## Introducing Crednet **Crednet** [name TBC] is addressing the need for a comprehensive reputation system within the permaweb. It operates on a dual-token model, leveraging decentralized IDs, ensuring profile portability and permanence across applications. The system is built on the ao hyper-parallel computer, operating on Arweave, and focuses on sybil resistance, proof of personhood, ad targeting, content recommendation, and influencer advertising. ## Description of tokens ### $REP token (reputation score) * **Nature**: Soulbond (non-transferable) * **Accumulation**: * Connecting various social networks and wallets (proof of uniqueness + open personal data) * Receiving endorsements by higher $REP holders * Stakeholders staking $CRED in the account by buying into its bonding curve * Participation in network quests and advertisers' campaigns, with possible NFT badges increasing the score * **Reduction**: * Inactivity leading to a linear decrease * Getting blocked by individuals or entities with a higher $REP * Negative $CRED burning by others (high-cost bad reputation signaling) * **Utility**: * Earn $CRED * Decentralized ID and portable profile * Recognition of enhanced influence within the network * Access to better ad inventory * **Use cases**: * Login in apps * Airdrops * Ad buying and targeting * Influencers sponsorship ### $CRED token (reputation credit) * **Nature**: Transferable currency * **Accumulation**: Proportional distribution based on one’s $REP score, with network-inflation mechanics * **Usage**: * Advertisers for campaign creation * Influencers for brand promotion and content creation * Staking to increase brand or product reputation ($REP score) ## Player perspectives and incentive alignment ### Users & influencers **Incentives**: Build and maintain $REP through engagement and network contributions, earning $CRED as a reward. Increased reputation leads to better advertising opportunities and higher earnings and performance * **Pros**: Enhanced visibility, earning potential, and network influence * **Cons**: Requires continuous engagement to maintain or increase $REP ### Advertisers & brands **Incentives**: Purchase $CRED for advertising, benefiting from the highly targeted and reputation-based system. Brands can effectively reach their desired audience * **Pros**: Targeted and efficient ad campaigns, reduced expenditure on non-relevant impressions (e.g., bots) * **Cons**: Requires understanding and adaptation to the reputation-based system ### General users **Incentives**: Engage with content and platforms aligned with their interests, assured by the reputation system. Get better content, app and contacts recommendations * **Pros**: Relevant content and ads, enhanced user experience * **Cons**: Potential over-reliance on system's recommendation, limiting exposure to diverse content ## Prosocial aspect Crednet's design inherently promotes prosocial behavior by: * Promoting authentic engagement and rewarding quality content creation * Mitigating the impact of bots and fake accounts through soft sybil resistance * Offering a transparent and decentralized recommendation system ## MVP * **To be defined**: * $REP weights and decay rate * Paramaters for bonding curves * Inflation and distribution rate for $CRED * **Core features**: * Basic implementation of decentralized IDs * Initial rollout of the $REP scoring system, including social network connections and endorsement mechanisms * Introduction of the $CRED distribution system based on $REP scores * Initial framework for ad targeting and content recommendation * **Goals**: * To validate the dual-token model and its impact on user engagement * To establish foundational partnerships with key influencers, brands and media * To gather an initial userbase to get feedback for iterative improvements * **Focus**: * Establishing reliable and transparent scoring algorithms * Building partnerships with influencers, brands and media for early adoption * Launching initial marketing and community-building efforts * Ensuring a great user experience with dead-simple user interfaces ## Why Crednet ### A reputation-based decentralized ID Drawing insights from this [article on reputation-based systems](https://a16zcrypto.com/posts/article/reputation-based-systems/), Crednet can represent a critical solution in the evolving landscape of digital identity and reputation. Reputation systems are foundational in building trust and enabling more nuanced interactions in digital spaces. Crednet's approach to reputation not only enhances trust but also bridges the gap between digital identities and real-world applications. It offers an alternative to KYC approaches or biometric methods (like Worldcoin), while ultimately remaining compatible with them as optional extra modules of verification. ### Liquifying reputation through $CRED The unique aspect of Crednet is its use of soulbound $REP tokens, which are non-transferable and represent the reputation score of an entity. To add a layer of liquidity and incentivization, $CRED tokens are regularly earned based on the $REP score. This system ensures that while reputation remains non-transferrable and genuine, it also rewards active and reputable participants in a tangible way. ### Soft sybil resistance While Crednet initially lacks a KYC process, making it not entirely sybil-resistant, the high cost of creating and maintaining multiple high-reputation identities acts as a deterrent against sybil attacks. In the future, centralized providers could add layers of real-world ID verification to further strengthen this aspect. ### Expanding reputation beyond individuals Crednet's vision extends reputation to be atomic – not just limited to individuals but also applicable to content pieces, mediums, apps and platforms. This approach acknowledges that in a post-AI world, every content piece, whether an image in an article or a clip in a video, has its own identity and reputation. This granular approach to reputation provides a more comprehensive and accurate representation of an entity's influence and trustworthiness. ### Normalization and application of reputation scores For enhanced clarity and usability, reputation scores can be normalized on a 100-point scale along with the raw $REP score. Entities can set thresholds for interaction (e.g., only engaging with content or individuals above a certain reputation score), which can lead to higher-quality interactions and content dissemination. This approach also mitigates issues like bot-driven impressions in programmatic advertising, adding efficiency and value to advertising expenditures. ### Investing in reputation and content An intriguing aspect of Crednet is the possibility of staking against the reputation of content by investing in its bonding curve. This feature turns content ranking into a dynamic and participatory process, much akin to TCRs (Token Curated Registries), where stakeholders are incentivized to identify and support content they believe will be impactful or culturally significant. ### Decentralized recommender system By allowing entities with a determined reputation to invest in content based on its perceived value and impact, Crednet is effectively creating a decentralized recommender system. This system is fueled by the collective judgment and participation of its users, leading to a more democratic and transparent process of content curation and recommendation.