Lessons from Fetch.ai’s first staking auction

Oct 18, 2019

On 10th October this year, Fetch.ai launched their staking program with its first staking auction. This is a preliminary implementation of Fetch.ai’s novel consensus known as Proof-of-Stake with unpermissioned Delegation (PoS-uD). In this article, I discuss the outcomes of the auction and the consequences ahead of the launch of the main network at the end of the year.

Are you pleased with how the auction went?
We’re delighted that the auction ran smoothly. Many people took part and more than $2m worth of FET tokens were staked in our smart contract. Most of the feedback we received was also very positive and suggests that the theoretical work that we put into designing the protocol is paying off.

How much FET was staked and what are the implications for the security of the network?
The contract holds just under 48 million FET (or 4.16% of the total supply). Binance is matching this figure with an identical quantity of tokens, leaving us with around 8% of the tokens staked. In the longer term, we are aiming for at least 15% of the circulating supply to be staked, so there is still some distance to go in promoting the program and encouraging participation. One positive benefit is that early adopters of the staking program will be rewarded with a return of more than 22%. We anticipate that, thanks to the direct participation of Binance, we’ll easily meet the 15% target in the next auction.

What did you learn from holding the auction?
Building and running a distributed application is quite a complicated process that involves security, cloud deployment, user-interfaces and back-end databases. Then there are the human aspects such as coordinating all of the different stakeholders in the project, marketing and communications. We’ll face similar challenges upon the launch of the mainnet, so the experience has been very useful preparation for that.

What improvements are you planning to make for the next auction?
We’re working on improving the user interface and are also planning on deploying a new contract that enables Binance and the Fetch.ai foundation to participate in the auction directly.

What are your plans for staking as you move towards mainnet?
We’re looking into tying the auction more closely to the operation of our test network with additional incentives for running validator nodes. We will release details on this program before the next auction begins.

What effect do you think Binance joining your staking program will have?
Our partnership with Binance is very important to the success of the project and we’re pleased they are on board with the staking program from the very start. It will also have quite a significant impact on the wider blockchain space, as exchanges can encourage new users and institutional investors to participate in staking programs. At the same time, this development gives exchanges a lot of control over the operation of the different networks. Binance are a very professional outfit, so I would expect them to do a good job of operating nodes. However, in the longer term it does raise some questions over the governance and decentralization of the protocols. Binance are themselves conscious of these concerns, and we’ll be working with them to make sure our protocol fulfils the goals of decentralization.

Are you concerned that with Binance controlling 100 slots in the auction and another address controlling an additional 45 slots that the protocol is too centralized?
One of the reasons for running the staking program prior to the launch of the main network is that it gives us an insight into these questions before they have any serious consequences. We’re strong believers in free entry to the protocol as any type of permissioning typically leads to even greater centralization. It’s also important to bear in mind the development of the ledger is a continuous process, and that our end-goal is to provide the perfect platform for industry and IoT applications. We are right at the start of that journey, with many of the tokens controlled by the Fetch.ai foundation and our partners at Binance. In the short-term, we’re going to work with Binance and our other partners to ensure that no single stakeholder controls more than 50% of the network. Our objective in the longer term is to keep developing the system so that it provides utility for many different applications. This will lead to ownership of the token becoming distributed across a wider number of participants and greater decentralization.

How does staking fit in with the other work you’re doing on DeFi?
Staking essentially involves putting up collateral in exchange for the right to earn profits from operating a validator node. We anticipate that extending this kind of auction mechanism to other financial applications, such as the metal exchange platform that we recently announced, can help bootstrap a decentralized financial ecosystem involving many other types of commodities. Artificial intelligence and agent-based systems approaches can support these kinds of activities by automating some of the decision-making.

The next Fetch.ai auction is due to be held in early November. We’re looking forward to sharing further updates on staking in the run-up to the start of the auction. If you have a question about the first staking auction that wasn’t answered here, let us know by posting it in the Fetch.ai Telegram channel.