Taking a look at two oracle projects: Witnet 2.0 versus Band Protocol

We’re starting a new series where we dive a little deeper into some oracle projects and go over how Witnet 2.0 differs. Our first one is Band Protocol.

Published in

The Witnet Oracle Blog

5 min readOct 26, 2023

Before we get into the meat and potatoes of this article, I should make note of the fact that while Witnet 2.0 is not live yet, we are choosing to compare it to other projects due to the fact that we have a solid foundation for how the network will look when its migrated and data and models based on that understanding.

The Layer 1

Witnet is a layer 1 blockchain oracle that was custom built in the Rust programming language. It was single-purpose and designed to do one thing very well; fulfill data requests in a fully decentralized way. Every design decision was built around doing one thing incredibly well, similar to how Bitcoin was built to do one thing incredibly well: act as money. Witnet was built with the same ideals and values in mind as Bitcoin; permissionless nature, censorship resistant, immutable, unconfiscatable, unstoppable, decentralized, etc. This means Witnet was designed for the long term game. If people care about a free internet, they’ll care about Witnet.

Band, on the other hand, is a Cosmos SDK chain; BandChain. This makes it a layer 1 but built on top of another network that has its own consensus rules, block production rules, inflation, etc. This isn’t necessarily a bad thing, but it means that the actual oracle can’t be built in a specific method that is niche to the way oracles are. As mentioned earlier, Witnet was crafted with due diligence and research directly relating to the blockchain oracle problem and the blockchain trilemma. It was built to tackle these problems as they relate to data feeds. Band has to first and foremost follow consensus rules of the Cosmos network, making it more complicated to engineer specific rules unique to oracles. This also means that Band, and all the protocols using it, is/are tied to the fate of the Cosmos ecosystem.

Witnet as a layer 1 isn’t tied to the fate of any chain. Ethereum could go down tomorrow and Witnet would still operate on Kava, Cronos, Conflux, Syscoin, and all the other chains the oracle is live on.

$WIT and $BAND

Each oracle has a coin or token of their respective network. There aren’t much differences between the coins on the blockchain. Both coins act as a security measure to protect against manipulation by staking or collateralizing a data request, both coins are used to pay network fees within the network, and both coins are used as block rewards for the proposer.

Consensus Mechanism and Validator Selection

The consensus mechanism of the respective networks aren’t that different, but as mentioned prior, BandChain must first and foremost operate on the Cosmos consensus mechanism which might inadvertently have differing interests at the base compared to what an oracle might need.

Witnet (will) operate(s) on a Proof of Stake consensus mechanism that was single-purpose built for oracle needs. BandChain uses a Tendermint consensus mechanism that works very similar to a Proof of Stake system, but has somewhat stronger ties to delegating tokens to nodes for stronger security.

The way Witnet will operate via a Proof of Stake model, and the algorithm fundamental to the PoS system, is referred to as “power”. Power is the product of Total Stake and Stake Age. Total stake is the total amount of $WIT coins that a node has in the stake tracker (the way staked coins are accounted for) whereas stake age is a counter that keeps track of when a node has performed a task on the network (fulfilling a data request or adding a block). Stake age is reset when a task as been performed and is a simply yet elegant mechanism to divert any ability to control the network or allow any particular node with a high amount of stake to add most of the blocks- this algorithm helps to level the playing field.

Band has a mechanism that is somewhat similar, based on two fundamental concepts. Based on their documentation: 1) A validator with a higher voting power should have a higher chance of being chosen than one with a lower voting power and 2) Every validator should still have a chance to be selected. Validators are chosen from a pool based on random seed generation.

Data Sources

Witnet and Band have a somewhat similar approach to data, however Band uses a few data aggregators like CoinMarketCap and CoinGecko on top of first order sources. Witnet doesn’t use these second order APIs due to the strong nature of trust required. These sources are considered second order sources because they aggregate data from multiple places and there is a heavy amount of trust needed when using this data in a smart contract. Witnet also has its own aggregator method as part of the consensus model for data requests. Using only first order sources (like exchanges) allows multiple nodes to go to multiple different places to retrieve a piece of data and aggregate it together, those who are outside of a deviation are filtered out. Band has a similar aggregator system to Witnet. The only difference is where the data comes from (first order sources versus second order)

Validators and Nodes

BandChain has a seemingly permissioned set of around 100+ validators, most likely with some degree of KYC, that are randomly selected to perform functions and cryptoeconomically bound by the amount they stake. Additionally, because running a Cosmos or BandChain node/validator is complicated, average holders can delegate $BAND to their validator of choice. These delegators are trusting Band nodes to act in their best interests and the interests of the network as a whole.

Witnet has taken an entirely different approach. As mentioned before, Witnet built the network akin to Bitcoin’s fundamental goals. Witnet built the network entirely on cryptoeconomic foundations and there is no trust involved to operate the network, no KYC, and no asking for permission. The $WIT coin provides the incentive for nodes to remain honest and has been the backbone of the network for years. Additionally, many people run their own nodes so that they can verify the network security instead of trusting someone else to do it correctly.

Wrapping Up

I hope this was a helpful summary about the differences between Witnet and Band and will guide developers, stakers, and holders to learn more about each oracles and their respective architecture.

Connect with Witnet