Introduction

Traditional marketplaces tend to be rent-seeking, charging asymmetric fees relative to the value of the services they provide. These networks create negative externalities on society, reduce economic efficiency, prevent emerging markets, and fail to adequately fill incentive gaps in markets. Over time, marketplaces accumulate power and trend towards rents that are equal to the margins of their suppliers. As the platform grows, fees to buyers increase and earnings for sellers decrease. Marketplaces are therefore economically sub-optimal and fail to meet the demands of modern markets.

A decentralized, permissionless network leverages the local knowledge of a broader group of participants and effectively fills more incentive gaps in a broader range of markets. Such a network represents a paradigm shift in how commercial markets function.

We define the following core concepts for the Nosh economy.

• Block Reward: A token that is created when a buyer and a producer transact.

• Buyer: A user with an intent to purchase a service.

• Producer: A user with the ability to fulfill a buyer's purchase intent.

• Arbiter: An entity that resolves disputes between a buyer and a producer.

Overcoming Design Constraints

The primary constraint in the design of such a protocol is the inability to systematically verify a wide-range of real-world activity. Real-world activity is subject to the oracle problem, and tools that are available to traditional blockchain projects like consensus (a collective agreement on the state of a blockchain) and validity proofs (evidence of state transition correctness) are inadequate to verify a real-world exchange of goods.

A protocol that pays a financial reward to participants for commencing in an exchange of goods or services that does not verify transactions with robust service-proofs is subject to a number of game-theoretic challenges; including but not limited to: self-dealing, collusion, and malicious actors.

As we explored the design space for real-world service-proofs, we identified a number of possible verification strategies for last-mile delivery networks. Specifically, a combination of KYC, zero-knowledge location-proofs, code exchanges, and random driver assignment together provide a robust proof-of-delivery mechanism for the current state of delivery networks. This double-blind system with random driver assignment ensures that neither the provider nor the customer can confidently predict or influence the matching process. If the provider and customer are known to be unique, cannot systematically predict the assignment of the third party, and all three parties require cooperation to submit a valid service-proof then there is extremely low collusion risk in mature markets.

However, even in the case of mobile food ordering, the majority of all orders are still pick up orders. Pick-up orders and in-store dining are much more difficult to verify. Because restaurants do not sell a commodity, provider assignment cannot be randomized. This makes it easy for a set of two cooperating attackers to collude and earn a block reward without doing the work required to justify the reward (in this case producing the food for the buyer). We could use a similar location-proof mechanism to verify that both parties are in the same region at the time of the transaction, but even if the customer is in the store of the restaurant, it is impossible to have a robust proof-of-work mechanism that verifies with high confidence that the service was performed.

Furthermore, any narrowly defined attempt to codify a verification strategy for a single vertical would impose limitations on the potential scope of the network and potentially be vulnerable to external market movements - for example, the proof-of-delivery mechanism described above could be rendered invalid if robotic delivery became the predominant means of distribution in food delivery networks.

Given these limitations, we adopt a strictly economic approach to verification.

Design Considerations

Generality and Longevity

To maximize the potential of the network and it's supported applications, we opt for a general solution to the DePIN verification problem. A good design should build for the networks of tomorrow. A technically robust but narrowly defined verification strategy, like the proof-of-delivery mechanism described above, would be too opinionated to evolve with future markets and a wider set of emergent applications, limiting the scope and potential of the network.

Security

Security of the network should grow as the network grows. Agents should have increasing assurance that disputed services will result in fair and transparent resolutions. This assurance is rooted in several key mechanisms that improve as the network grows:

• A decentralized insurance AMM
• economic incentives that reward honest behavior
• slashing mechanisms for malicious actors
• a global reputation system that favors participants with good behavior

Integrated Selfishness

Selfish actors amplify the value and security of the network. In order to earn asymmetrically large rewards that are greater than the fees collected by the network, participants must increase their eigenvector centrality ranking. In this paradigm, malicious actors that masquerade as both parties in a two-sided marketplace transaction would exist in an "island", disconnected from the rest of the graph, and therefore have low eigenvector centrality rankings. If they want to increase their rank, and earn large rewards, they must create transactions with the core network. This graph-theoretic approach to token rewards increases competitive dynamics in immature markets and the outcome of selfish actions is a more successful and secure network.

Note: our model requires an honest majority assumption.

Immutability

We minimize governable parameters. A self-regulating reward mechanism automatically adjusts to the unique needs of emerging markets. A credibly neutral, self-optimizing protocol decreases complexity and potential conflicts.

Nosh Core Infrastructure

• Dynamic Transaction Graph: Transactions are represented in a dynamic on-chain graph, where producers and buyers are represented by nodes, and weighted edges capture the activity between peers.

• Multi-Signature Escrow: A multi-signature escrow contract holds funds until both parties attest to service completion. This mechanism ensures that payments are only released upon mutual agreement, reducing the risk of fraud and enhancing trust. Mutual attestations can be achieved through various formats, including randomly generated pins, NFC taps, and RFID.

• Reputation: A game-theoretic approach to rewards ensures that selfish actors maximize the value of the network for all participants. By aligning individual incentives with the overall health of the network, the protocol encourages behaviors that enhance security and robustness, ultimately benefiting all users. Self-interested agents must perform real transactions with real users in order to improve their earning potential and rights to future dividends in the network.

Economic Model

The core economic algorithm of Nosh ensures that demand and supply evolve in tandem, addressing the demand-side acquisition challenges typical in early decentralized infrastructure projects. The protocol automatically self-optimizes for both acquisition and retention and is therefore not temporally biased - rewards will find their optimal distributions in both immature and mature market conditions.

The core utility of the economic infrastructure is to:

• Bootstrap Supply: Our design aims to incentivize producers to join as co-owners and actively contribute to the platform's success by marketing it to buyers and recruiting additional producers.
• Bootstrap Demand: Attracts buyers by rewarding them with tokens for engaging with producers. These incentives help generate demand, ensuring that buyers are motivated to participate actively. Buyer's are not directly subsidized but earn rewards relative to their relationship to the network. Buyer's are encouraged to be loyal to the entire graph rather than a particular producer.
• Increase the Volume of Transactions: Network value is measured as the sum of the connections of buyer and producer peers, encouraging frequent transactions.
• Dispute Resolution and Arbitration: Disputes are handled by a decentralized network of insurance providers, removing the need for a universal arbiter in a centralized network.