Karma3 Labs is building an open reputation protocol that uses the EigenTrust algorithm. The protocol enables a ranking and reputation infrastructure for web3 apps and marketplaces, leveraging on-chain data and social attestations.

Developers can use the protocol to enable key features such as search and discovery, personalized recommendations, trust scores and sybil resistance. The data and algorithm used in computing the ranking and reputation scores can be easily configured by developers or communities. The public infrastructure relieves developers from the cost and time of building custom in-house rankings, which may also not be open and composable.

Reputation attestations are high-value signals, enabling a trust layer for human coordination. As an important public utility, our reputation infrastructure will enable computation of social interactions at scale by providing a measurement of trust across p2p communities and applications. Application developers can leverage this as a trust, safety or recommendation layer to improve the end user experience and drive user growth and scale.

EigenTrust is a reputation-based trust management algorithm designed to address the problem of malicious nodes in decentralized networks. Developed by Sepandar Kamvar, the protocol was initially designed to address the issue of malicious nodes in peer-to-peer file-sharing systems such as Napster and Gnutella. Due to the record industry's efforts to flood (sybil) these networks with malicious peers, a reputation system was necessary to help users identify trustworthy nodes to download songs from and avoid unreliable ones.

EigenTrust's fundamental concept is that a person's reputation is recursively determined by how much they are trusted by the individuals around them, with each of those opinions weighted by the opinion holder’s reputation itself. The EigenTrust computation leverages p2p attestations or vouching data as input signals. Additionaly the compute can be executed in a distributed manner to minimize trust and make the results verifiable.

EigenTrust computes a global trust value for any given user by aggregating how everyone in a network trusts you. The basic principle behind EigenTrust is, if I trust my friend’s opinion, to a degree I also trust my friend’s friend’s opinion. This establishes a local trust value between two users. Since this level of trust applies to all my friends, their friends and their friends’ friends, there’s a level of transitive trust that occurs between me and my friend’s friend.

In the end, every person in the network will have a global trust value, a measurement of trust between any two users in a network.

web3 presents the perfect setting to use on-chain, social attestation data and apply EigenTrust computation to enable an open, configurable and decentralized ranking and reputation infrastructure.

Developing reputation infrastructure in-house for every app and use case is redundant and costly, especially because most of the data is either on-chain or available openly over a shared data infrastructure.

It is also difficult for developers to shift focus from their core strengths and value propositions towards designing and maintaining proprietary algorithms. It takes a significant amount of time and skill to craft the right heuristics and the marginal benefits diminish over time. Having a centralized computation back-end also restricts developers in many ways — centralized liability and point of trust and perpetuates lack of trust from the community.

We believe that the reputation layer in web3 should not be based on private data and proprietary algorithms but on a verifiable and shared ranking and reputation infrastructure.

We want to make it easy for developers to access context-specific on-chain data to determine reputation. Developers should also be able to use open-source, verifiable algorithms like EigenTrust to determine rankings and recommendations, which can be adopted or configured by community consensus.

We believe that social attestations will be a powerful input signal for creating reputation and trust systems in web3. Attestations enable high-signal validation of a trust vector between any two individuals or entities, allowing for the implementation of a fair computation. The attestations can be either existing on-chain transactions between any two wallets or new explicit attestations issued within an application.

The computation of trust values will depend on different trust signals, which are easily customizable to the use case and context of reputation. Developers can build their own framework to improve the fidelity of local trust levels.

For example, in social networks, many types of attestations can be proposed by developers. The topical axes bear higher fidelity to specific activities than, say, a simple follower-based pairwise trust, and in general, cater much better to users’ interests. We encourage developers to try to bring topical/fine-grained trust data as well as more general/coarse-grained trust data.

We believe that web3 needs an open and shared reputation infrastructure that enables apps and communities to scale quickly. Developers can save time and resources by using reputation data that is composable and open-source. Over time, rankings across different contexts drive innovation and composability which will create network effects for developers and users.

Developers should be able to bootstrap the reputation of their community by composing many different EigenTrust scores across contexts. For instance, for Optimism DAO to create their community reputation or ranking, they could use a combination of Uniswap attestation scores, community governance rankings and Optimism creator attestation scores.

Network effects start to accrue once developers start benchmarking reputation heuristics for their use case and start relying on EigenTrust computation for their rankings. There can be a curation market for different types of reputation rankings, which strengthens the use of the shared infrastructure for reputation. Just like programming languages accrue network effects, the customizable reputation heuristic layer will breed similar network effects.

We have started implementing EigenTrust rankings for a variety of use cases like web3 social recommendations, p2p attestation based reputation compute for communities and powering open rankings for NFTs and wallets for consumer apps and marketplaces. 

If you are a developer and want to use our alpha APIs, fill this form

A trustless reputation and ranking system is essential to the success and widespread adoption of web3, yet establishing trust in a decentralized system is a complex challenge. That’s why Karma3 Labs built a strong conviction by studying how the early design decisions around search, rankings and review-ratings systems helped web2 companies acquire, engage and monetize billions of users.

