Oasis is being designed to offer scalability and privacy-preservation at all layers of the protocol. Oasis Labs’ priority will be to support developers and node operators. On November 13, 2019, Anne Fauvre-Willis (strategy), EJ Jung (economics), and Reuven Gonzales (tech) joined us to answer our staking Oasis.
The way the Oasis Labs team sees things, digital service users today have no idea how their data is used, and the companies that run these services risk leaking their clients’ sensitive information. Data is siloed in order to avoid these risks, and so we lose the power that comes from data at scale.
The unique combination of blockchain and privacy technologies can help to solve these problems, according to Anne. The goal of the Oasis network is to build a better internet--one where users are in control, and innovation is not bound by locked data silos. Oasis is being designed to offer scalability and privacy-preservation at all layers of the protocol, and Oasis Labs’ priority will be to support developers and node operators.
DeFi (decentralized finance) and open finance, since app developers are reportedly looking for a way to incorporate and implement privacy and identity management. This article goes into some depth about Oasis Labs’ thinking about this, and mentions some examples, including a pilot project with a direct-to-consumer gene-sequencing company.
The team also expects to see data privacy APIs and services that leverage the Oasis network’s decentralized structure and confidential computation capabilities. What for? They’re seeing demand for apps that give users and companies more control over how their data is used, as well as the ability to run analytics on encrypted data. We can expect to learn a lot more about this soon.
However, the Oasis network will launch with a focus on staking and delegation, and on the consensus layer of the protocol. For security and stability purposes, the team’s proposal is to launch the mainnet in phases, adding new features and running a competition before each phase, with TEEs (trusted execution environments) being added in a later phase.
The Oasis network token will be used to provide economic security via staking and delegation, and it will also be used to pay for network transactions. I expect that the token will also be the primary method to incentivize secure computation when that functionality is enabled.
Some key supply information is missing here:
We also don’t know what proportions of the network’s token supply will be owned by different entities, such as investors or the foundation. Since token ownership is usually a good approximation for network ownership, I think that this information will be critical to Oasis stakeholders and potential Oasis stakeholders. The team’s plan is to share this information before mainnet launch.
Important note before beginning: all of these parameters are subject to change prior to mainnet launch, based upon observations and feedback gained from the incentivized testnet.
When you stake Oasis, you can expect your tokens to have a 14-day unbonding period. During the unbonding period (ie. unstaking), you won’t earn rewards.
Staking rewards are automatically added to your existing stake once per day (approximately), so your staking rewards will compound. However, that means staking rewards are subject to the same 14-day unbonding lock-up period, so they will not be liquid.
We should expect the rewards rate to be 15% APR initially at launch before gradually decreasing linearly to 10% APR by the end of Year 1, with 10% rewards being the lower bound.
At this time the team does not expect that staking participation rate will affect the rewards rate. Since a rewards rate is what’s being targeted, I think we should assume that at least one of these two scenarios will play out:
The supply’s inflation will likely be volatile if new tokens are minted to ensure that stakers receive a targeted rewards rate. Why this assumption? Because if eg. 20% of the supply are participating in staking at 10% APR rewards rate, then inflation only needs to be 2% to pay the rewards. However, if a participant with 10% of the supply begins staking, suddenly inflation will increase to 3% to fund these new rewards.
Block rewards will (at some rate) continue to decrease toward zero. Why? If block rewards have been allocated in advance, then at some point that allocation will be exhausted.
There’s also the possibility that some combination of the above could be enacted. However, my guess is that we’ll see Scenario 2, since EJ informed us that they have already set aside tokens for the target reward rate.
At mainnet launch, the team expects that stakers can only be slashed for equivocation (aka “double-signing”), and it’s only the minimum 100-token stake at mainnet launch. This is a very small amount and arguably does not provide much security. We should therefore expect a proposal for the broader network community after mainnet launch to increase the penalty. That penalty amount will be a fundamental piece of information for those who want to assess staking risks.
The protocol will support delegation, so stakers won’t have to run a validator. Delegators can choose to back an existing Oasis validator instead, with rewards distribution being handled automatically by the protocol. That means that validators can only impact upon delegators by 1) the commission fee they set, 2) whether or not they’re signing blocks, and 3) whether or not they get slashed for breaking the equivocation rule.
How will validators be rewarded for producing blocks and securing the network? The protocol will pay block rewards, and as the network’s transaction volume increases, transactions will yield transaction fees.
