Overview of Cosmos

Cosmos, being developed by the Interchain Foundation (ICF), is a decentralized network of parallel blockchains. Parallel blockchains are those that are connected by a central blockchain but do not otherwise interfere with each other's operations. When necessary, they can exchange information through the central blockchain. ICF has a strong focus on developing a scalable system to improve the interoperability between different blockchains, essentially creating an internet of blockchains.

In today’s blockchain industry, it is hard to witness direct value transfers or co-operation across different blockchains. It may not be a significant problem today, but at the speed that the blockchain industry is advancing, lack of Interoperability will be likely to halt economic activities in the long run. This is where Cosmos comes in. Cosmos is a network connecting many independent blockchains, called zones, and the first zone on Cosmos is called Cosmos Hub. The zones are powered by Tendermint Core which is a Practical Byzantine Fault Tolerance-like (PBFT) consensus engine. Because more validators can be added to the network, Tendermint Core is suited for scaling Proof of Stake (POS) blockchains. Blockchains that utilizes other consensus algorithms can be connected using adaptor zones. When connected, all transactions between zones will go through the hub.

Blockchain Interoperability

Nowadays the safest way to exchange crypto assets requires traders to trust third parties such as centralized exchanges. The market demand made it one of the most profitable businesses in today’s blockchain industry. However, they are vulnerable to attacks as we have witnessed the occurrence of hacking multiple times.

One solution to the problem mentioned above is through the use of on-chain atomic swaps. This method requires that both parties, on different blockchains, involved in the exchange of asset to sign off on the transaction to make it happen. The technology eliminates the third party and makes the transaction trustless. In September 2017 the first atomic swap was completed between Decred and Litecoin. Successful swaps were also confirmed between Ethereum and Bitcoin, Litecoin and Bitcoin later in the same year. However, the atomic swap has its limitations. The fastest processing time of such transaction depends on the slower block time of the two blockchains involved. Also, it is not applicable for some blockchains as it requires the smart contract on both sides has similar capabilities. Furthermore, on-chain atomic swaps reduce privacy as the payments on different chains can be linked.

A similar solution to atomic swap is the off-chain Lightning Networks (LN). Compared to the on-chain method, it is much faster (instant), cheaper for smaller transactions, and has better privacy properties. Cross chain swaps are also possible by opening two payment channels. Litecoin was one of the first cryptocurrencies to adopt this technology and can potentially help it to become a widely adopted currency. However, the use of LN requires the funds to be committed to the network for use thus somewhat removes the trustless property of previously mentioned atomic swaps. The commitment also makes the funds only useable on the LN until the payment channel is closed.

Another substitution to centralized exchanges is decentralized exchanges (DEX). As the name suggests, DEXs do not have central servers to maintain order books. Therefore, they are much more hacker proof. However, DEXs are not easy to use for beginners, at least at the current stage. On top of that, processing speed, as well as processing capacity, are just not as good as their centralized opponents. Those drawbacks hindered the adoption of DEXs and further led to low volume and low liquidity. Currently, one of most popular DEX, IDEX, processes US $2 million daily while a centralized exchange like Binance processes over US $900 million daily. Other popular projects like Kyber and 0x take different approaches to the liquidity problem but were not yet succeeded in bringing comparable volume.

One more method to improve blockchain interoperability is sidechains. A sidechain is a secondary chain that is connected to the main blockchain, using a two-way peg. Sidechains are maintained by their own miner/validator, thus provide an extra layer of security. The connection allows the information to free flow between connected chains and the exchange of assets from the main chain to sidechain and back. However, there is no actual transfer happening when using the two-way peg to exchange assets. Instead, token A is locked by smart contracts on Chian A and then some token B of equivalent value is released on chain B. The Cosmos blockchain we are covering in this article improves upon the sidechain technology, and we will cover how it works in detail in the next section.

The Inner Workings of Cosmos

To connect different zones to the hub, Cosmos uses an Inter Blockchain Communication (IBC) protocol. This setup allows each blockchain to work in parallel while retaining their security properties. The hub is a multi-asset distributed ledger that keeps track of the total number of tokens in each zone. It allows tokens to be transferred from one zone to another safely and quickly. The direct transfer means that liquidity from crypto exchanges are no longer needed to move value from one blockchain to another, thus, improving security.

Because new zones can be added to the hub at any time, the architecture design of Cosmos made it possible for the current blockchains to interoperate with future blockchain technologies. While each zone keeps their security characteristics and is separated from the failures of other zones, the security of the hub is paramount to the stability of the whole network. Therefore, at launch, the Cosmos Hub will require 100 decentralized validators that are globally distributed. Another 200 validators will be gradually added to the system in 10 years time. We will discuss the consensus algorithm in detail in a later section of this article.

Economic Model

The current proposed Cosmos economic model uses a single token called Atom. It is the only staking token on the Cosmos hub, and also serves as the fee token on the network in the current implementation. Atom holders can vote on governance, participate in block validation, and delegate to validators for income. The income for validators and delegators comes from the inflation of Atom and transaction fees. The reward paid to them is proportional to the number of Atoms they have staked against the total staked Atoms on the blockchain.

The annual inflation rate of Atom is adjustable depending on the percentage of total Atom supply that is staked on the whole network. When more than ⅔ of the total supply of Atoms is staked, the inflation can go as low as 7%. When less than ⅔ of the token is staked, the inflation can be as high as 20%. This mechanism is implemented to encourage staking tokens. Because those who do not participate in staking will not receive any inflationary Atoms nor transaction fees, their holdings are effectively diluted by the inflation rate every year. However, the inflation mechanism is somewhat experimental and may not guarantee a high staking percentage. Because, there exists a possibility that a low staking percentage, thus high inflation, could lead to quick devaluation of the Atom token. The drop in value may disincentivize some investors to hold the token.

