representation
Enable an AI agent to recognise a visual or tangible representation, assess what it renders and how reliably, and decide how it may be interpreted, displayed, compared or altered.
Bundle → Layer → Finding → Questions Filled
6 bundles · 11 layers · 18 findings · 28 questions
Subject and rendering role Establish what is being rendered and why this artifact is treated as a representation.
Referent identification
Identify the subject at the specificity justified by available evidence.
Supported subject identification
Record whether the subject is a particular entity, a class, an imagined entity or unresolved, preserving disputed identifications.
- Who or what does the artifact render, and at what level of specificity can that subject be identified? definition
- Which inscription, creator statement, contextual record or perceptible feature supports each proposed identification? provenance
Representational status
Establish the rendering role and separate it from other roles the artifact may have.
Basis for rendering role
Record the evidence that the artifact functions as a rendering, including cases where functional, decorative or symbolic roles coexist.
- What establishes a rendering relationship rather than an accidental resemblance or an unsupported viewer association? boundary
- Does the artifact also function as a tool, replica, symbol or decoration, and which features specifically support its representational role? definition
Perceptible rendering Describe how the subject becomes available through visible or tangible form.
Rendering features
Identify the surfaces, marks, shapes and spatial arrangements that carry subject information.
Subject-bearing elements
Distinguish elements that render the subject from supports, frames, captions and incidental material features.
- Which contours, colours, textures, volumes or spatial relationships make the subject recognisable? definition
- Which parts render the subject directly, and which merely support, frame or explain that rendering? boundary
Perceptual access
Identify the conditions under which the rendering can be perceived and inspected.
Access-dependent legibility
Record how viewing position, lighting, distance, tactile access or obstruction affect access to subject information.
- Under which viewing or touch conditions are the features needed for recognition accessible? measurement
- What inspection or presentation change would reveal currently inaccessible subject features without damaging the artifact? action
Correspondence and transformation Make explicit what the rendering preserves, changes and leaves out.
Feature correspondence
Relate rendered features to subject features through an explicit comparison basis.
Comparison basis and scale
Record the reference evidence, relevant subject state and any scale or spatial mapping used to assess correspondence.
- Against which observation, reference artifact or documented subject state can the rendered features be compared? provenance
- Where is scale or spatial correspondence established, and where do perspective, deformation or missing dimensions prevent measurement? measurement
Selection and alteration
Identify departures from the subject and the limits introduced by selection.
Rendering transformations
Record supported instances of omission, emphasis, stylisation, compositing or reconstruction without assuming the creator's motive.
- Which subject features are omitted, exaggerated, simplified, combined or reconstructed, and what evidence supports that assessment? provenance
- Which conclusions about the subject become unsupported because of those transformations? boundary
Interpretation and use Connect interpretations and proposed uses to the evidence the rendering actually supplies.
Interpretive context
Separate observable content, attributed intention and audience interpretation.
Qualified interpretation
Record how labels, conventions and audience knowledge influence interpretation, with attribution for claims about intended meaning.
- Which interpretations follow from perceptible features, and which depend on a caption, convention or contextual account? definition
- Who attributes an intended meaning to the representation, and what evidence supports that attribution? provenance
Use adequacy
Assess a concrete proposed use against the representation's supported content and limitations.
Supported use and qualifications
Record which aspects of a proposed use are supported and what contextual or external constraints govern proceeding.
- For the proposed use, which required subject features are supported, missing, ambiguous or transformed? boundary
- What qualifications, additional verification or referenced permission constraints must the agent address before using or displaying this representation? action
Representational change Track changes that affect the readable subject, apparent correspondence or attributed meaning.
Meaning-bearing condition
Assess condition in terms of the subject information that survives or has changed.
Loss or change of subject information
Record which representational features are obscured, missing or altered and distinguish observed condition from inferred original appearance.
- Which identifying or interpretively important features are currently missing, obscured or altered? measurement
- What prior image, examination or treatment record establishes that these features changed rather than being absent from the original rendering? provenance
Intervention and derivation
Assess how proposed modifications or derived renderings affect representational continuity.
Effect of proposed intervention
Identify whether annotation, restoration, cropping, recolouring or reproduction preserves, changes or speculatively adds subject information.
- What subject information would the proposed intervention preserve, remove, alter or add without direct evidence? action
- Would the result remain a changed instance of this artifact or require recognition as a distinct derived representation, and what criterion supports that decision? boundary
Evidence and external alignment What the world already says about this thing, gathered so the model can be checked against it.
