MakerDAO's Annual Update, Reality Check, Next Chapter
It's been about one year since Rune's nascent ideas about stablecoin technology coalesced into what we know as MakerDAO. That's one year of frontline instruction in the principles of blockchain security, decentralized governance, open-source project management, and financial innovation. I'd like to take a step back at this point and give the wider community an overview of what we've achieved during the last year and where we are headed for the next one. If you saw my presentation at Devcon 2, a lot of this will be familiar with the addition of plenty of helpful links to the different corners of our ecosystem.
The first lesson we had to abruptly learn was just how new the Ethereum platform really is. Like the heady days of the early World Wide Web, the fantastic potential of this technology far outweighs implemented solutions to even the most basic problems of software development.
If you wish to make an apple pie from scratch, you must first invent the universe.
Carl Sagan, Cosmos
The original dev team saw this lack of sufficient tooling as a dependent task blocking Maker's important work of creating a robust stablecoin (which is itself a dependent task hindering most interesting economic activity in the network). Rather than waiting around for someone else to solve it, they got to work creating a development environment where it could be possible to create the high quality product that was envisioned for Dai. To that end, the Maker community subsidized the development of two open-source projects that make up the foundation of our ecosystem and benefit the Ethereum platform as a whole.
The first is dapple, a developer multi-tool that allows (among many other things) adequate code reuse properties that are otherwise not found in Solidity. Once dapple was functional, we had the ability to create an arbitrarily deep tree of dependencies in our projects. This enabled us to create the second component, dappsys. Dappsys is a highly composable framework of smart contracts that take a Unix-style approach to functionality by producing a lot of small easily understood building blocks that are then assembled into a much more complex dapp. These two technologies have unlocked more productive potential than anything else in Maker, and both were undoubtedly a wise investment of resources at the start of the project.
Once the foundation was laid with these two components, we were able to start building out the core of the Maker ecosystem. First, dappsys building blocks were composed into an upgradeable MKR token. The token and the entire framwork were then audited last February by Piper Merriam, and all of Maker's deployed core code was again audited last August by Peter Vessenes.
Next, we followed by releasing an OTC exchange dapp called Maker Market, where MKR and other ERC20 tokens can be trustlessly traded against ETH on the blockchain. Maker Market doesn't charge fees beyond gas usage, making it another instance of something we built out of necessity that additionally serves as a beneficial public resource for the whole Ethereum community. The smart contracts can be found here and the frontend website code can be found here. Maker Market is still in its early stages, so you can expect many usability improvements here in the coming months.
During the summer, we stepped up our research efforts and published two papers on the intended design for some important components of the ecosystem. The first was a description of The Sentinel Stake-voting System. This will be the root authority that grants the MKR holders the ability to manage all modifiable aspects of the stablecoin engine. It is intended to be one of many governance solutions that will empower the MKR holders to carry out their important responsibilities as regulators of the system. The second was an in-depth paper describing the mechanics of how control theory can be applied at the heart of the engine to enhance Dai stability over time. This paper walks through different simulations of Dai supply & demand dynamics to illustrate how the incentives in the system should react to ensure stability. In addition, we published a revised whitepaper that explains Maker's core value proposition more clearly and with updated mechanics.
Next, we got to work creating another core element of the Maker ecosystem: the Continuous Splitting Auction contracts. This is a system that performs an auction of a parcel of ERC20 tokens, but a bidder can choose to bid on a portion of the parcel and thus create a sub-auction. This functionality is recursive, which makes it optimized for autonomous price discovery. It will serve as the foundation of both the collateral liquidation system and the Buy&Burn system within the stablecoin engine. We are currently testing the contracts at this website, where anyone can participate in MKR auctions. The dapp's code can be found here.
Throughout this entire year, the Maker community and dev team have been remarkably realistic about the scale and difficulty of the project before them. Dai is a compelling idea and there is much to do, both on the blockchain and off, to ensure that it will work properly when it is released. An entire cast of actors needs unique tools to follow their proper incentives, and it would be dangerous and irresponsible to declare Dai ready for launch before verifying the system extremely thoroughly. As a community, we have always prioritized security and caution over developer velocity and cheap marketing, and I think that this is the primary strength upon which our reputation is built. What this means however is that Dai is not Coming Soon™ in the overzealous promotional sense. It will be ready when the whole thing is ready and when that time comes it will be built like a castle.
On the other hand, I will be the first to admit that the Maker community has gotten complacent in our "avoid unnecessary hype" philosophy and external communications have suffered as a result. We have grown a bit too insular in our chat room and this may have caused prospective newcomers and casual observers to become discouraged when trying to determine the status of the project or stay current with development updates. This lapse in messaging has been noted and you can expect it to be corrected going forward. We will be providing updates at a more regular cadence in a variety of media formats that will make it much easier to keep up to date on our progress, and we will improve the onboarding experience so that more interested people can get involved in the project. We learn a lot at MakerDAO every day and we want to ensure that the Ethereum community can benefit from our hard work as much as possible.
Regardless of the release date of Dai, Ethereum's genuine demand for a functioning stablecoin solution grows by the day. In light of this, we are working on a transition technology that will bridge the gap while Dai is being implemented and refined. This project is called Simple Stablecoin. It is an independent platform that allows anyone to deploy a centrally administered token controller and offer their own unique stablecoin solution. The administrator will be able to pick their own accepted forms of collateral, their own price feed providers, whitelists for issuers and holders, and their own custom rules such as fees or unique token characteristics. This will allow a competitive market to form, with the end result being a stablecoin that works well for Ethereum's current, relatively simple, needs. For example, a project raising money in a crowdsale will be able to hedge against ETH volatility and a dapp that requires a generic "store of value" can use a simple stablecoin during its beta program. Obviously, the more ambitious use cases for Ethereum that envision massively scaled systems will have to rely on something much more trustless like Dai to work properly, but that world isn't here yet and simple stablecoins should do fine for the immediate future. We are already in talks with a few different groups that have ideas about the tokens they want to offer, so you can expect to see developments on this topic soon. We are currently running the controller factory on the Morden testnet here and you can check out the contracts here.
In closing, Maker has learned much while trailblazing through the blockchain wilderness, but there is a ways to go until we reach the promised land. We have approached the problem of stability from engineering first principles, we are gaining momentum in pursuit of our goal, and we are proud to share our solutions with the wider world. We will increase their accessibility going forward. Look out for a new welcome wiki in the coming weeks. If you are interested in learning more about the project or getting involved, join our chat room and say hello.