In essence, Atom is not meant to be a medium of exchange nor store of value, but rather a tool for Cosmos governance. Since it has the low-liquidity design from the start, a fee token was proposed by ICF’s partner, the Tendermint development team. The fee token is called Photon where its core utility is paying for transactional costs. It is designed to have much higher liquidity than Atom tokens. To further accommodate other blockchains to connect through Cosmos hub, Tendermint team also proposed to include a whitelist of tokens (bitcoin, ethereum, etc.) to be accepted as fee token. However, those fee tokens are not part of Cosmos hub yet. At mainnet launch, Atom will be the only token, and all fee tokens need to be voted in by governance.

Consensus

Cosmos uses a POS BFT-base consensus algorithm that requires 100 validators at the launch of the network. Similar to other POS blockchains, validators have to stake (staking is called bonding on Cosmos) Atom tokens to participate in the consensus process. Those who do not wish to be validator can participate by staking their tokens and delegate to any of the 100 validators to share the block reward and transaction fees. Based on the implementation of the Tendermint consensus engine, block time can be as low as 1 second.

On other networks, delegators tend to vote for the validators who offers the most return. However, on Cosmos, several mechanisms are put in place to punish unreliable validators as well as their delegators. When a validator gets hacked or violates network protocols, both the validator and its delegators will get slashed a percentage of their staked Atom tokens.

Slashing

There are three main slashing conditions. First, double signing. If someone on chain A reports that a validator signed two blocks at the same height on chain A and chain B, this validator will get slashed on chain A. The second condition is based on unavailability. If a validator did not sign the past X blocks, then this validator will get slashed the amount proportionate to X. If X is above certain limit Y, then the validator will get unbonded which means the validator lose the right to produce blocks. The third condition is the occurrence of non-voting. If a validator failed to vote on a proposal and was reported, its stakes will receive a minor slash. The slashing conditions stipulate that validators should maintain a 100% uptime to avoid slashing. The slashing percentages for the conditions we have talked about are not yet available. However, no matter the percentage, the slash amount can mount up quickly if the validator is consistently unreliable.

Worth noting, Cosmos also encourages the community to exploit vulnerabilities that validators have. Once proven the exploit is successful, the validator and their delegators will get slashed 5% of their staked Atoms. The hacker will get rewarded the slashed Atom. This “Hacker Incentives” is put in place to identify weak links before they could do real damage. Also, this mechanism could help prevent whales on the network since the slash/reward is based on a percentage they are more attractive to hackers. Considering the risks, delegating to reliable and trustworthy validators, instead of those who offer low fees, is the best option for delegators.

Incentive model

The incentive model for validators and delegators are still in development and can be changed at any time. In the current implementation, both parties are rewarded the inflationary Atoms as well as transaction fees generated in the system. Validators are allowed to charge commision fees on the reward issued to their delegators to cover the cost of running a full node. There is a 2% tax on all transaction related rewards charged by the system to save in a reserve pool to reward future development proposals.

Suppose there is a validator, SpaceRocket, with 5% of the total staked Atoms on the Cosmos hub. Of the 5%, 40% is self-bonded, and 60% is delegated. Let’s further assume that the inflation on the network is 10%, and the validator charges a 15% commission fee. Then the income from inflationary Atoms for the first year can be calculated as follows:

Atom supply at launch: 230,000,000

Inflation: 230,000,000 * 10% = 23,000,000 Atoms

Total reward available to SpaceRocket: 23,000,000 * 1% = 230,000 Atoms

Commission charge: 230,000 * 60% * 15% = 20,700 Atoms

SpaceRocket’s total revenue: 230,000 * 40% + 20,700 = 112,700 Atoms

Revenue from transaction fees has a similar calculation as shown above only with a 2% tax charged by the governance model. Since the mainnet is not yet live, there is no information about how much a transaction would cost. We will not calculate the revenue from transactions at the moment and will update this article as soon as information is available.

Conclusion

Cosmos is an ambitious project that can bring interoperability to the blockchains. It could be the answer to some of the most challenging problems the industry is facing such as scalability and upgradability. The technology is intriguing; however, its development is not without hurdles. The Cosmos mainnet launch has been postponed multiple times. According to the published roadmap, the current launch schedule merely is set in the “future” without a specific date.

While the Cosmos is still in its infancy, we can already see its potential interoperability space. Based on how the centralized financial system has evolved, direct value transfers and interoperability is the path forward for blockchains. We shall be hopeful that Cosmos will have all of its functionality in place soon and attract mainstream adoptions.

References

https://blog.cosmos.network/economics-of-proof-of-stake-bridging-the-economic-system-of-old-into-the-new-age-of-blockchains-3f17824e91db

https://blog.cosmos.network/the-internet-of-blockchains-how-cosmos-does-interoperability-starting-with-the-ethereum-peg-zone-8744d4d2bc3f

https://cosmos.network/docs/resources/delegator-faq.html#revenue

https://cosmos.network/roadmap

https://figment.network/cosmos/hubble/gaia-7005

https://github.com/cosmos/cosmos/blob/master/VALIDATORS_FAQ.md

https://medium.com/tendermint/proposed-cosmos-fee-token-codename-photon-e0927daf5c4c