Intuition 백서 전문

e such as Amazon can be represented as a literal interpretation. For instance, adding a product to a marketplace can be as straightforward as creating a Triple like [Product] [listedOn] [Amazon]. Expanding on this concept, various stakeholders can publish their interpretations of product reviews. For example, a company might aggregate all reviews and provide a summary score, while an independent consumer protection group could highlight reviews mentioning safety concerns. 5. Algorithms (a) Users can choose from various algorithms to ﬁlter and sort reviews based on their preferences. For instance, an algorithm can prioritize reviews from users with similar purchase histories or highlight reviews that mention speciﬁc product features. 6. Reality T unnel 49 (a) Users can compose together interpretations and algorithms to create curated ‘views’ of data, allowing them to ﬁnd the voices they trust among thousands of anonymous voices. 7. Portable Identity and Data (a) Identities and data become universal, and can be referenced and integrated anywhere across the web. Users can see a product’s aggregated reviews across Amazon, Facebook Marketplace, Craigslist, or eBay, provided the same canonical identiﬁer for the products is used in each context. And, even if Amazon were to shut down, users’ data and lists of favorites would still exist, ensuring continuity and control over their digital interactions. By integrating these primitives into contexts and interactions that cannot yet be fully ‘trustless,’ Intuition establishes a decentralized trust layer. This layer allows users to interact with conﬁdence, knowing that the information they rely on is transparent, veriﬁable, and free from centralized manipulation. For instance, when buying a product, users can access a wealth of credible reviews, ﬁltered and presented according to their preferences, all while maintaining control over their data and identities. By leveraging these tools, we can build a more trustworthy and user-centric digital ecosystem, addressing many of the inherent ﬂaws in the current Web2 trust stack and paving the way for a more decentralized and equitable internet. 8.4 Design Decision Rationale Having introduced the core primitives underpinning the Intuition system, before we transition to a more detailed examination and implementation details, it is imperative to delve deeper into the rationale behind these design decisions. By exploring the reasoning and guiding principles that informed the creation of these primitives, we can better appreciate their signiﬁcance and the critical role they play in addressing contemporary challenges in digital trust. 8.4.1 The Vision: A More Semantic W eb In our pursuit of a trustful interaction layer, it is evident that building trust requires more comprehensive and reliable agent-centric information. The vision of a semantic web is central to achieving this goal. By ensuring that data is not only abundant but also structured and accessible, we can foster an environment where trust can ﬂourish through informed decision-making. To illustrate the vision of a more semantic web, consider the process of adding notes about a book to a reading list. Typically, this might involve jotting down the book title, the author, and perhaps some additional information in raw, unstructured text. Alternatively, one might paste the URL of the book from Amazon, which provides Amazon’s link, data, and reviews. However, this approach restricts access to the data of a single platform, rather than leveraging the entire internet’s wealth of information. Imagine if this book had a unique, globally persistent, canonical digital identiﬁer that could be referenced universally and directly. Using this identiﬁer, it would be possible to instantly access all related data from across the web—ratings, reviews, purchase links, and more—regardless of the platform where the data resides. This identiﬁer would serve as a uniﬁed reference point, aggregating information from multiple sources, thus providing a comprehensive view of the book. This would transcend Amazon’s perspective, allowing access to reviews from Goodreads, purchasing options from various retailers, academic critiques, related articles, and even social media discussions. Additionally, any data contributed about this book—such as personal notes, reviews, or tags—would be easily discoverable by others, enriching the global dataset. When annotating the book with thoughts or adding tags for easier referencing, these contributions become part of the shared knowledge pool. Other users, regardless of where they are on the web, can beneﬁt from these inputs, just as they contribute their own. Standardizing the structure of data about the book ensures consistency and interoperability across platforms. For instance, reviews can be categorized by sentiment, ratings can follow a universal scale, and purchase links can be formatted uniformly. This standardization allows for seamless integration and comparison of data from various sources, making it easier for users to interpret and utilize the information. Incentivizing more and better expressions is also crucial. In this system, users should be encouraged to contribute high-quality data. Detailed and insightful reviews, comprehensive annotations, and accurate tagging are all examples of contributions that might be rewarded. This creates a virtuous cycle where the richness and reliability of the dataset continually improve as more users participate and contribute. 