• Website

• Join our team

• Twitter

• Gitbook

Karma3 Labs is building an open reputation protocol that uses the EigenTrust algorithm. The protocol enables a ranking and reputation infrastructure for web3 apps and marketplaces, leveraging on-chain data and social attestations.

Developers can use the protocol to enable key features such as search and discovery, personalized recommendations, trust scores and sybil resistance. The data and algorithm used in computing the ranking and reputation scores can be easily configured by developers or communities. The public infrastructure relieves developers from the cost and time of building custom in-house rankings, which may also not be open and composable.

Reputation attestations are high-value signals, enabling a trust layer for human coordination. As an important public utility, our reputation infrastructure will enable computation of social interactions at scale by providing a measurement of trust across p2p communities and applications. Application developers can leverage this as a trust, safety or recommendation layer to improve the end user experience and drive user growth and scale.

EigenTrust is a reputation-based trust management algorithm designed to address the problem of malicious nodes in decentralized networks. Developed by Sepandar Kamvar, the protocol was initially designed to address the issue of malicious nodes in peer-to-peer file-sharing systems such as Napster and Gnutella. Due to the record industry's efforts to flood (sybil) these networks with malicious peers, a reputation system was necessary to help users identify trustworthy nodes to download songs from and avoid unreliable ones.

EigenTrust's fundamental concept is that a person's reputation is recursively determined by how much they are trusted by the individuals around them, with each of those opinions weighted by the opinion holder’s reputation itself. The EigenTrust computation leverages p2p attestations or vouching data as input signals. Additionaly the compute can be executed in a distributed manner to minimize trust and make the results verifiable.

EigenTrust computes a global trust value for any given user by aggregating how everyone in a network trusts you. The basic principle behind EigenTrust is, if I trust my friend’s opinion, to a degree I also trust my friend’s friend’s opinion. This establishes a local trust value between two users. Since this level of trust applies to all my friends, their friends and their friends’ friends, there’s a level of transitive trust that occurs between me and my friend’s friend.

In the end, every person in the network will have a global trust value, a measurement of trust between any two users in a network.

web3 presents the perfect setting to use on-chain, social attestation data and apply EigenTrust computation to enable an open, configurable and decentralized ranking and reputation infrastructure.

Developing reputation infrastructure in-house for every app and use case is redundant and costly, especially because most of the data is either on-chain or available openly over a shared data infrastructure.

It is also difficult for developers to shift focus from their core strengths and value propositions towards designing and maintaining proprietary algorithms. It takes a significant amount of time and skill to craft the right heuristics and the marginal benefits diminish over time. Having a centralized computation back-end also restricts developers in many ways — centralized liability and point of trust and perpetuates lack of trust from the community.

We believe that the reputation layer in web3 should not be based on private data and proprietary algorithms but on a verifiable and shared ranking and reputation infrastructure.

We want to make it easy for developers to access context-specific on-chain data to determine reputation. Developers should also be able to use open-source, verifiable algorithms like EigenTrust to determine rankings and recommendations, which can be adopted or configured by community consensus.

We believe that social attestations will be a powerful input signal for creating reputation and trust systems in web3. Attestations enable high-signal validation of a trust vector between any two individuals or entities, allowing for the implementation of a fair computation. The attestations can be either existing on-chain transactions between any two wallets or new explicit attestations issued within an application.

The computation of trust values will depend on different trust signals, which are easily customizable to the use case and context of reputation. Developers can build their own framework to improve the fidelity of local trust levels.

For example, in social networks, many types of attestations can be proposed by developers. The topical axes bear higher fidelity to specific activities than, say, a simple follower-based pairwise trust, and in general, cater much better to users’ interests. We encourage developers to try to bring topical/fine-grained trust data as well as more general/coarse-grained trust data.

We believe that web3 needs an open and shared reputation infrastructure that enables apps and communities to scale quickly. Developers can save time and resources by using reputation data that is composable and open-source. Over time, rankings across different contexts drive innovation and composability which will create network effects for developers and users.

Developers should be able to bootstrap the reputation of their community by composing many different EigenTrust scores across contexts. For instance, for Optimism DAO to create their community reputation or ranking, they could use a combination of Uniswap attestation scores, community governance rankings and Optimism creator attestation scores.

Network effects start to accrue once developers start benchmarking reputation heuristics for their use case and start relying on EigenTrust computation for their rankings. There can be a curation market for different types of reputation rankings, which strengthens the use of the shared infrastructure for reputation. Just like programming languages accrue network effects, the customizable reputation heuristic layer will breed similar network effects.

We have started implementing EigenTrust rankings for a variety of use cases like web3 social recommendations, p2p attestation based reputation compute for communities and powering open rankings for NFTs and wallets for consumer apps and marketplaces. 

If you are a developer and want to use our alpha APIs, fill this form

A trustless reputation and ranking system is essential to the success and widespread adoption of web3, yet establishing trust in a decentralized system is a complex challenge. That’s why Karma3 Labs built a strong conviction by studying how the early design decisions around search, rankings and review-ratings systems helped web2 companies acquire, engage and monetize billions of users.

• Website

• Join our team

• Twitter

• Gitbook