A complete reconstruction of the W3C PROV-O provenance ontology in MDLD format, fully projected from the authoritative PROV-O specification with comprehensive narrative documentation.

This project presents the W3C PROV-O standard reimagined for modern provenance workflows, featuring:

• Complete PROV-O Implementation - All classes, properties, and relationships faithfully reconstructed from prov-o.ttl
• Single Consolidated Document - Build script generates prov-o.md, a complete readable MDLD ontology with all explanations
• Flow-Based Organization - Components organized by conceptual flows for better understanding
• Rich Documentation - Narrative summaries explaining purpose and use cases
• Real-World Examples - 11 practical scenarios demonstrating provenance patterns

Start here: PROV-O MDLD Catalog

Build the complete ontology document:

node consolidate.js

This generates prov-o.md - a single consolidated document containing all categories and components with full narrative explanations.

The PROV-O ontology is presented through two complementary organizational approaches that serve different purposes.

Categories organize terms by learning complexity, providing a progressive path for understanding provenance concepts. Starting Point Terms introduce the core concepts needed for immediate use — Entity, Activity, Agent, and their basic relationships like prov:wasGeneratedBy and prov:used. These form the foundation for creating simple provenance descriptions. Expanded Terms then offer additional patterns for more complex scenarios, including specialized classes like Person and Organization, temporal properties for tracking entity lifetimes, and structures for handling collections and bundles. Finally, Qualified Terms provide enhanced relationships with contextual detail through qualified forms that capture additional information about binary relations, enabling more precise provenance descriptions.

Components organize terms by functional modules, grouping related provenance concepts based on their role in describing different aspects of provenance. The Entities and Activities component covers the core building blocks and lifecycle management, modeling the fundamental "what" and "when" of provenance. Derivations focuses on entity transformation and lineage tracking, describing how entities derive from one another through various transformation processes. Agents, Responsibility and Influence addresses accountability and authority structures, capturing who is responsible for activities and entities. Collections handles dynamic membership and provenance of groups, while Bundles enables nested provenance and provenance-of-provenance for describing provenance about provenance. The Alternate component deals with multiple perspectives and specialization relationships, linking entities that represent different aspects of the same thing.

These two views interact: each PROV-O term belongs to both a category (based on complexity) and a component (based on function). For example, prov:wasGeneratedBy is a Starting Point term (simple, core concept) in the Entities and Activities component (deals with entity-activity relationships). This dual organization supports both progressive learning and functional reference, allowing users to approach the ontology from whichever perspective suits their needs.

Browse the complete catalog at index.md for detailed exploration of all categories, components, and 11 real-world examples.

• Complete Ontology Projection - The authoritative prov-o.ttl (W3C PROV-O specification) is fully projected to MDLD format with all explanations and narrative
• Universal Build Script - consolidate.js dynamically discovers and consolidates all linked markdown files from index.md, making it reusable for any MDLD project
• Single Consolidated Document - Generates prov-o.md containing all categories, components, and examples in one readable document
• 35 Entities, 16 Activities, 16 Agents - Comprehensive instance data for validation
• Flow-Based Summaries - Each component includes narrative explanations of conceptual flows
• Progressive Learning - Categories organized from basic to advanced concepts
• Modern Tooling - Built with MDLD for enhanced readability and validation

The consolidate.js script creates a complete, single-file ontology document by:

1. Starting from index.md - Reads the main catalog file as the entry point
2. Extracting markdown links - Finds all relative .md links using the pattern [text](path.md)
3. Recursively following links - Visits each linked file and extracts its links, building a complete tree
4. Infinite recursion protection - Tracks visited files to prevent processing the same file multiple times
5. Anchor ID generation - Creates unique anchor IDs for each file based on its path (e.g., categories-starting-point-index)
6. Internal link conversion - Converts internal markdown links to anchor links (e.g., [text](path.md) → [text](#anchor-id))
7. Link normalization - Normalizes image and resource links with ./ prefix preservation
8. Combining with separators - Files are joined with MDLD section separators to maintain document structure

This approach is universal for MDLD authoring - any MDLD project with an index.md file can use this script to consolidate all linked documentation into a single file with preserved navigation.

Result: prov-o.md - a complete, readable ontology document containing the entire PROV-O specification with all narrative explanations and working internal navigation, suitable for:

• Single-file reference with working internal links
• Easy distribution
• Comprehensive review
• Integration into documentation systems

Clone the repo via Intelligraphs app

mkdir mdld-prov
cd mdld-prov
git clone https://github.com/davay42/mdld-prov.git 
curl -O https://app.intelligraphs.com/ig-cli.js
chmod +x ig-cli.js
mv ig-cli.js /usr/local/bin/ig-cli
ig-cli --help
ig-cli stats

This reconstruction maintains full compatibility with the W3C PROV-O standard while adapting it for modern MDLD workflows. The authoritative source is prov-o.ttl (the official W3C PROV-O specification), which has been fully projected to MDLD format.

Ontology Statistics (via ig-cli):

• 57 supported files
• 2,830 RDF quads across 52 named graphs
• 226 subjects, 87 predicates
• 39 RDF types including all PROV-O classes and properties
• 35 Entities, 16 Activities, 16 Agents (instance data)

Adaptations for MDLD:

• Changed owl:Class to rdfs:Class for better MDLD integration
• Excluded dm and n links for cleaner documentation
• Enhanced with narrative summaries and flow-based organization
• Added category-based learning progression (Starting Point → Qualified → Expanded)
• Component organization by functional modules with conceptual flows

Build Achievement: The consolidate.js script successfully combines the entire ontology into a single readable document (prov-o.md), making the complete PROV-O specification with all explanations available in one file for easy reference and distribution.