50 This vision encompasses a more decentralized, semantic web, where data is seamlessly aggregated and ac- cessible through universally recognized identiﬁers. Such a system fosters a dynamic and interconnected information ecosystem, vastly improving the eﬃciency and richness of digital interactions. By transcending the silos of individual platforms, we move towards a uniﬁed digital space where information ﬂows freely and is easily accessible to all, ultimately enhancing the ability to ﬁnd, share, and build upon knowledge. The Challenge: Achieving Consensus In an entirely permissionless environment, where new data structures, schemas, formats, standards, and identiﬁers can be created and leveraged without permission, coordination is a complex challenge. The decentralized nature of the web aligns with the core principle of avoiding any single authority. Instead, consensus must emerge organically from the community, reﬂecting the diverse and often conﬂicting interests within it. This decentralized approach is crucial to maintaining the openness and democratic nature of the system. In traditional systems, a centralized authority might impose standards and ensure adherence. For example, the International Standard Book Number (ISBN) system is centrally managed, ensuring that every published book has a unique identiﬁer. However, relying on a singular authority to enforce such standards contradicts the ethos of decentralization. Centralization places an entity in a position of extreme power, which goes against the principles of promoting openness and avoiding monopolistic control. Therefore, it is imperative that the process of reaching consensus be more democratic and community- driven. This involves the collective participation of users, developers, and stakeholders who contribute to and maintain the system. This approach encourages diversity and inclusivity, allowing for a broader range of perspectives and innovations. However, it also means that agreement on standards and identiﬁers can be slow and contentious, requiring extensive collaboration and negotiation. To alleviate the coordination issues around achieving decentralized social consensus, mechanisms to facilitate collaboration and negotiation among a diverse set of participants must be put in place. Economic incentives, transparent governance models, and other consensus mechanisms play crucial roles in this process, driving participants towards mutually beneﬁcial agreements while maintaining the decentralized nature of the system. The Market-Driven Approach to Consensus The market-driven approach to consensus is a founda- tional principle within the cryptocurrency ecosystem. This method utilizes economic incentives and social consensus to establish standards for digital assets, contracts, or even entire blockchain networks. Mecha- nisms like Proof of Work (PoW) and Proof of Stake (PoS) are designed based on game-theoretic principles to ensure that participants’ actions align with the network’s best interests. These principles can extend beyond ledger states to include the establishment of canonical identiﬁers and the convergence on arbitrary industry standards. For example, in the cryptocurrency realm, determining which PEPE coin should be considered the canonical version among thousands of identical deployments is primarily inﬂuenced by the market, given the absence of tangible fundamentals. The PEPE coin with the largest market capitalization becomes the recognized standard. This outcome is driven by the collective actions and decisions of the community, reﬂecting a preference for the most widely adopted version. Prediction markets also demonstrate this market-driven consensus by converging on the correct answer as more data becomes available. Participants in these markets are economically incentivized to align their predictions with the most accurate outcomes, resulting in a consensus that mirrors the collective intelligence of the participants. The ERC-20 token standard further illustrates this approach. Before its introduction, the Ethereum ecosys- tem lacked a uniform protocol for tokens, leading to fragmentation and interoperability issues. Developers and users recognized the beneﬁts of adopting a common standard: interoperability, ease of integration, and enhanced security. By adhering to the ERC-20 standard, developers ensured that their tokens could seam- lessly interact with various wallets, exchanges, and decentralized