notes: research/papers/papers-revocation

This data as json

path title type status classification words body
research/papers/papers-revocation Revocation summary v0.1 — 2026-10-15: drafted by Claude from the markdown file in the Papers repository; unread by Mark; marked partner on Mark's word 2026-10-21, paraphrase only partner 566 # Revocation **Abstract.** A white paper (version 0.4, 13 June 2026) exploring the design trade-offs for dynamic revocation of verifiably issued digital entitlements in SEDI (State Endorsed Digital Identity), a citizen-sourced, state-endorsed identity system. It compares proofing mechanisms (bare attached signatures versus anchors in verifiable data structures), frames the goals of citizen control, control over value and privacy, and proposes bulk issuance with per-issuance registries. For policy makers and architects of government digital identity. ## Summary The overview defines SEDI as a system in which citizens create and control identifiers derived from key pairs, bring them to the State for identity assurance, and receive a verifiable endorsement, the core SEDI entitlement, to which other State issuances such as a driver's license or proof of age can be linked. It defines entitlement, credential, endorsement, issuance and revocation, and distinguishes dynamic revocation (before expiration) from expiry, with suspension treated as a subset. The section on revocability ties the need for revocation to the lifetime of an issuance, noting that short-lived, automatically reissued instances still need an overt act of revocation when the underlying entitlement is long-lived. The proofing-mechanism section classifies authentication into shared secrets, digital signatures, and verifiable data structures built from signatures. It argues that bare attached signatures are ephemeral because keys must be rotated, which forces reissuance at scale or centralized "phone home" verification, and that anchoring issuances in an append-only, hash-chained key event log gives perpetually verifiable proofs. Applied to revocation registries, it contrasts a signed bitstring status list with ACDC (Authentic Chained Data Container) registries: one independent registry per issuance, each a transaction event log (TEL) anchored in the issuer's key event log (KEL) and served by a Registrar. The goals section puts citizen control first, meaning portability across open infrastructure, strong security, persistence and recovery; then control over value through a duty of data loyalty and accountability; then privacy, discussed through the PAC (privacy, authenticity, confidentiality) trilemma and the argument that super-correlators defeat technical anonymity, so legal recourse matters more. The final sections argue that unlinkability is not uncorrelatability, that sparse Merkle trees can provide range proofs, and that bulk issuance, a set of mutually exclusive identifier sets per citizen with registries updated in bulk for herd privacy, supports enforcement of data loyalty through chain-link confidentiality. ## Key terms SEDI, dynamic revocation, entitlement, cryptoperiod, bare attached signature, anchored issuance, revocation registry, transaction event log, Registrar, PAC trilemma, bulk issuance, duty of data loyalty ## Claims and their evidence - Bare attached digital signatures are ephemeral proofs because key rotation invalidates what they signed, which creates a reissuance-at-scale problem. [informative] - Issuances anchored in a key event log remain verifiable beyond the cryptoperiod of the signing keys. [informative] - Cryptographic accumulators as unlinkable revocation registries have proven impractical for performance reasons. [unverified] - Unlinkability largely amounts to "privacy theater" because any real-world exchange leaks correlatable information (section "Tradespace of Proofing mechanisms"). [informative] - Verifiable credential systems that ignore signature ephemerality are fundamentally flawed from a security perspective. [informative] ## Related ACDC specification (Trust over IP), KERI specification, ACDC with Static Age Predicate Proof SMT, ACDC Affordances for Confidential (Sensitive) Data Protection and Management, ACDC IPEX Negotiation. ## Provenance Drafted by Claude on 2026-10-15 from the markdown file `whitepapers/Revocation.md` (V0.4 2026/06/13), read in full. The document gives no author line; it is attributed to Samuel M. Smith by the repository. Not yet read by Mark.

Links from other tables