API3: AMA
I hosted an AMA for API3 team on t.me/panamacrypto on January 12th 2021 at 5PM UTC. Burak Benligiray, Saša Milić participated from API3; and myself as moderator. Below are the questions and answers from the resulting AMA:
1. We all heard about Chainlink and its success, but most of us don’t know the basics, so let’s start with an ELI5, more so, it would be nice to receive explanations in a “noob friendly” way, rather than using buzz words and tech explanations which most people won’t understand. The more I or We, can understand a product the more i can like it, use it, recommend it etc… so:
- What are oracles?
Saša:
An oracle is a piece of software that brings off-chain data onto the blockchain. It’s composed of an off-chain component (the node), and an on-chain component (the protocol implemented as smart contracts).
- Why do they matter/ why are they useful?
Saša:
The data that natively exists on the blockchain is ownership data; i.e. who owns what token on the blockchain? This is why you see cryptocurrency as the first and still the major application of the blockchain. Expanding the kinds of blockchain applications possible (beyond cryptocurrency) necessitates having additional types of data on-chain (e.g. price data for more complex financial instruments, flight data for flight insurance, etc.)
- Where can oracles be used? basically who are the customers paying for oracles?
Saša:
Pretty much any blockchain application that isn’t strictly exchanging tokens requires off-chain data. For example: algorithmic stablecoins require knowing off-chain price data (since the price of a token isn’t actually known on-chain, it’s decided by off-chain markets). Most, if not all, major DeFi projects need off-chain price information. These DeFi projects are currently the biggest customers of blockchain oracles, although we think that will change as the number and diversity of blockchain projects grow.
2. Can you give us some concrete examples of how ecosystem functions?
Burak:
An oracle ecosystem is composed of data providers and data consumers. Here, data providers are traditional businesses monetizing their data through a technical interface called an API. Data consumers are smart contracts and decentralized applications running on blockchains. For example, a centralized exchange API can provide ETH/USD price to a stablecoin contract. These data providers and consumers are mapped many-to-many, which means the stablecoin will want data from multiple APIs, and the APIs will want to serve other projects as well.
One thing of note here is that third-party oracle-based solutions also introduce a middlemen class in this ecosystem called “oracle node operators”. These intermediaries buy data from the data providers, and resell it to the data consumers.
3. Why is a decentralized solution such as API3 better than a legacy one?
Saša:
Short answer: a fully decentralized solution means no single points of failure. This includes engineering failures as well as governance failures (from a central entity governing the protocol).
4. API3 is relatively new on the market and competition seems to be growing, what do you bring different to the table? It is a known fact that, for example BAND copied lots of Chainlink code, to the extent that it is being called a copycat.
Saša:
There are multiple notable differences, but I’ll center on arguably the two main ones: a first-party oracle centric solution and decentralized governance.
Our solution has API providers themselves operate (first-party) oracles and transmit their data on-chain. First-party oracles are operated by the API providers themselves. Note: third-party oracles are not operated by the owners of the data they serve, acting as middlemen between the data source and the blockchain. Compared to third-party oracle solutions (e.g. Chainlink and Band), this is an improvement by: (1) reducing attack surface area by eliminating third-party nodes situated between the data source and the blockchain, (2) improving Sybil attack resistance (since, by default, you know who is operating each oracle), (3) reducing costs (since you don’t have multiple third-parties querying data from the same source, which is the case in existing solutions).
The decentralization governance (DAO) aspect of our solution adds another improvement: (4) transparency. In particular, by making data feed construction explicit and by making the operations of the DAO (the governing entity of the protocol) open to the public.
I recommend reading our articles:
5. Which competitor do you like and why and which competitor do you resemble with (more or less) tech wise?
Burak:
The Coinbase oracle (https://blog.coinbase.com/introducing-the-coinbase-price-oracle-6d1ee22c7068) is an interesting experiment that provides similar security guarantees to first-party oracles, so it can be said that it resembles the API3 solution. However, the method used suffers from a lack of data source variety, which prevents it from providing security in practice…
In addition, I find the UMA oracle staking solution rather elegant, though it’s designed around third-party oracles.
Saša:
It’s encouraging that many people recognize the critical importance of connecting off-chain data with the blockchain :) I agree with Burak and also, I’ll add that I find the NEST Protocol pretty neat and novel, although they only do price data.
6. Could you give us background on your team members; does anyone have previous experience with oracles? How many people are on the team and what did they do previously?
Burak:
A large part of the founding members were previously working on the Honeycomb API Marketplace.
