Part of the WebMEM® Protocol
Fragment Class: IdentityFragment
Location: /protocol/fragments/identityfragment/
Status: Current Draft
Last Updated: 2026-08-24
Overview
An IdentityFragment is a WebMEM fragment class for publishing the canonical, structured identity of an entity.
Each IdentityFragment establishes what the entity is, the identifiers by which it can be distinguished, its canonical name or label, its entity type, and the roles, affiliations, classifications, domains, or other structural attributes necessary to resolve that identity consistently.
Unlike a PersonaFragment, which represents a contextual human model, archetype, or scenario, an IdentityFragment represents the identity of an actual entity or explicitly defined conceptual entity.
IdentityFragments provide stable publisher-declared identity anchors that allow other WebMEM fragments, Machine Fact Objects, relationships, indexes, and resolver structures to refer consistently to the same entity.
Semantic Purpose
The semantic purpose of an IdentityFragment is to state:
This is the entity, these are the identifiers and structural attributes by which it is known, and these relationships establish its identity within the publisher’s knowledge structure.
An IdentityFragment may establish:
- the canonical identity of the entity;
- the canonical name or label of the entity;
- the entity type;
- one or more stable identifiers;
- canonical resources associated with the entity;
- primary and secondary roles;
- domain or operational scope;
- organizational affiliations;
- governance or institutional relationships;
- current and historical classifications;
- superseded or retired identity relationships;
- version information where identity declarations change over time;
- and provenance supporting the identity declaration.
The IdentityFragment preserves these elements as one coherent identity knowledge object.
IdentityFragment Within an SDT
An IdentityFragment exists as one modular knowledge object within a Semantic Data Template (SDT).
The relationship is:
Web Resource → SDT → IdentityFragment → Canonical Entity Identity
An SDT may contain one IdentityFragment or multiple IdentityFragments when a resource represents or depends on multiple identifiable entities.
Other fragments within the SDT may reference the entity established by an IdentityFragment rather than independently redefining that entity each time it appears.
This allows identity to remain stable while DataFragments, IndexFragments, DirectoryFragments, PolicyFragments, and other knowledge objects describe different aspects of the same entity.
IdentityFragment Requirements
A conforming IdentityFragment should establish enough information to distinguish the entity from other entities and connect that identity to the knowledge objects that describe or reference it.
| Element | Purpose |
|---|---|
| Fragment Class | Identifies the knowledge object as an IdentityFragment. |
| Fragment ID | Provides a stable identifier for the IdentityFragment within the published knowledge structure. |
| Entity ID | Provides a stable identifier for the entity represented by the fragment. |
| Entity Name | Provides the canonical human-readable name or label of the entity. |
| Entity Type | Identifies the structural type or class of entity being represented. |
| Canonical Resource | Identifies the canonical resource associated with the entity when one exists. |
| Structural Attributes | Identifies roles, affiliations, domains, classifications, or other attributes necessary to establish the entity’s structural identity when applicable. |
| Provenance | Identifies or references the sources and lineage supporting the identity declaration. |
Additional identifiers, aliases, roles, affiliations, historical identities, version information, and relationships may be included when required to represent the entity accurately.
Entity Identity and Fragment Identity
The identity of an IdentityFragment is distinct from the identity of the entity represented by that fragment.
For example:
- Fragment ID: identifies the particular IdentityFragment published within the SDT.
- Entity ID: identifies the entity represented by that fragment.
Multiple publication instances may describe the same entity while preserving the same entity identity.
This distinction allows entity identity to remain stable independently of a particular page, fragment, serialization, or publication instance.
Canonical Identity
An IdentityFragment should identify the entity as precisely as the publisher’s knowledge permits.
Canonical identity may be established through:
- a publisher-defined stable identifier;
- an authoritative external identifier;
- a contract or product identifier;
- an organization identifier;
- a geographic identifier;
- a canonical URI or web resource;
- or another identifier appropriate to the entity type.
