DCMIThe Learning Resource Metadata Initiative (LRMI) provides a lightweight, web-native vocabulary for describing educational resources in ways that make them more discoverable to learners, educators, and search engines. Now maintained by the Dublin Core Metadata Initiative, LRMI bridges the gap between educational content publishers and the structured data ecosystem of Schema.org, enabling learning resources to be found through general web search with rich, education-specific metadata.
Background
LRMI was launched in 2011 as a joint project of the Association of Educational Publishers (AEP) and Creative Commons, funded by the Bill & Melinda Gates Foundation and the William and Flora Hewlett Foundation. The initiative recognized that while enormous amounts of educational content were being published online, there was no standardized way to describe key educational attributes — such as target audience, educational level, or alignment to learning standards — using web markup that search engines could understand.
Between 2011 and 2014, the LRMI project developed a specification of properties for educational resource description, many of which were proposed to and adopted by Schema.org. In 2014, stewardship of LRMI was transferred to the Dublin Core Metadata Initiative (DCMI), which continues to maintain and develop the specification through its LRMI Working Group.
Purpose & Scope
LRMI defines classes, properties, and concept schemes specifically tailored for the description of educational resources. The specification is designed to work within the Schema.org ecosystem, extending it with education-specific semantics. Key descriptive dimensions include:
- Educational alignment — Connecting resources to curriculum standards and learning objectives
- Educational level — Target audience by age, grade, or educational stage
- Learning resource type — Categorizing materials (lesson plan, textbook, assessment, etc.)
- Interactivity type — Whether the resource is active, expositive, or mixed
- Time required — Typical learning time needed
The concept schemes provide controlled vocabulary terms as SKOS concepts, suitable for use as values with LRMI and Schema.org properties in web markup (RDFa, JSON-LD).
Serializations & Technical Formats
LRMI is designed for use as structured data embedded in web pages:
- JSON-LD — The primary recommended format for embedding LRMI metadata in HTML
- RDFa — Inline HTML attribute-based markup, also widely supported
- The LRMI Terms specification is published in RDF and builds on Schema.org types
The concept schemes are declared using the W3C's Simple Knowledge Organization System (SKOS), providing standardized hierarchical and associative relationships among terms.
Governance & Maintenance
LRMI is maintained by the LRMI Working Group, a community group within DCMI. The specification has evolved from its original version 1.1 (the version transferred from the LRMI project in 2014) to the current LRMI Terms, which are maintained as a living specification in RDF. DCMI's Usage Board provides oversight for the vocabulary's formal status.
DCMI documents, including LRMI, are licensed under a Creative Commons Attribution 4.0 International License.
Notable Implementations
LRMI metadata is used across a wide range of educational platforms and repositories:
- Schema.org — Several LRMI properties have been adopted directly into Schema.org's core vocabulary
- Google Search — Recognizes LRMI-compatible structured data for educational content discovery
- OER Commons — Open educational resource repository using LRMI for resource description
- MERLOT — Multimedia Educational Resource for Learning and Online Teaching
- Learning Registry — US Department of Education initiative that supported LRMI adoption
- Publishers and content platforms display "LRMI enhanced" badges to signal their use of the specification
Related Standards
- Dublin Core Metadata Terms — LRMI's parent vocabulary framework within DCMI
- Schema.org — The web vocabulary that LRMI extends with educational properties
- IEEE LOM — The earlier Learning Object Metadata standard; LRMI provides a lighter-weight, web-native alternative