However, a node will need to sign at least 75% of blocks per epoch in order to be eligible for the staking rewards for that epoch. This is intended to incentivize uptime without penalizing validators when the network first launches. Even though there isn’t a slashing penalty for downtime, stakers delegating to offline validators will miss out on potential rewards. The team is planning to work with validators to consider additional ways to discourage downtime as the network matures.
If a validator double-signs, my understanding is that their node will be frozen permanently, the operator will need to use a new validator address to resume operations, and their backers will need to change delegations to resume earning rewards.
At this time, there won’t be a maximum stake-backing per validator, so we can expect one entity to operate one validator. The team is currently addressing Oasis governance questions, and the answer to the stake decentralization is key. The team will be testing some strategies to dissuade stake-centralization in the incentivized testnet competition, but in general, they are looking forward to more community involvement here.
The minimum requirements for validating at launch? Oasis Labs’ recommendations:
2.0 GHz x86-64
Must have AES-NI support
4 GB ECC RAM
500GB High Speed Storage
The plan is to roll out the network in phases, with validators running the consensus layer in the first phase. We’ll likely see new features in future phases, beginning with a proposal to expand the role of nodes to support other aspects of the architecture, including confidential computation and perhaps storage. It is possible that validators would need to upgrade their infrastructure to run confidential computation in later phases, but this has not yet been announced. Updates will be published here: https://docs.oasis.dev.
For mainnet we should expect minimum stake required to be 100 tokens (and there’s currently no minimum for the public testnet), but there will only be 100 active validators initially participating in the consensus committee at any given time. The active set will be selected by the protocol based upon stake-backing size.
Again, all of these initial parameters and design decisions are subject to change, particularly based upon the results and feedback that arise from The Quest, an Oasis incentivized testnet competition.
If you’re interested in participating as a validator, this is an opportunity to win token allocations necessary to participate as a mainnet validator. There will also be a pool set aside for grants awarded for other contributions to Oasis. Here’s how to sign up: www.oasis-protocol.org/quest and see more rules and specifics here: https://docs.oasis.dev/operators/staking-competition-rules.html#overview
The Quest competition, slated to launch in January, will be used to stress test the network and to test some expectations and assumptions about the decentralized network, staking dynamics, and incentives. The team stated a number of times that they ultimately value community feedback from validators and stakers.
We may see members from the Oasis Labs team running validators, since both the competition and the network is open for anyone to participate, though Anne suspects that the team will be too busy helping the network in other ways to compete as validator operators.
I’ve mentioned a few times, so far, that initial launch parameters are likely to change with proposals, which we can expect will be managed off-chain by the Oasis Foundation. Oasis will not initially support on-chain governance.
We can expect the same for upgrades after mainnet launch--the foundation will use a form of off-chain governance to gather input and feedback from Oasis community members before proposing a change. I assume that updated software would then be released to validator operators, who can then choose whether or not to run the new software.
We don’t know how decisions will be made within the Oasis Foundation, or if/how stakeholders may become foundation members. We also don’t know if/how the foundation will delegate its reserve tokens.
As I mentioned, the mainnet (for security and stability purposes) should launch in phases, with new features and a competition prior to each phase. The initial mainnet is expected to focus on staking and delegation, while TEEs (trusted execution environments) will be added in a later phase, and are currently available for DApp developers to use on the Oasis DevNet.
When mainnet? We don't know. The team has some tests to run, including a validator and stress test first, so after the incentivized testnet goals are fulfilled, the mainnet will be ready to launch.
Oasis’ consensus is currently based upon Tendermint, so there’s inter-blockchain communication (IBC) potential. What’s that mean? Oasis could be made interoperable with the Cosmos ecosystem while remaining sovereign. Anne noted that they have a grants program, and one of the tools that they'd love to get someone to work with them on is IBC integration. Anyone interested should check out: www.oasis-protocol.org/grants
Foundation website: www.oasis-protocol.org
Oasis Labs website: www.oasislabs.com
Staking competition, The Quest: https://docs.oasis.dev/operators/staking-competition-rules.html#overview
Slack channel: www.oasis-protocol.org/slack
Join the #nodeoperators channel to discuss as a community, share feedback and thoughts, and work together to troubleshoot.
Initial documentation is here: https://docs.oasis.dev/operators/overview.html
Research papers: www.oasis-protocol.org/researchpapers
Special thanks to Anne Fauvre-Willis for spending an hour with Staking Hub to answer our many questions.
Thanks to our Staking Hub community for the thoughtful and impactful questions that inspired high-quality answers. Since you’ve read this far, you might as well join us over in the Staking Hub Telegram channel.
Hopefully you found this useful. Feedback is always welcome! I’m on Twitter.