Authentication¶
Core Idea¶
The pattern of binding an asserted identity or origin to admissible evidence through a verification procedure that outputs a verdict — is this who or what it claims to be? — before trust, access, or weight is granted.
How would you explain it like I'm…
Are You Really You?
The Identity Check
Binding Claim to Evidence
Broad Use¶
- Information security: Passwords, hardware tokens, biometrics, and certificate chains bind a claimed identity to evidence.
- Law and finance: Signature verification, notarisation, and chain-of-custody for evidence.
- Art and collectibles: Provenance research, materials testing, and expert attestation bind a claimed origin to evidence.
- Medicine: Patient identification before a procedure and sample chain-of-custody.
- Journalism: Source authentication and the vouching of anonymous sources.
- Biology: Molecular self/non-self discrimination as a literal authentication mechanism.
Clarity¶
It separates three operations otherwise conflated: authentication (is this who it claims?), authorisation (what may it do?), and verification (does it conform to spec?) — so a failure in one cannot hide behind the others.
Manages Complexity¶
The trust-on-record contract performs the expensive binding once and amortises it over a window, shifting identity cost from per-interaction to per-session, with a verdict calibrated to the stakes.
Abstract Reasoning¶
It enables threat-model reasoning (which impostors does this stop?), trust-anchor reasoning (what would compromise the root?), and freshness reasoning (does it resist replay?).
Knowledge Transfer¶
- Security to medicine: A clinician reads a wristband-plus-verbal-confirmation as a two-channel procedure with a known anchor and freshness window.
- Security to art appraisal: An appraiser reads attestation-plus-testing-plus-provenance as a multi-channel procedure with a mixed anchor.
- Universally: The four-channel taxonomy (known, held, inherent, attested) is a sharp lens for single-channel thinness.
Example¶
A TLS handshake binds a server's claimed domain to evidence (a private key plus a CA-signed certificate), validated up to a root certificate, with a nonce resisting replay and a verdict licensing the session.
Relationships to Other Primes¶
Foundational — no parent edges in the catalog.
Children (2) — more specific cases that build on this
- Capability Separation is a kind of Authentication — capability_separation's cross-ref is attestation and its nearest is traceability (0.892), both of which the file severs. But the deeper load-bearing relation the file draws is to authentication: capability_separation is the issue/verify-asymmetry ARCHITECTURE within which authentication's deliberate verifiable mark operates, and a public-key signature "exhibits both." This is closer to a sibling than a clean is-a. Recording child_of authentication is the medium-conviction read (issue/verify split presupposes a verifiable-mark mechanism); a sibling_of attestation reading is the alternative. Flagging medium because the direction is genuinely arguable -- LEAVE is acceptable if reviewers prefer not to force it.
- Provenance decompose Authentication — The file: provenance is 'one evidence channel a multi-channel authentication procedure weighs' — input to authentication, not the procedure. provenance is a major existing prime (broader: art, data lineage), so this is a part-of/uses, not a reparent.
Not to Be Confused With¶
- Authentication is not Verification because authentication binds an identity or origin to evidence, whereas verification checks an artefact against its specification; conforming content from a forged identity passes one and fails the other.
- Authentication is not Access Control because authentication establishes who a party is, whereas access control decides what an established identity may do.
- Authentication is not Provenance because authentication is the procedure that weighs evidence and renders a verdict, whereas provenance is the recorded chain of origin that serves as one evidence channel.