to be ARA-mazing.


This post was a collaboration between

Jorge Izquierdo, Aragon One

  • Jorge Izquierdo
  • Aragon One