applications (dApps). Conversely, deploying a token that did not follow the ERC-20 standard would lead to signiﬁcant challenges. Such a token might not be compatible with existing wallets, making it diﬃcult for users to store or transfer it. It could also face integration issues with exchanges and dApps, limiting its usability and adoption. This lack of interoperability would create friction for both developers and users, discouraging the adoption of non-standard tokens. In these examples, the power of incentives is evident: users naturally converge when it best serves their interests. The collective actions of community members, driven by their individual economic motivations, 51 result in the organic selection of the most robust and valuable options. This market-driven approach ensures that the best solutions rise to prominence through organic consensus rather than imposed mandates. The Role of Economic Incentives in Intuition Intuition leverages this market-driven approach by incorporating economic incentives to drive convergence on data structures, schemas, standards, and identi- ﬁers, as well as to encourage participation. Every action within the system incurs a nominal fee, similar to a gas cost in blockchain networks, which is used to grant fractional ownership of the data being interacted with and to compensate all other current owners. This mechanism ensures that participants are ﬁnancially motivated to use and support canonical identiﬁers and standardized methods of expression, as they beneﬁt from future interactions involving these elements. For example, when a user makes a claim about a concept or entity, they gain fractional ownership of that claim and its associated data. Subsequent interactions that reference this data will generate rewards for all those who have previously engaged with it, thus incentivizing the use of widely accepted and recognized identiﬁers and standardized expressions. This alignment of individual economic interests with the collective goal of creating a cohesive and interoperable system is central to Intuition’s design. By adopting this market-driven approach, Intuition ensures that the most eﬀective and reliable data struc- tures, schemas, standards, and identiﬁers naturally gain prominence through community consensus rather than top-down directives. This system encourages the adoption of standardized identiﬁers and data struc- tures, reducing fragmentation and enhancing interoperability across the digital landscape. Furthermore, the economic incentives within Intuition create a dynamic and adaptable ecosystem capable of evolving in response to changing needs and challenges. As participants continuously interact with and con- tribute to the system, they collectively shape and reﬁne the standards and practices that deﬁne the ecosystem. This is a signiﬁcant departure from more centralized solutions, such as schema.org, which rely on predeﬁned schemas and centralized control over data standards. Intuition’s decentralized and incentive-driven model allows for a more ﬂexible and responsive approach. As new types of data emerge and the community’s needs evolve, participants can propose and adopt new standards organically. Economic incentives ensure that the most useful and widely accepted standards are adopted, while less eﬀective ones are naturally phased out. This adaptability is crucial in a rapidly changing digital landscape where new technologies and use cases are constantly emerging. Reducing Fragmentation & Solving the Problem of F olksonomies The concept of community- driven data structuring, exempliﬁed by folksonomies, is not new. Folksonomies are user-generated systems for classifying and organizing information, that often lead to signiﬁcant fragmentation. Unlike structured and hierarchical taxonomies, folksonomies allow individuals to use their own terms and descriptions for the same concepts. While this ﬂexibility can be beneﬁcial for personal organization, it frequently results in varied tags or labels being applied to the same item, creating a fragmented data landscape that is diﬃcult to reconcile. For instance, in a folksonomy, one user might tag an article about climate change with “global warming,” another might use “climate crisis,” and yet another might simply tag it “environment. ” Although all these tags refer to the same overarching topic, the lack of standardized labeling leads to scattered information, making it challenging to aggregate and analyze data comprehensively. Intuition addresses this problem by incentivizing convergence on popular choices. By employing economic incentives, Intuition encourages users to adopt and promote canonical identiﬁers and descriptions. Every interaction within the system, such as tagging or classifying an entity, involves a small fee that contributes to fractional ownership of the data being interacted with. Users are motivated to use established identiﬁers because doing so increases their potential for economic rewards