Skip to content

Chapter Two: Making Claims (And Protecting Them)

A XID is just an identifier: a pseudonymous label that allows for consistent reference over time. Keys and a provenance mark support it, but they're ultimately infrastructure: the keys control the identifier, and the provenenance mark supports updates.

For an identifier to truly become an identity requires more: it requires a rich collection of data that define and describe the real person behind the identifier (or at least the persona that they embody when they interact using the identifier).

A variety of content will be added to an identifier over this course, and the first of those will be self-attestations: claims made by the owner of the identifier themself. However, these first, simple claims also offer the first danger of a breach in the cloak of pseudonymity that a XID offers, so the addition of claims must go hand in hand with the discussion of how to properly protect them.

Major Objectives for this Chapter

After working through this chapter, a developer will be able to:

  • Register additional keys in their XID.
  • Create attestations that are publicly verifiable.
  • Advance provenance marks.
  • Commit to claims.
  • Encrypt claims.

Supporting objectives include the ability to:

  • Understand the fair witness methodology for making credible claims.
  • Recognize correlation risks.
  • Know the difference between a variety of types of attestations.
  • Choose between a variety of methods for handling sensitive information

Table of Contents