Reported evidence
Findings from the breadth pass, kept separate from the structural claims.
Kinds and varieties
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these kinds and varieties hold for the sense of representation this model covers, and on what evidence? provenance
Identifiers and schemes
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these identifiers and schemes hold for the sense of representation this model covers, and on what evidence? provenance
Standards and regulation
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these standards and regulation hold for the sense of representation this model covers, and on what evidence? provenance
Real-world use
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these real-world use hold for the sense of representation this model covers, and on what evidence? provenance
Typical measurements
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these typical measurements hold for the sense of representation this model covers, and on what evidence? provenance
Failure modes and hazards
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these failure modes and hazards hold for the sense of representation this model covers, and on what evidence? provenance
Regional variation
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these regional variation hold for the sense of representation this model covers, and on what evidence? provenance
Neighbouring kinds and how to tell them apart
Reported by the breadth pass; each item needs checking against its source before it becomes normative.
- Which of these neighbouring kinds and how to tell them apart hold for the sense of representation this model covers, and on what evidence? provenance
Classifiers Filled
- Family
- Thing Registry
- Category
- Physical world and living systems
- Entry kind
- noun.artifact
- Plane
- PHY
- Domain
- PHY.OBJ
- Tags
- artifact
- Facets: substance
- material
- Facets: origin
- manufactured
- Facets: agency
- inert
- Facets: mobility
- portable
- Facets: scale
- handheld
- Facets: affordances
- graspable
- Other names and narrower kinds
- adumbration, anaglyph, angiocardiogram, angiogram, arteriogram, arthrogram, audiogram, autoradiograph, backdrop, background, backcloth, beefcake
What it is Filled
A human-made object or image produced so that its optical or spatial form depicts a recognisable person, place, thing or event, as distinct from that referent and from the shared visual prototype that copies, casts and digital surrogates may carry.
In scope
- Identification of the artifact and the someone or something it renders, including imagined or unidentified subjects
- Visual and tangible features through which the subject is represented
- Selection, abstraction, scale, viewpoint and other transformations of the subject
- Evidence for attribution, intended meaning and the relationship to a referent
- Legibility, misleading appearances and suitability for a proposed representational use
- Changes to the artifact that preserve or alter what it represents
Out of scope
- The represented person's identity, biography or current condition
- The represented object's own composition, operation or lifecycle
- The full production workflow, tools and materials supply chain
- General ownership, licensing and contractual records, which this model only references when they constrain use
- Purely functional objects or decorative patterns with no established representational role
- Abstract propositions and symbolic systems considered independently of a visual or tangible rendering
Why it exists Filled
Enable an AI agent to recognise a visual or tangible representation, assess what it renders and how reliably, and decide how it may be interpreted, displayed, compared or altered.
Distinguishing features Filled
- An identifiable created artifact renders someone or something; merely being visible or tangible does not establish representation.
- The artifact can be distinguished from its subject: a portrait is a rendering of a person, while the person is its referent.
- A representational relationship is supported by perceptible correspondence, an established convention, contextual evidence or an attributed intention; resemblance alone may be inconclusive.
- A rendering may selectively distort, simplify or invent features and still represent a subject; exact duplication is not required.
- A functional replica, symbol or decorative object qualifies only insofar as a rendering role can be established alongside its other roles.
What robots and AI may and may not do Filled
Must not
- Present a depiction as evidence that an event happened as shown.
- Create or alter depictions of real people to deceive or demean them.
- Remove or damage representations in collections without authority.
- Misattribute authorship of a work.
Only with a human decision
- Displaying depictions of sacred subjects or the dead in contexts their communities object to.
May
- Describe what a representation depicts and who made it, citing provenance.
- Distinguish the object from its subject and from copies.
Moral aspects Filled
- Images shape how people and events are remembered.
- Depictions of people raise questions of consent and dignity.
Who is affected
- People depicted
- Makers and owners
- Communities with sacred images
Owners Filled
Steward
The owner of the object; makers hold rights in the work.
