Skip to content

Records-First Authenticity And Digital Originality Design Note

Status

Draft conceptual design note, 4 September 2026.

Purpose

This note explains why OpenETR uses a records-first approach to authenticity and digital originality. It places that decision in historical context while preserving a clear boundary between historical analogy, technical evidence, institutional recognition, and legal effect.

The central thesis is:

A Digital Original is not one privileged copy of a file. It is an exact Digital Artifact whose consequential state can be independently derived from authentic records.

The shorthand is:

Copies are plural; consequential state is singular.

This is not a claim that every legal regime must use the term Digital Original or give an OpenETR record a particular effect. OpenETR supplies independently verifiable evidence and defined state-transition rules. Recognition regimes decide which actors, records, and derived states they accept and what follows.

Design Decision

OpenETR begins with the record rather than requiring a resolved institutional identity as the gateway to the record:

exact Digital Artifact
  -> attributable signed events concerning it
  -> Evidence Graph
  -> state-transition rules
  -> Consequential State
  -> recognition for a stated purpose
  -> external effect

The minimum identity fact required by the protocol is the public key that signed an event. A recognition system may subsequently determine that the key represents a particular person, organization, warehouse, authority, service, device, or software agent.

An identity-first architecture commonly begins by resolving a subject and its credentials, keys, roles, or service endpoints before evaluating a statement. That is useful where identity assurance is the primary objective. OpenETR has a different primary objective: preserving independently verifiable records of consequential actions concerning an exact digital artifact.

Identity is therefore important but not foundational in the same way:

  • signatures establish cryptographic attribution;
  • profiles and external systems associate operational meaning with keys;
  • registries, credentials, attestations, and trust frameworks support recognition; and
  • law, policy, contract, and institutional rules determine effect.

What-First Interoperability

The records-first decision also establishes a what-first approach to interoperability. A verifier begins with the thing being presented and asks:

What exact Digital Artifact is this?
What signed Evidence Graph records concerns it?
What Consequential State follows under the identified rules?
What additional evidence does the proposed use require?

An identity-first or delegation-first exchange may instead require the relying party to understand a chain of subjects, roles, relationships, permissions, and delegated authority before it can evaluate the transaction. Such machinery may be necessary inside the organization that authorizes an action, but it does not necessarily need to cross the system boundary with the resulting record.

OpenETR therefore adopts this interoperability principle:

Do not require the upstream authority model to be interoperable when its consequential result can be made independently verifiable.

This principle does not eliminate identity, authorization, or delegation. It keeps those concerns where they are established and brings them into verification only when the applicable policy requires them. A host system may use accounts, passkeys, organizational directories, approval chains, or agent delegations to authorize use of a signing key. OpenETR preserves the resulting signed evidence concerning the artifact. Another verifier evaluates that evidence and introduces only the recognition inputs needed for its purpose.

Portable Consequences, Precisely Stated

The shorthand make consequences portable requires a precise technical meaning. OpenETR does not copy an authoritative state value between systems. It makes portable the inputs required to reproduce a consequential result:

digest-identified artifact
  + signed Evidence Graph records
  + identified rules and evaluation parameters
  -> reproducible Consequential State

Two conforming verifiers using the same sufficient evidence, ruleset, and parameters should derive the same Consequential State. They may still reach different conclusions about recognition or external effect because those conclusions depend on their respective legal, institutional, commercial, or operational contexts.

The resulting boundary is:

portable: artifact identity, signed evidence, graph relationships, rules basis
derived:  Consequential State
local:    recognition and effect

This avoids making a private database, API, organizational directory, or delegation service the only place from which a consequential answer can be known. The upstream bureaucracy can remain upstream; the independently verifiable record can travel downstream.

The "Cult Of The Original"

The phrase "cult of the original" describes a modern tendency to locate authenticity and authority in a singular artifact with a privileged physical history. Walter Benjamin observed that a traditional original possesses a unique presence in time and place, while technical reproduction substitutes a plurality of copies for that unique existence. His analysis is concerned with works of art, but the distinction helps expose a problem with digital records: perfect reproduction is an ordinary property of the medium, not an exceptional act.

Historical scholarship also shows that authenticity has not always depended exclusively on modern forensic ideas of singular authorship, physical origin, and an unbroken material object. In some medieval settings, authority could be relational, institutional, customary, devotional, or functional. Copies and reconstructions could participate in an accepted tradition even where modern historical methods would question their material origin.

OpenETR does not adopt a medieval proposition that accepted utility, belief, or spiritual truth may substitute for factual provenance. Nor does it treat a plausible reconstruction as equivalent to authentic evidence. The comparison identifies one narrower and useful insight:

The authority associated with a record need not reside exclusively in one privileged physical embodiment.

OpenETR combines that relational insight with modern cryptographic exactness. It can therefore be understood as a model of cryptographically rigorous relational authenticity.

Relocating Historical Testimony

For a physical artifact, changes in condition, custody, and ownership may be carried by the object's unique material history. A byte-for-byte digital copy does not carry a visibly distinct patina or naturally privileged location.

