Harmony Open Staking is around the corner! As the community participates in staking and running validators, it means that Harmony has really entered the permissionless state. From now on, the blocks of this chain are built by the community.

1. EPoS, The Optimization of PoS

The different consensus mechanisms will lead networks into different directions. In earlier days of blockchain technology, decentralization, scalability ,and security is a famous trilemma you can not achieve at the same time. PoW and PoS head to different directions of this triangle, and Harmony chose PoS to be the consensus, which has emphasized more on scalability. And in order to avoid the centralization problem, Harmony came up with the EPoS model.

Ethereum is experiencing a transition from PoW to PoS, its founder Vitalik Buterin understanding to PoS can describe the mechanism precisely, he wrote in Casper the Friendly Finality Gadget: In a PoS system, a blockchain appends and agrees on new blocks through a process where anyone who holds coins inside of the system can participate, and the influence an agent has is proportional to the number of coins (or “stake”) it holds. This is a vastly more efficient alternative to proof of work (PoW) “mining” and enables blockchains to operate without mining’s high hardware and electricity costs. He pointed out the efficiency of PoS.

POS was born on July 11, 2011. A user on the Bitcointalk forum named “QuantumMechanic” first proposed the Proof of Stake in an article “Proof of stake instead of proof of work”. Many solutions he mentioned in this one are still popular, such as “voting power would accumulate with trusted delegates instead of miners”, “new bitcoins and transaction fees could be randomly and periodically distributed to delegates”, etc.

On August 19, 2012, Peercoin, the first blockchain project based on the PoS was born. From the proposal to launch on production, PoS has experienced about 1 year of development. After that, an increasing number of people has reached consensus on PoS, and some POW networks have also begun to shift to POS, Ethereum is one of them.

In the meanwhile, some problems of PoS have been found in practicing, like because insufficient insurance or price drops, community hasn’t been incentivized, also on decentralization PoS is far less perfect than PoW’s consensus. Thus, lots of projects try to optimize PoS, and proposed LPoS, BPoS, Ouroboros, ect. DPoS is an important one among them, it brought PoS into a more pragmatic stage.

DPoS, Delegate Proof of Stake, is proposed by Daniel Larimer in April 3rd 2013, to voting validators instead of the PoS which the randomness of election might be manipulated in PoS:“why don’t let stake holders elect validators who sign new blocks? Holders vote to signers by CDD(coin-day-destroyed)”.

Kamie, a KoL from China and founder of staking service provider Wetez said, “DPoS is respecting the basic rule of professionalism, it is like a representative system or deputy to people’s congress in country governance”, and he also pointed out that centralization is inevitable in this system when holders delegate huge amounts of token to single nodes.

Harmony began building the public network since 2018, and has learned lessons from previous projects. During the past 2 years, Harmony team has made lots of effort to figure out the best solution to achieve decentralization, scalability and security, which is to break the blockchain trilemma. So Harmony proposed its own economic model EPoS(Effective Proof-of-Stake):

  • Electing top stakers: Most of the current PoS blockchains have a concept of committee which is a group of validators who have the right to produce and validate blocks. Usually, the committee has a limited number of seats, like 21 for EOS and 100 for Cosmos. The Harmony committee adopted the same mechanism. As a PoS sharding blockchain, Harmony has 4 shards at launching phase and each shard contains 250 seats. 80 seats among the 250 are available to the outside community, which means 320 seats open the opportunities to the external community. Validators will be elected based on the highest amount staked. Thus, the top 320 stakers will obtain the 320 seats (4 shards * 80 seats) and become the validators across the shards. We believe using staked tokens as criteria is the most efficient and economically secure way for validator election, as the stakes closely tie the validators incentives with the well-being of the blockchain itself.
  • Encouraging Participating: Decentralization is Harmony’s north star goal since the first day of designing the network. We have been seeking to build a network everyone can participate in. Harmony allows delegation. Thus, people who hold large amounts of ONEs can build their own validators, and meanwhile, people who hold small amounts of tokens can participate in open staking by voting to choose validators who can produce blocks.
  • Punishments preventing centralization: To avoid the right centralization which occurred in DPoS, the EPoS Mechanism has been introduced to the network. Simply put, the higher ranked validators will be actually economically punished from staking too much in a single validator. Validators will be rewarded less after they surpass the effective median stake amount, and the ratio of return will decrease.