Master systems
- Museum and collection catalogues
Links to other meta-models Filled
parent
- oewn-03133774-n - registry parent class
What else AI and robots need to interact with it Filled
Identity and identifiers required Filled
- Vercy registry: vr.tr.representation-artifact
- Open English WordNet: oewn-04083958-n
Direct properties required Filled
- Referent resolution: identified individual; identified class; imagined subject; disputed; unresolved
- Rendering modality: visual; tangible; both
- Representational role evidence: Links to creator statements, labels, commissions, conventions or documented interpretations
- Subject correspondence: Feature-level mappings between rendered elements and subject features, with evidence and limitations
- Representational transformation: Examples include cropping, simplification, exaggeration, idealisation, compositing and reconstruction; multiple values allowed
- Declared linear scale: Dimensionless artifact-to-subject length ratio, qualified by feature or region; variable, unknown or not applicable
- Representational legibility: assessable; partially obscured; unassessable; untested, qualified by audience and viewing or touch conditions
- Use-specific adequacy: Assessment linking a proposed use to supported features, missing information and required qualifications
Recognition required Filled
- A made object or image whose form shows a recognisable person, place or thing.
Capabilities and actions required Filled
- Identify or propose the rendered subject while retaining competing interpretations and their evidence
- Inspect visual or tangible features and relate them to specific features of the subject
- Compare representations of the same subject by viewpoint, selection, scale and transformation
- Assess whether a proposed identification, measurement or explanation is supported by the rendering
- Display or describe the representation with qualifications that prevent unsupported inferences
- Propose annotation, restoration or adaptation and identify how it would change the represented meaning
Hazards and failure modes required Filled
- Inaccurate or selectively framed depictions can mislead people or agents about the represented subject.
- Confusing a representation with its referent can produce incorrect spatial, identity or operational decisions.
- Unauthorised copying or disclosure can infringe rights or expose private information.
- Fragile or unstable physical representations can break, fall or damage nearby property.
Standards and interfaces required Filled
- The Copyright Act of 1976, enacted by the United States Congress, governs copyright protection and permitted uses of qualifying original pictorial, graphic and sculptural works in the United States.
Context of use required Filled
- Representations appear in artworks, illustrations, models and exhibits that depict people, places, objects or events.
- They support communication, teaching, planning and interpretation when the depicted subject is absent or needs selective emphasis.
Sources Filled
- E36 Visual Item (CIDOC CRM, including P138 represents / P62 depicts) - Separates the physical carrier, the immaterial visual item, and the depicted entity; defines representation as a visual relation with an optional mode.
- Visual Item 3D (CIDOC CRM Issue 522) - Extends visual items to shared surface shape of sculptures (Rodin Le Penseur: many casts of different size as one visual item).
- depictions (visual works), AAT 300248231 - Getty definition of depictions as images recognisable as representations of visible entities, with Dutch/German/Spanish equivalents.
- Categories for the Description of Works of Art (CDWA) - Museum cataloguing of depicted subject, related visual documentation, and image measurements as distinct from the work itself.
- Depicted Subject, Iconography Authority (Getty Vocabulary Editorial Guidelines) - Practice of indexing what a work depicts versus what it is about; ofness versus aboutness and isness.
- Wikidata:WikiProject Visual arts/Item structure - How individual representations are typed (painting, sculpture, print) and linked to subjects via depicts (P180) and Iconclass (P1257).
- Commons:Structured data/Modeling/Depiction - Distinguishes a digital file, a faithful digital representation of a 2D work (P6243), and the depicted artwork.
- replica (Q1232589) - Replica as an object created as a copy; AAT 300015642 scope note on same-dimension reproductions versus facsimiles and versions.
- Compendium of U.S. Copyright Office Practices, Chapter 900 (Visual Art Works) - Legal class of pictorial, graphic and sculptural works, and the hazard of registering a photograph of a sculpture as if it were the sculpture.
- Adapting Aspects of FRBR for Artworks in Wikidata - How museums treat print and sculpture editions as related instantiations of one work rather than wholly unique objects.
- Open English WordNet 2024, synset oewn-04083958-n - sense and gloss of the concept
- Wiktionary: representation - word senses and usage
Open questions
- Does the PHY / PHY.OBJ registry entry include digitally displayed, projected or transient renderings, or only artifacts with a persistent physical embodiment?
- What minimum evidence establishes representational status for found objects, natural casts, ambiguous abstractions and resemblance adopted after creation?
- Where do existing Vercy models place overlaps with photographs, diagrams, maps, symbols, replicas and scale models, and which model should own their shared representational facts?
- What evidence and audience-specific criteria should support legibility and use-adequacy assessments without implying universal recognition or fidelity scores?
- Which identity rules distinguish restoration of one representation from production of a derivative, particularly when missing subject features are reconstructed?
Machine files
Provenance
thing registry research (pass 2) · unreviewed
Built from: models/things/publications/thing-representation-artifact/spec.json