from future interactions. This economic model promotes a more uniﬁed and coherent data structure by minimizing the fragmentation inherent in folksonomies. When users converge on standardized methods of expression - a concept which can be likened to the concept of language - it becomes easier to aggregate, query, and analyze data. The result is a more organized and accessible information ecosystem. For example, if multiple users are tagging articles about climate change, they are incentivized to use the most popular and widely accepted tag, such as “climate change,” rather than creating new or slightly diﬀerent tags. This convergence reduces the scattering of related information and enhances the overall usability of the data. 52 Intuition’s economic incentives naturally create a token-curated registry (TCR) for expression, composed of TCRs for identiﬁers, schemas, formats, and standards. As users interact with and endorse speciﬁc identiﬁers and data structures, they gain ownership in these entities. This ownership makes the most useful and accepted identiﬁers and frameworks more easily discoverable and widely adopted, guiding the community towards standardized terms and improving the overall coherence and utility of the data. Keynesian Beauty Contest in Intuition The Keynesian Beauty Contest principle is central to how Intuition fosters social consensus. In economics, a Keynesian Beauty Contest is a scenario where participants are rewarded for choosing options they believe others will choose, rather than their personal preferences. This concept is applied to Intuition’s selection of identiﬁers and modes of expression, ensuring that the most popular and widely accepted options become canonical. Application in Identiﬁer and Description Selection Judgment Based on Popularity In a Keynesian Beauty Contest, participants aim to predict the most popular choices among all participants. This method has been shown to be highly eﬀective in reaching social consensus, as it aligns individual incentives with collective preferences. The principle was ﬁrst introduced by John Maynard Keynes in his 1936 book “The General Theory of Employment, Interest, and Money,” where he described a newspaper beauty contest in which participants were asked to choose the most popular faces rather than their personal favorites. In Intuition, this concept translates to users being incentivized to choose identiﬁers and descriptions they believe will be widely adopted by others. This ensures that the most useful and accurate representations gain prominence. For instance, if a majority of users prefer the identiﬁer [New York City] over [NYC] or [NY], the system will naturally converge on [New York City] as the canonical identiﬁer. Evidence of Eﬀectiveness The eﬀectiveness of the Keynesian Beauty Contest in predicting popular choices and achieving consensus has been demonstrated in various studies and real-world applications: 1. Experimental Economics : Studies have shown that participants in Keynesian Beauty Contest experiments often converge on the same choices, reﬂecting a strong predictive ability about the preferences of others. For example, in an experiment conducted by Nagel (1995), participants were asked to choose a number between 0 and 100, with the goal of selecting the number closest to two- thirds of the average choice. Over multiple rounds, participants quickly learned to anticipate others’ choices, leading to convergence around the predicted equilibrium. 2. Prediction Markets: Prediction markets, which operate on similar principles to Keynesian Beauty Contests, have been shown to accurately forecast outcomes in various domains, including elections, sports, and ﬁnancial markets. These markets reward participants for correctly predicting popular outcomes, and their accuracy is well-documented. For instance, a study by Berg, Forsythe, Nel- son, and Rietz (2008) found that prediction markets outperformed traditional polling methods in forecasting the results of the U.S. presidential elections. 3. Crowdsourcing and Consensus Mechanisms : Platforms like Wikipedia and Stack Overﬂow leverage similar consensus mechanisms to ensure the accuracy and reliability of information. Con- tributors are incentivized to align their contributions with the community’s expectations, leading to high-quality, consensus-driven content. Research by Kittur and Kraut (2008) highlighted the eﬀec- tiveness of Wikipedia’s collaborative model in achieving high information quality through collective intelligence. Practical Example in Intuition Imagine users need to describe a widely known concept such as “Machine Learning. ” Multiple identiﬁers could exist: [ML], [MachineLearning], [AI], [ArtiﬁcialIntelligence]. Through the Keynesian Beauty Contest principle, users are incentivized to converge on the identiﬁer they believe will be most widely adopted, and not necessarily the one they personally prefer. Over time, economic incentives may drive the majority to use [Machine