An entity may have multiple identifiers when different systems or authorities identify the same entity differently.
When multiple identifiers are published, the IdentityFragment should make their relationship to the canonical entity explicit rather than treating each identifier as a separate entity.
Entity Type
An IdentityFragment should identify the structural type of entity it represents when that type is necessary to interpret the identity correctly.
Examples may include:
- Person;
- Organization;
- Dataset;
- Product;
- Plan;
- Contract;
- Provider;
- Program;
- Policy;
- Geographic Entity;
- Defined Term;
- or another entity class defined by the publisher or applicable domain.
Entity type should describe what the entity is rather than what a particular page says about it.
Roles and Structural Classification
IdentityFragments may represent roles or structural classifications when those attributes are necessary to understand the entity’s identity within a particular domain or organizational context.
For a person, these may include:
- a primary professional role;
- secondary roles;
- organizational positions;
- professional domains;
- institutional responsibilities;
- or historical roles.
For an organization or other entity, structural classifications may include:
- organization type;
- operational role;
- parent or subsidiary relationships;
- regulatory classification;
- governance position;
- or other publisher-known structural attributes.
Roles and classifications should remain distinct from the entity’s canonical identity. A role may change while the underlying entity remains the same.
Affiliations
An IdentityFragment may identify formal affiliations between the entity and other identifiable entities.
Affiliations may include:
- employment relationships;
- organizational membership;
- ownership relationships;
- parent and subsidiary relationships;
- institutional appointments;
- governance relationships;
- professional associations;
- or other explicitly established structural relationships.
Affiliations should identify the related entity using a stable identifier when one is available.
Where the affiliation has a temporal scope, that scope should be preserved so that historical and current relationships are not conflated.
Example IdentityFragment
The following example represents the structural identity of an organization using the current WebMEM HTML serialization.
<template
data-webmem-fragment
data-fragment-class="IdentityFragment"
data-fragment-id="identity-example-organization">
<section
data-entity-type="Organization"
data-entity-id="org:example-institute"
data-provenance-ref="#provenance-example-organization">
<h3 data-role="entity-name">
Example Research Institute
</h3>
<span data-role="canonical-role">
Research Organization
</span>
<ul data-role="domain-scope">
<li>Artificial Intelligence</li>
<li>Knowledge Engineering</li>
<li>Information Architecture</li>
</ul>
<a
data-role="canonical-resource"
href="https://example.org/">
Example Research Institute
</a>
<meta
data-role="identity-version"
content="1.0">
</section>
</template>
The fragment establishes that:
- the knowledge object is an
IdentityFragment; - the fragment has a stable identity;
- the entity has its own stable entity identity;
- the entity is explicitly classified as an organization;
- its canonical name and structural role are represented;
- its domain scope is explicit;
- a canonical resource associated with the entity is identified;
- the identity declaration is versioned;
- and provenance supporting the identity can be resolved through the referenced provenance object.
Multiple Identifiers
An entity may be known by multiple identifiers across different systems.
An IdentityFragment may expose those identifiers while preserving their relationship to the same canonical entity.
<ul data-role="identifiers">
<li
data-identifier-type="contract-id"
data-identifier-value="H5521">
H5521
</li>
<li
data-identifier-type="publisher-id"
data-identifier-value="carrier-example-health">
carrier-example-health
</li>
</ul>
Different identifiers should not be treated as interchangeable unless they actually identify the same entity at the same level of scope.
For example, a contract identifier and a plan identifier may participate in the same resolution structure while identifying different entities.
Aliases and Alternate Names
An IdentityFragment may identify aliases, alternate names, abbreviations, historical names, or other labels associated with an entity.
Alternate labels should remain subordinate to the canonical entity identity.
Where a name is historical or no longer current, that status should be represented explicitly when necessary to prevent an obsolete label from being interpreted as the current canonical name.
Identity Versioning
An entity’s identity may remain stable while its roles, affiliations, classifications, names, or other structural attributes change over time.
