Term reference for the Transparent Research Object Vocabulary (TROV). Lists every class, property, and predefined value in the current draft, with descriptions and cardinality constraints.
The warrant chain is TROV’s mechanism for accountability. When transparency claims are made about a TRO, the warrant chain records who is responsible for them.
TRO attribute
← warranted by → TRP attribute
← warranted by → TRS capability
Class
Description
trov:TREAttribute
Base class for attributes of a TRS, TRP, or TRO.
trov:TRSAttribute
An attribute of a TRS. Subclass of TREAttribute.
trov:TRSCapability
A declared ability of a TRS to enforce a specific transparency condition. Subclass of TRSAttribute.
trov:TRSPolicy
A TRS-level assertion about a capability. Subclass of TRSCapability. Under redesign — see Known Limitations.
trov:TRPAttribute
A transparency claim about the conditions under which a specific TRP was conducted. Subclass of TREAttribute.
trov:TROAttribute
A transparency claim about a TRO as a whole, warranted by attributes of its constituent TRPs. Subclass of TREAttribute.
The TRS’s public key (optional). Required for GPG signing (binds the key to the signed declaration content). Not needed for X.509/CMS (the certificate chain is in the .p7s file).
trov:customTerm
TrustedResearchSystem
Resource
Declares a custom term that extends a TROV class. Each value is a resource with an rdfs:subClassOf triple connecting it to the TROV class it extends. See TROV Extension Guide.
One or more hashes of an artifact’s contents or of a composition’s sorted artifact hash values. Each object contains trov:hashAlgorithm and trov:hashValue.
trov:hashAlgorithm
(within trov:hash object)
Literal (string)
The algorithm used to compute the hash (e.g. "sha256", "hmac-sha256").
The following capability types are currently defined. This list is designed to grow as new transparency conditions are identified by the research community.
TROV uses terms from the following external vocabularies:
Prefix
Namespace
Usage
rdf:
http://www.w3.org/1999/02/22-rdf-syntax-ns#
rdf:type
rdfs:
http://www.w3.org/2000/01/rdf-schema#
rdfs:comment, rdfs:Class, rdfs:subClassOf
schema:
https://schema.org
schema:CreativeWork, schema:Organization, schema:creator, schema:name, schema:description, schema:dateCreated. Optional — used in tro-utils for TRO and TRS metadata but not required for conformance. See TRO Declaration Format — dependency boundary.
Namespace. The 0.1 namespace is https://w3id.org/trace/trov/0.1#; the pre-release namespace https://w3id.org/trace/2023/05/trov# is retired. See Pre-Release Vocabulary Reference for migration guidance.
Extensibility. The capability and attribute type lists are designed to be extended. New TRS capability types and corresponding performance/TRO attribute types can be added as new transparency conditions are identified. Existing TRS certificates and TRO declarations remain valid when new types are introduced.
Signing mechanism. TROV is agnostic to the specific signing technology. Current implementations use GPG and X.509/CMS. For GPG, trov:publicKey is required in the declaration to bind the key to the signed content. For X.509/CMS, the certificate chain in the .p7s file provides identity binding, so trov:publicKey is not needed. See TRO Packages — Signing identity and trust for details. A proposed trov:signingMechanism property is under review for 0.1.
Interoperability. TROV aims to be interoperable with, rather than to replace, current and future Research Object standards (e.g., RO-Crate), archival formats (e.g., BagIt), and repository layouts. TROV complements the W3C PROV-O ontology for describing general provenance relationships.
JSON-LD context. TRO declarations use a @context block to map short-form term names to the URIs defined here. See TRO Declaration Format for the structure and TRO Declaration Design for the dual-audience rationale.