Skip to main content

Strategies + Recipes

The generalized location proof protocol is designed as it is to accommodate the wide range of strategies for proving location on the decentralized web, while also provided a standard way of representing these proofs.

In this section of the documentation, we will define these different location proof strategies, and document how to contribute a new strategy to the protocol. (For a clearer idea of our design, see the Architecture page).

note

A contributing guide, along with documentation on specific strategies, is in the works. We've created an unstable example recipe implementation on Github, here. If you want to help us develop this, reach out!

Broad Approaches to Location Proofs

  • Authority-based Strategies: Authorized individuals attest to someone's presence. i.e. ticket checkers, event hosts
  • Social Strategies: People verify each other's claims, confirming or challenging their validity. i.e. permissionless social confirmation, perhaps with sybil-resistant scoring of confirmer accounts; PIN exchange
  • Near-field Machine Strategies: Hardware devices attest to the presence of another device through information exchange or cryptographic 'endorsement' of a claim. i.e. RFID; NFC; Bluetooth
  • Network Machine Strategies: Nodes on a local network or hardware wallets triangulate position and sign confirmation attestations. i.e. Time of Flight; Time Difference of Arrival
  • Sensor Data Strategies: Where devices sense local networks, environmental conditions, images + audio, accelerometers, etc, and location is discerned from analysis of this evidence. i.e. radio frequency / optical / inertial / acoustic / magnetometer localization

In addition to these different categories of strategies, various verification schemes can be employed to endow location proof recipes with a different attributes, such as privacy, game theoretic assurances of validity, etc.

  • Time-based Schemes: Involves waiting or challenge periods.
  • Cryptographic Schemes: Zero-knowledge proofs verified on-chain or in peer nodes.
  • AI-based Schemes: Analyzing photos or contextual information.
  • Economic Schemes: Requires payment or staking value for proof creation.

This is an incomplete list, and we hope to support a growing set of experiments in location proof design.