OpenETR relocates the relevant historical testimony from a particular storage instance into an Evidence Graph: one signed event or a graph of signed and linked events concerning the artifact.

The model separates four things that physical practice often combines:

Concern OpenETR treatment
Exact content A cryptographic digest identifies the Digital Artifact.
Attributable history Signed events identify the keys making assertions or taking actions.
Consequential history Event links and transition rules permit a verifier to derive state.
Accepted authority and effect A recognition context decides which actors, graph, and state it accepts for a purpose.

The Evidence Graph does not create a new physical original. It preserves the evidence from which the artifact's consequential history can be reconstructed and evaluated without relying exclusively on the application that first displayed or stored it.

Originality Without A Privileged Copy

Byte-for-byte copies having the same digest represent the same Digital Artifact. Moving, backing up, transmitting, or rendering those bytes does not create a second artifact identity and does not independently duplicate its control state.

Consequential state attaches to the digest-addressed artifact through its validated Evidence Graph. It does not attach separately to every file-system instance, download, attachment, or storage location.

OpenETR therefore separates:

  • originality from uniqueness of physical embodiment;
  • artifact identity from storage location;
  • authenticity from recognition;
  • cryptographic attribution from legal identity; and
  • protocol-derived state from external effect.

A Digital Original is consequentially singular even when its exact bytes are widely reproduced. A relying party can obtain any exact copy, identify it by digest, retrieve or receive the relevant signed evidence, and independently evaluate the candidate state.

A Records-First Account Of Forgery

Under this model, copying the bytes of an artifact is not itself forgery. A perfect copy remains an instance of the same digest-addressed Digital Artifact. The relevant failures become more precise:

Conduct or defect Result
Altering the artifact bytes Produces a different digest and therefore a different Digital Artifact.
Claiming that another key signed an event Fails signature verification.
Signing an action without recognized authority Produces cryptographically attributable evidence that may be invalid under transition policy or denied recognition and effect.
Publishing an unlinked or malformed transition Produces evidence that does not validly continue the candidate control graph.
Constructing a conflicting history Produces competing branches or claims that a verifier must report and a recognition policy must evaluate.
Misrepresenting the real-world subject behind a key Creates an identity or recognition failure, even if the signature itself is valid.
Omitting relevant evidence Creates an evidence-completeness and retrieval-coverage problem rather than proving that the omitted event does not exist.

The operative question changes from:

Is this the one authentic physical copy?

to:

What exact artifact is this, what authentic signed records concern it, what consequential state follows under the applicable rules, and under which recognition context is that state accepted?

OpenETR verifiers should report those dimensions separately. A valid signature does not prove authority. A valid transition does not compel recognition. An apparently complete graph does not prove universal completeness. Each proof proves only what it proves.

Design Consequences

The records-first decision has the following consequences for OpenETR:

  1. The artifact is content-addressed. The digest, not a platform record id or storage location, establishes the artifact identity used by the control layer.
  2. The history is evidence-addressed. Signed event ids and explicit links preserve inspectable evidence of assertions and state transitions.
  3. Consequential state is derived. No application's mutable status field is the exclusive source of truth for the control state.
  4. Copies remain useful. Any exact copy can be checked against the same digest and Evidence Graph records.
  5. Identity stays composable. DIDs, credentials, KYC results, registries, Nostr Web of Trust attestations, TRQP responses, and host-system accounts may support recognition without becoming mandatory dependencies of the base record protocol.
  6. Recognition remains contextual. Different relying parties may apply different rule books to the same authentic evidence and may reach different conclusions about recognition or effect.
  7. Infrastructure is non-exclusive. Signed records may be obtained from public relays, private relays, archives, evidence packages, or local stores and verified without the originating application running.

Scope And Limits Of The Historical Analogy

The historical comparison is explanatory, not normative. It does not establish legal equivalence between medieval documentary practice and a modern electronic transferable record. It does not relax OpenETR's cryptographic or state-transition requirements. It does not imply that social acceptance can repair altered bytes, invalid signatures, or broken graph links.

The analogy helps distinguish two questions:

  • Must authenticity depend on one unique material carrier?
  • Can authority and effect instead arise from verifiable relationships, attributable actions, and a recognized context?

OpenETR answers the first question no and makes the second question explicit. Cryptography supplies evidence of exact content and signed actions; an identified ruleset derives candidate Consequential State; external institutions determine recognition and effect.

Working Maxims

  • Copies are plural; consequential state is singular.
  • A Digital Original is consequentially original, not physically uncopyable.
  • The digest identifies the artifact; the Evidence Graph records what happened to it.
  • The control graph carries consequential history without becoming a claim of universal authority.
  • Identity informs recognition; it does not exclusively own the record.
  • Cryptography establishes attributable evidence, not legal effect.
  • Each proof proves only what it proves.

Sources And Further Reading

Historical sources provide context rather than normative requirements:

Related OpenETR documents: