Data minimization is the practice of limiting the data you share to only what’s necessary for a specific purpose. It follows the principle: “Share what you must, protect what you can.”
Beyond Anonymity and Pseudonymity
Pseudonymous Trust Building is another Identity Design Principle, but neither it nor a fully anonymous digital is sufficient in itself. That’s because anonomized data can be de-anonymized, while pseudonyms, which accumulate data histories over time can eventually be linked to real identities, especially if they contain coontextual information (and if they reveal it unncessarily).
Data minimization addresses these limitations by reducing all data shared to the minimum needed for each specific interaction.
Elision as a Data Minimization Tool
Though zero-knowledge proofs and others sorts of more complex, less-tried cryptography can be used to ensure data minimization, Blockchain Commons suggests a simpler methodology, Elision Cryptography.
This methodology is built into Gordian Envelope, which allows the selective removal of specific pieces of information while maintaining the cryptographic integrity of the whole.
Data Maximalism Risks
Data minimization is intended to resolve a number of human rights risks with data sharing. It supports:
- Privacy: The right to control what personal information is shared and with whom.
- Autonomy: The ability to make choices without undue influence based on profiled data.
- Non-discrimination: Protection from judgments made on irrelevant personal data.
- Security: Reduced attack surface for identity theft and other harms.
Privacy Risks of Excessive Data Sharing
Every piece of information shared increases potential privacy risks:
- Correlation: When data from different sources is combined, it can reveal far more than intended. Even seemingly harmless details can complete a revealing puzzle about a person.
- Secondary Use: Once data is shared, it may be repurposed beyond its original intent, potentially in ways that harm the subject’s interests.
- Disclosure Risks: Sharing excessive data can create prejudice or disadvantage, particularly for marginalized individuals or communities.
- Digital Permanence: Unlike conversations that fade from memory, digital data can persist indefinitely and be copied without limit.