2. The Beginning of Permissionless Blockchain

As a permission blockchain, Libra revealed its powerful committee in 2019, which is composed by dozens of branded companies. Harmony is different from Libra in that we are a permissionless blockchain that started from community, and we have been building a decentralized network which is accessible to every community member.

Harmony’s delegation mechanism, also the economical punishment and compensation rules from EPoS allow retail investors, wales, funds, exchanges and wallets those who hold different amounts of token to join the network together. At the same time, Harmony still can make sure it is both a decentralized and fair or sufficient incentivized network.

The open staking launch is the start of permissionless network, multiple kinds of community members’ incentives will band together with harmony’s growth:

  • Retail investors:Harmony has been listed by Binance as a IEO project at June first 2019. We have gained a huge amount of token holders from Binance. Although there have been fluctuations in past months, Harmony’s flexible staking mechanism offers a long term chance for investors to make profit.
  • Foundational nodes: we were seeking to have an organic and globally distributed network of node operators to ensure that our network is decentralized from the start. So, in January 2019, we launched our foundational nodes program. People who joined this program committed for a 22months lock period but obtained our token at a fair price at that time. At the end, we have more than 70 individuals and institutions joined us, and influential funds like Hashkey are among them. Because part of their token is still in locking, those foundational nodes will join the staking program again to gain extra reward. In a network’s starting phase, there are frequent updates and resets, it is a huge challenge to Harmony’s technology and operation, when those nodes are globally distributed and run by different level partners and individuals. We removed lots of frictions in the past year, either on technology or operation, and we are ready for the launching mainnet.
  • Exchanges: Exchanges are the inherent nodes of PoS consensus. Our token holders can join staking conveniently via this platform, also if exchanges can manage well their deposit reserves, they can join the staking to earn extra reward. After Binance, Huobi, Kucoin, Bitmax also listed ONE token, and they are the partners of Harmony open staking now.
  • Institutional investors: Several Crypto funds participated in Harmony’s early fundraising, who care about stable and long term return. When acknowledging our open staking is approaching, they joined the staking testnet, showed the wills to join staking and continue holding our token.
  • Wallets:Some users are accustomed to holding their token by wallets rather than actively trade. It is our duty to coordinate wallets providing staking service within their product. Based on that, we partnered with Trust Wallet, Math Wallet. And we are glad for more wallets to join our ecosystem.
  • Staking service providers: People who are not willing to deal with technical details can delegate their token to staking service providers, who will operate the nodes to them. It is the same as people who hold a small amount of ONE tokens can vote to those nodes run by service providers. So far, we partnered with more than 10 global providers, such as Hashkey, Infpool, Stakefish, Wetze, Everstake, SesameSeed and etc. our users can choose service from any of them.

In Harmony’s road map, Harmony will enter the adoption phase, all the partners and nodes we have connected in staking will help us with mass adoption.

3. A model balances fairness and incentives is ready to run

The token economic model on a chain is similar to a social welfare and taxation system. The welfare and taxation system guides the social wealth’s creating and distributing, as well as the token economic model guide the token holders’ way to node running and token usage, by block reward and slash rules. At the beginning of mainnet launching, it is key to balance decentralization and sufficient incentives to ensure the network will run smoothly.

Harmony’s staking is the first practice of EPoS economic model from concept becoming a reality.

Fairness

The combination of PoS and delegation, enables everyone who holds our token to participate. There is obvious fairness in our access control, which is different from ETH’s requires a minimum of 32 ETH, and the super node system of EOS and Tron.

But there is a huge gap between individuals and institutions in token holding amounts. Also we elect the top ranked stakers and the compounding design which re-staking the reward in the next epoch, both of them obviously lead to the dilemma of “the rich get richer”.