An IdentityFragment may therefore include version information when the publisher needs to preserve the evolution of an identity declaration.
Versioning should distinguish between:
- changes to the published identity representation;
- changes to attributes of the entity;
- changes to roles or affiliations;
- and replacement of one entity by a genuinely different entity.
A change in role or affiliation does not necessarily create a new entity identity.
Historical and Retired Relationships
An IdentityFragment may preserve historical, superseded, or retired relationships when those relationships are necessary to understand the entity’s evolution.
Examples may include:
- former organizational affiliations;
- previous professional roles;
- historical names;
- retired web resources;
- superseded organizational structures;
- or other identity relationships that were previously valid.
Historical relationships should be explicitly distinguished from current identity attributes.
Retirement does not necessarily mean deletion. A historical identity relationship may remain useful for understanding continuity between previous and current representations of the entity.
IdentityFragment and PersonaFragment
IdentityFragment and PersonaFragment may both contain information about people or human contexts, but they serve fundamentally different purposes.
| IdentityFragment | PersonaFragment |
|---|---|
| Represents the canonical identity of an entity. | Represents a contextual human model, archetype, or scenario. |
| Answers: Who or what is this entity? | Answers: What circumstances, goals, challenges, or values define this human context? |
| Uses stable identifiers and structural relationships. | Uses contextual characteristics and scenario framing. |
| May represent an actual person or organization. | May represent an illustrative, composite, or research-derived persona. |
A PersonaFragment should not be used as a substitute for canonical identity, and an illustrative persona should not be represented as though it were an actual identifiable person.
IdentityFragment and MetaFragment
An IdentityFragment and a MetaFragment may both describe identifiable objects, but their semantic purposes differ.
An IdentityFragment establishes what the entity is.
A MetaFragment describes an information resource or dataset and its publication metadata.
For example, an IdentityFragment may establish the identity of an organization while a MetaFragment describes a dataset published by that organization.
IdentityFragment and IndexFragment
An IndexFragment may enumerate identifiers or entities belonging to a defined collection.
An IdentityFragment may establish the canonical identity associated with one of those indexed members.
For example:
IndexFragment → Entity ID → IdentityFragment → Canonical Entity
The IndexFragment establishes membership. The IdentityFragment establishes what the indexed entity is.
IdentityFragment and DirectoryFragment
A DirectoryFragment may contain structured entries representing identifiable entities.
An IdentityFragment may provide the canonical identity for a directory entry when richer identity information is required.
For example:
Directory Entry → Entity ID → IdentityFragment → Canonical Entity Resource
This allows directory-specific attributes to remain within the DirectoryFragment while canonical entity identity remains independently represented.
IdentityFragment and Relationships
IdentityFragments participate directly in the WebMEM relationship model because identity is often established partly through relationships to other entities.
An IdentityFragment may relate an entity to:
- a parent organization;
- a subsidiary;
- an employer;
- a professional role;
- a governing organization;
- a product or service;
- a geographic entity;
- a dataset;
- a canonical resource;
- a previous identity representation;
- or another explicitly identified entity.
Relationships should preserve the identities of both participating objects and the nature of the relationship between them.
Read the Relationships specification →
IdentityFragment and Resolution
IdentityFragments play a central role in the WebMEM Resolution Model because resolution depends on knowing what an identifier represents and how that identity connects to other information objects.
A resolution path may take the form:
Identifier → IdentityFragment → Canonical Entity → Canonical Resource
or:
Indexed Member → Entity ID → IdentityFragment → Related Entity
or:
Alias → IdentityFragment → Canonical Name → Canonical Entity
or:
Historical Identity → IdentityFragment → Current Identity Representation
The IdentityFragment provides the identity anchor. Resolver structures may use that identity to connect identifiers, aliases, indexed members, relationships, historical representations, and canonical resources to the same entity.
