Semantic integrity
Claims, relationships and representations do not undermine one another without a visible explanation.
03 · ODI for knowledge infrastructure
Digital identity as an object of inquiry.
Ontology of Digital Identity is a distinct theoretical, terminological, research and IP layer. It examines the conditions under which digital identity becomes recognizable, stable, relationally grounded, interpretable and reconstructable.
A distinct layer
ODI is not a footnote to DIO.
DIO engages a specific identity with an optimization objective. ODI examines the conditions under which digital identity exists, can be known and can break down—without requiring immediate application.
The two layers share concepts and the K6 model, but answer different questions. DIO asks how to intervene. ODI asks what must first be defined before identity, intervention and the resulting reconstruction can be discussed coherently.
Epistemic restraint
Not “objective truth,” but reliable and auditable reconstruction.
ODI treats digital identity as a distributed, emergent object. Different observers work from different traces and rules; the goal is therefore not a metaphysically final description, but transparent conditions for compatible interpretations.
Provenance
It is possible to trace who created a claim, what supports it and the context in which it applies.
Confidence
The distinction among data, interpretation and inference remains visible.
Identification consistency
Different systems can arrive at compatible definitions of an entity and its relationships.
K6 as a shared language
Six dimensions of identity’s relational field.
The dimensions do not form a sequence. Any change in representation can alter interpretation, trust, relationships and the resulting reconstruction—while also redefining how the entity itself is bounded.
Entity
What is stabilized as a distinguishable interpretive node, and where its boundaries lie.
Representation
The traces, signs, data and statements through which the entity becomes present to each observer.
Interpretation
The meaning produced from representations within a specific mode of reading and decision-making.
Trust
How reliability, authority, evidential support and resilience to contradiction are established.
Relationship
How connections to other entities, works, roles and sources anchor meaning.
Reconstruction
How the whole is reassembled from available traces, and with what loss or uncertainty.
K6 is the operational and conceptual framework of DIO/ODI. Its universal empirical validity remains open to external evaluation.
Research agenda
Questions that can be refined, operationalized and falsified.
ODI gains value not by expanding abstraction, but by translating concepts into research problems, metrics, domain adaptations and open expert critique.
Identity
Under what conditions do distributed traces stabilize as one entity—and when do they split into multiple nodes?
Modes of reading
When are human, search, graph and model-based reconstructions compatible—and when are they not?
Time
How can an identity update be distinguished from a discontinuity, conflation or illegitimate overwrite?
Evidence
How should provenance and confidence affect the weight of a claim, relationship or derived interpretation?
Intervention
How can we tell that a change in representation improved reconstruction rather than merely hiding an inconvenient contradiction?
Existing reference infrastructure
ODI already has a versioned conceptual and machine-readable foundation.
Reference Vocabulary v0.1.0, Reconstruction Failure Modes and Reference Model v0.3.1 connect the publication layer with a canonical namespace and open data formats.
Reference Vocabulary
A controlled vocabulary of concepts, definitions and relationships with persistent identifiers.
Reconstruction Failure Modes
Four failure families distinguish the referential, semantic, temporal and contextual levels of reconstruction.
Versioned reference model
The published release connects concepts, model relationships, documentation and distributed artifacts.
Stable concept identifiers
The namespace https://vocab.danielberanek.cz/dio-odi/# gives the terms an addressable and linkable form.
SKOS, Turtle and JSON-LD
Machine-readable serializations make the reference layer available to ontologies, graphs and other knowledge systems.
RAG × DIO and HEM × DIO
Application Mappings connect DIO/ODI concepts with a RAG pipeline and the Heterogeneous Entity Matching taxonomy while remaining separate from the canonical vocabulary.
From a reference foundation to domain applications
Follow-on work can develop metrics and protocols, domain mappings, research projects, standardization and specifically defined licensed uses.
Canonical core and derived layers
The reference vocabulary preserves stable meanings; application profiles can translate them into specific systems without being conflated with the canonical layer.
A suitable partnership
Critical, interdisciplinary and specifically bounded.
ODI is not looking merely for agreement. It seeks environments where its concepts can be refined, challenged and translated into usable structures.
- Critical expert review
- Formalization and operationalization of concepts
- Research and grant partnerships
- Domain adaptation
- Licensing or standards collaboration
ODI for knowledge infrastructure
Bring a question, discipline or system in which identity needs a more precise account.
The proposal should define the research or institutional problem, the available capability and the expected form of the result.