Honeycomb was essentially an API-centric oracle marketplace, but the underlying oracle solution wasn’t built to enable first-party oracles, and was limiting us. Seeing that none of the solutions on the market satisfied our needs, we decided to build our own oracle solution to meet this need.
It was affirming that we were not alone in this view. The security and the data source-transparency that first-party oracles provided is seen as an obvious solution to a lot of the problems that existing projects suffer from. For example, Sasa and many other founding members reached similar conclusions independently, and got involved with the project simply because it complemented their personal vision.
The DAO has 20+ employees and multiple external teams working for it at the moment and we’re currently hiring.
7. API3 is advertised as cross chain, how far is this from being accomplished? At this moment API3 is just an ERC20 token if I am not mistaken.
Burak:
API3 is primarily the governance token of the API3 DAO, which resides on Ethereum mainnet. So the API3 token will stay as an ERC20 token, and does not need to be ported to other chains for us to serve there because it’s not a payment token.
There are two prerequisites for calling a data source decentralized: (1) The data is aggregated from multiple sources trustlessly. (2) The governance of the structure that achieves this is decentralized. So the main difficulty here is achieving decentralized governance across chains, but this is completely overlooked by other projects that are governed centrally even on Ethereum mainnet. Therefore, our cross-chain plans are beyond deploying nodes on other chains or porting the token.
8. Tell us about the nodes? What is their purpose in the system and who can become a node? and how?
Burak:
The API3 solution depends on first-party oracles. This means that there is no middleman on the data path, and the data source also operates the oracle node. Then, you would need to be an API provider to operate a node that will serve the API3 DAO. We reach out to API providers for this, but we are also receiving inbound demand from a variety of API provider types.
9. what is the Decentralized safety net? this sounds like an interesting feature, so please do an ELI5 on it
Saša:
I think you mean the insurance component of the project? Data feeds can be optionally insured, which means you have a lower bound on how much money you can trust the data feeds to secure. Insurance claims are settled in a decentralized manner via Kleros, a blockchain dispute resolution protocol. I should note insurance is a novel feature; I am not aware of any other blockchain data feeds that are insured.
10. What is the token utility in the ecosystem? Does the token have any other purpose beside governance?
Burak:
Staking the token at the DAO provides voting power and staking rewards. The staked tokens are used as collateral for the insurance feature Sasa mentioned just now. Therefore, the token secures the data that API3 provides, but it does this at the governance-level rather than the oracle-level. For more information about our tokenomics you can refer to this post: https://medium.com/api3/api3-tokenomics-update-f032d6e49b30
11. Is there anything else about the product that we didn’t ask and that you would like us to know?
Burak:
People are extremely DeFi-oriented when they think about oracles. Our thesis is that oracles are for integrating APIs to smart contracts. Because of this, we are working on a much more encompassing protocol, node and business model in general. In essence, we are working on a solution to the API connectivity problem, rather than solely building data feeds for DeFi.
12. How much funds did you raised thus far? On how many rounds? and how are the tokens distributed among investors and future users/nodes?
Burak:
We closed our seed funding round at $3M USDC (for 10% of supply, vested over 2 years), and our public distribution with an additional $23M USDC (for 20% of supply, in circulation at the moment). We have an ecosystem fund composed of 25% of the token supply. For more details you can refer to:
Note that since this post, we increased the vesting period for the founding members from 2 years to 3 years.
13. Why do you believe that formulating API3 as a DAO will have greater chance of success? At this point in time governance is more or less a gimmick and far from being a winning formula.
Saša:
I don’t believe DAOs are gimmicks; I think they are the future of institutional organization. I recommend reading my article here:
To quote from the article: “The creation of a DAO is meant to assist a blockchain protocol in dealing with the unknown. That is, if a protocol could be perfectly engineered from its inception, then a DAO — created to oversee changes to the protocol — would be entirely unnecessary.”
That is, if you are creating a complex protocol, then it will undoubtedly require modifications and changes in the future. DAOs are essential in making sure that changes to the protocol are implemented in a decentralized way.
There are other also more practical things a DAO can enable: organizational scaling, organizational transparency, regulatory arbitrage, and removing central (governance) points of failure. These are all discussed in that article I linked above.
14. What are the revenue streams for API3 the company?
Burak:
The DAO will monetize access to the data it curates, and will also be providing security in the form of insurance that the users will be paying premiums for.
15. Could you tell us more about your tokenomics? Are they inflationary or deflationary, fixed finite supply or flexible issuance? What are the incentives for token investors and holders and what is the value accrual mechanics of the API3 token?
Burak:
The token has an inflationary and a deflationary mechanic. Staking at the DAO yields inflationary staking rewards. The DAO burns its revenue, causing deflation. This model is adapted from EIP-1559 and fixes significant incentive issues that revenue-generating DAOs face. See https://medium.com/api3/api3-tokenomics-update-f032d6e49b30 for more details.
Since the entire inflation is paid out to the stakers, token holders don’t need to be concerned about it as long as they stake. However, staking comes with the risk of being slashed when insurance is paid out. This is reconciled with a floating inflation rate. The DAO sets a stake pool size target, and inflation rate is regulated automatically to ensure that it’s met.
16. Is there any chance of API3 token being categorized as a security at a later point in time?
Saša:
There is no way to guarantee a regulator’s categorization or interpretation of existing guidelines, across jurisdictions, but the API3 founders took numerous mitigating measures for purposes of securities regulations, like: outsourcing the token distribution and specifically geo-barring Americans, directing all proceeds from the distribution to the multisig DAO wallet, strictly prohibiting any token price speculation on API3 comms, zero hierarchy and broad decentralization/distribution in tokenholders governing the protocol.
17. What stage is the project or product? Do you have MVP and are you planning to offer more services in the future?
Burak:
The API connectivity problem is an ecosystem problem much more so than a technical problem. Therefore, I don’t consider our technical solutions to be significant milestones for assessing progress. For example, our main goal is to become the no-brainer solution if you’re an API provider who wants to monetize their data on blockchain, or if you’re an on-chain data consumer in need of data from APIs. Releasing a particular technical product does not provide that.
To give a more concrete answer, we have a feature-complete oracle node, and we’re working on getting it production ready. We also want to extend it with a generalized pub-sub protocol, which no other oracle node supports right now. In addition, multiple teams under the roof the API3 DAO are working on different aspects of the solution simultaneously, such as the integration platform, insurance, governance, etc.
18. What are your main milestones over the next 6–12 months?
Burak:
Our priority at this stage is to launch the authoritative DAO, which will implement the staking functionality I have mentioned before. In the meantime, we’re simultaneously working on our oracle solution and integrations. We have always been an extremely agile team and don’t really believe in keeping your head down to build something for an entire year. Experimenting and using the gathered data to correct course is extremely important when working on bleeding edge tech. Specifically for our case, we are working on enabling use cases that don’t exist at the moment simply because the off-chain data is not there. This requires us to work with users that demand this unavailable kind of data in a very tightly-coupled kind of way, and be flexible to take opportunities.
19. Do you have real world customers already and how do you see their transition to DAO? How did you gauge demand for this type of service/product?
Burak:
It’s very clear to me that we’re at the brink of DAOs going mainstream simply because of the increasing numbers and sizes of new DAOs being found. Unless the governance of a project and its funds is not decentralized, the project is not really decentralized, and the idealists are uniting under a variety of DAOs against this. It’s unthinkable for a project to decentralize its own governance, and then depend on a centralized solution for a critical service such as oracles. Therefore, I think that there will be a landslide demand for actually decentralized oracle solutions, and API3 is by far in the best position to provide that.
20. How do you plan to encourage people to use your solution? What are your marketing plans?
Burak:
If a project needs off-chain data, it doesn’t really have any option to not use it. It will either receive it from an oracle solution or the project will simply not exist. There are two opportunities for us here. By providing data that was practically not available before, we will make a lot of use-cases possible, and essentially be the only option for those. In addition, we will also provide existing kinds of services in a fully decentralized way, and with the security and cost-efficiency of the first-party oracles.
21. What are the biggest challenges you expect to face and how do you plan to overcome those challenges?
Burak:
Oracle projects are a lot more operations-heavy than other blockchain projects. Specifically, there is a lot of business development and technical integration work that need to be done all the time. Doing these under a DAO structure is definitely not something trivial. However, the same DAO structure also presents a significant opportunity. If you think of the alternative, which is a centralized company, it can’t feasibly scale up to serve all blockchain projects across all chains. Despite not being as efficient as a centralized company, a decentrally governed project has the potential to scale up limitlessly, and cover the entire blockchain ecosystem like a fungi network. This is not something that was achieved before, but it’s very exciting to think about.
22. Do you have any interesting partnerships that you would like to tell us more about?
Burak:
As I’ve mentioned before, I’m most excited about strategic partnerships with other DAOs. That is a lot more powerful than projects partnering up because the cross-pollination between the members of the DAOs align their incentives in a way that is very difficult to detach again. So I’d recommend you to be tuned in for these relationships, as I expect these to be increasingly built as more decentrally governed projects pop up.
23. Everyone knows that a big part of $LINK success (price wise) is because Chainlink managed to build a strong community of die-hard supporters. Do you guys have any plans to do the same? It can sound nice to be a part of “API3 Cavalry” for instance.
Saša:
I don’t think you can “plan” something like a die-hard community. (E.g. The “link marine” community arose organically, and I think primarily because it was the first oracle project and people really grabbed onto the potentiality of solving the “oracle problem”.) Planning for such a thing (and not letting it develop organically) usually comes across as corny. As soon as crypto marketing is approached in a corporate way, it really loses its effect, from my experience.
I think the most important thing is to be solving an important problem and doing so in a good way. With that said, in addition, having likable faces associated with the project has a large effect as well. The core team is busy with development right now, but we’ll be putting more effort into community engagement and more visible and vocal presence in the “cryptosphere” in the coming weeks/months.
Community questions:
24. Hi it was really interesting AMA, I wanted to ask what is strongest advantage that you guys think will make your team leading the market compared to other projects?
Burak:
We believe that we have modeled the problem a lot more accurately compared to other projects and as a result were able to come up with a much more suitable solution with first-party oracles and decentralized governance.
25. API3 — It’s a DAO built with the intention of providing data feeds that are governed in a decentralized manner. Can you tell us about the degree of decentralization and governance model of this project? What incentives will be there for participating in the Governance activities?
Saša:
The token has an inflationary and a deflationary mechanic. Staking at the DAO yields inflationary staking rewards. The DAO burns its revenue, causing deflation. This model is adapted from EIP-1559 and fixes significant incentive issues that revenue-generating DAOs face. See below for more details.
Since the entire inflation is paid out to the stakers, token holders don’t need to be concerned about it as long as they stake. However, staking comes with the risk of being slashed when insurance is paid out. This is reconciled with a floating inflation rate. The DAO sets a stake pool size target, and inflation rate is regulated automatically to ensure that it’s met.
26. Does the API3 team have any business development / sales team in place (or planned) to gain customers, partnerships, etc? Seems like the market is just beginning for real world data in smart contracts so need to be aggressive at gaining adoption for api3’s solution. Any conferences or presentations upcoming? Weekly newsletter with progress updates and future events / milestones to focus on?
Burak:
We currently have a BD team already experienced due to our previous work and we’re extending it to better fit API3’s needs. At the moment we’re a bit limited in terms of traveling so we are mostly posting articles and doing AMAs such as this one, but we will be attending a lot of conferences when things got better.
27. “The collateral utility has the participants share API3’s operational risk and incentivizes them to minimize it. Finally, the governance utility gives the participants the ultimate instrument to enact these incentives.” What would be a example of this both from a governance perspective and from the 1st party perspective. If there’s bad data from 1st party? Would be a vote to have them removed if not fixed or other ideas?
Saša:
If there’s bad data supplied to a contract, then that contract can create an insurance claim if they had opted in for insurance. And, yes, it’s possible that the DAO votes to remove that data provider, but it would depend on the type and severity of the error.
28. How can API3 providers get setup with an airnode? I get this is a one-off thing, but I assume that any given API provider will likely set themselves up with their own airnode via your (future) tooling?
Burak:
We have guides that will allow providers to get started with operating their own nodes even by themselves:
29. Does that mean if there’s a specified period for staking but something goes wrong, does that mean the staking program would be stopped?
Burak:
Essentially the model is structured in a way that stakers receive both the rewards of well-operated data feeds and the risks of badly-operated ones. Since stakers also hold the voting power, they are responsible to vote in a way that will result in secure data feeds. The whitepaper explains this in a more detailed way.
But the API3 team will also be helping with the integration and deployment of the nodes.
If anyone would like to follow up further on API3:
Website: https://api3.org/
Telegram: https://t.me/API3DAO
Twitter: https://twitter.com/API3DAO
Looking forward to see you guys active on the API3!
If you enjoy the read, follow me on Twitter, and use my referral links:
PrimeXBT, Binance, Bitmax, Bitmex, Bybit, Kucoin, FTX
Is good for Karma!
This is not a financial advise and you should do your own research before investing, and understand if you are willing to lose this investment, as there is always risk in innovative markets as technology, fintech, and cryptocurrencies, due to regulations, geopolitical issues, and what not. That being said, besides being invested in API3, I am being compensated for my time writing this article.