This makes identity resolution distinct from simple identifier matching. An identifier provides a reference. The IdentityFragment establishes what that reference identifies and preserves the structural context necessary to distinguish it from other entities or levels of identity.
Read the WebMEM Resolution Model →
IdentityFragment and Resolvers
An IdentityFragment and a resolver serve complementary roles within WebMEM.
The IdentityFragment answers:
Who or what is this entity?
A resolver answers:
How can a machine move from the identifier or context it currently has toward the appropriate entity or information resource?
A resolver may therefore use an IdentityFragment to establish the entity represented by an identifier before continuing toward related resources, collections, facts, or other resolvable objects.
For example:
Contract ID → Resolver → IdentityFragment → Contract Entity → Contract Resource
or:
Plan ID → Resolver → IdentityFragment → Plan Entity → Plan Resource
The resolver provides the path. The IdentityFragment provides the canonical identity encountered along that path.
Read the Resolvers specification →
Provenance
An IdentityFragment should preserve the evidentiary basis for the identity declaration and the structural attributes it publishes.
Provenance may identify:
- the authoritative or publisher-controlled source establishing the entity’s identity;
- the source of external identifiers;
- the source supporting names, roles, classifications, or affiliations;
- the organization or authority responsible for the identity information;
- the canonical source location;
- publication, effective, or revision dates;
- retrieval or access dates;
- version information;
- the temporal scope of roles or affiliations;
- and other lineage information necessary to establish the basis of the identity representation.
When different identity attributes originate from different sources, the provenance model should preserve those distinctions rather than implying that every attribute was established by the same authority.
Publisher-declared roles, classifications, or identity statements should be distinguishable from identities or classifications established by external authoritative sources when that distinction is material.
Read the Provenance specification →
Conformance
A conforming IdentityFragment must:
- identify itself as an
IdentityFragment; - have an identifiable fragment scope;
- provide a stable identity for the entity being represented;
- provide a canonical name or equivalent identifying label;
- identify the entity type when necessary to distinguish what the identifier represents;
- preserve the distinction between fragment identity and entity identity;
- preserve the distinction among identifiers that represent different entities or different levels of entity scope;
- identify canonical resources when they are part of the published identity representation;
- preserve applicable roles, affiliations, classifications, historical relationships, or temporal scope when those attributes are necessary to interpret the identity correctly;
- identify or reference provenance supporting the identity declaration;
- preserve relationships to other entities or knowledge objects when those relationships form part of the structural identity;
- and conform to the applicable WebMEM serialization requirements.
An IdentityFragment represents the publisher’s structured identity declaration. Conformance does not imply that every published role, affiliation, classification, or relationship has been independently verified unless the associated provenance establishes that verification.
Relationship to WebMEM 1.x
The IdentityFragment class originated in the WebMEM 1.x Semantic Data Template specification and remains a defined fragment class in the current WebMEM Protocol.
Earlier specifications described IdentityFragment as a mechanism for stabilizing machine classification through canonical roles, secondary roles, domain scope, affiliations, governance positioning, intent classification, retired nodes, versioning, and retrieval hints.
The current protocol preserves the semantic purpose of the class while separating that purpose from retired WebMEM 1.x retrieval weighting, memory conditioning, reclassification, and agent-classification mechanisms.
Canonical roles, secondary roles, domain scope, affiliations, governance relationships, historical or retired relationships, and versioning remain valid identity structures when they are necessary to represent the entity accurately.
The current protocol broadens the class beyond role stabilization by making explicit the underlying identity architecture: entity identity, entity type, stable identifiers, canonical resources, structural attributes, relationships, temporal continuity, and provenance.
The current protocol also makes explicit the central role of IdentityFragment in machine resolution. An identifier can participate in resolution only when the publisher can establish what that identifier represents and how the identified entity relates to other knowledge objects or canonical resources.
IdentityFragment continues to represent the canonical structured identity of an entity and the identifiers, roles, affiliations, classifications, relationships, and provenance necessary to distinguish and resolve that entity consistently.