With the introduction of effective stake, we can bypass that dilemma. The higher ranked validators will be actually economically punished to stake too much in a single validator and the lower-ranked validators are enjoying extra reward for their stake. To prevent a validator with a large amount of stakes to overtake a single shard (a.k.a. 1% attack). In the meanwhile, encourage small stakeholders to join the network operation and governance.

At this stage, harmony runs 4 shards and each shard opens 80 seats to the external community. Let’s say every shard owns 100% voting power and the portion distributing to 80 seats is by their effective stake.

We use median_stake to denote the amount of stake at median in the ranked list of top 320 stakers, and actual_stake is the actual stake held by the validator.\

Here, c is a protocol parameter (for example, c = 0.15). The effective stake of a validator is basically its actual stake bounded by the upper limit of (1 + c) * median_stake and the lower limit of (1 — c) * median_stake.

Besides the block reward, the voting power of each validator in the consensus is also determined proportionally by the validator’s effective stake:

For compounding, the validators in the yellow area are economically incentivized to spin up new validator machines to “compound” their rewards. In terms of staking pool, this design is forcing the staking pools to decentralize themselves and avoid single point of failure. The nodes in the blue and green area can directly compound by re-staking their reward in the same validator.

Sufficient incentives

Sufficient incentives are important for launching. BTC who is based on PoW was easy to produce blocks in the early stage. And the decentralized world outlined by BTC has convinced a large number of people to be the miners.

PoW mining by computing power, the remaining amount of token is the incentive. In contrast, PoS’s incentive comes from new insurance. It is key for PoS networks to distribute a sufficient amount of rewards to miners and equally important to figure out the way of sustainable development. PoS networks usually adopt an insurance ratio with linear change, which gives out enough reward at the beginning to attract validators and gradually decrease insurance ration to 1%.

ETH’s year insurance ratio is 3%,Tezos and EOS is aiming at 5%,Cosmos sets the range between 7%~20%, overall, most of the networks insurance ratio at the beginning is around 5%~10%。

We focus on sustainability in the token model we proposed last year, so aimed at a very low insurance ratio:

Linear Adjustment of Issuance based on Staking Percentage (old model)
  • We believe a target stake ratio in the range of 35% is optimal, and the ratio of return will dynamic change with the stake ratio: Per shard block reward is 18 ONE tokens. For every 1% increase (or decrease) in stake, the block reward decreases (or increases) by 0.4 ONE tokens. Although issuance shifts linearly, the fact that it shifts inversely to supply staked allows the yield to change exponentially. This creates strong downside defense from having too low of a stake as yields become very attractive at lower levels. Based on that assumption, we set our insurance ration around 3%~1%。
  • In this model, yield = (block reward * multiplier *blocks per year * 4/320) / (new circulating*%supply staked/320)
  • But in the survey we did in 2020 among our partners, we found out that the 35% stake ratio is a very low target for our community. When the ratio reached 70%, running a validator is unprofitable. Rewarding period ends too early is negative to participating.
  • And from the formulation above, it is obvious that yield is decided by 2 factors: one is the changing reward which is decided by stake ratio, another is the change of our token supply, when the supply increases the yield will decrease, as our token is still in the process of unlocking. Those factors will easily bring the yield down.

Thus, in March 2020, we modified the model:

Yield curves for the first 3 years of rewards. (new model)
  • Yield= annual ONE issuance/percentage of supply staked*circulating supply.
  • Under the new model, block reward is constant annual 441M ONE regardless of changes in underlying variables such as block time and staking ratio. We increased the reward per block from 18 to 28 to attract validators, at the same time, kept the insurance at 3.5%. Balanced both incentives and sustainability.
  • Since the annual model has fixed rewards, the yield is only dependent on the percentage of supply staked and circulating supply. This means yield has fewer variables than the previous model. The new model shows when 70% of the 12.6B total token supply is staked the yield will be 5%, meaning the community can engage in staking profitably.

We believe this model takes care of both decentralization and incentivization, initiates the permissionless network ,and will maintain the network going forward.