Dublin Core
Dublin Core
About DCMI Themes Members Governing Board Usage Board Directorate
Twitter YouTube GitHub RSS Feed

Memorandum of Understanding

Memorandum of Understanding between the Dublin Core™ Metadata Initiative and the IEEE Learning Technology Standards Committee

Introduction

The Dublin Core™ Metadata Initiative (DCMI) and the IEEE Learning Technology Standards Committee P1484 (IEEE LTSC) express their joint commitment to collaboration on the development of interoperable metadata for learning, education and training. In particular, both organisations see benefit in maintaining compatibility between the ongoing development of the IEEE LTSC Learning Object Metadata (LOM) and the Dublin Core™ Metadata Element Set (DCMES). It is the view of both parties that it is vital that the learning, education and training communities continue to be able to take advantage of tools and services that emerge from both initiatives – the DCMI approach to developing metadata standards for resource discovery across domains and flexible detailed descriptions for ‘learning objects’ as supported by the IEEE LOM. Both DCMI and IEEE LTSC have long held certain fundamental assumptions regarding the development of metadata specifications for the web, and our ability to collaborate is directly tied to a mutual commitment to these assumptions.

Philosophy

DCMI and IEEE LTSC LOM constituents have substantially overlapping technical and descriptive metadata requirements and are therefore best served by a single architecture that presents a minimum of barriers to the creation, interchange, and use of metadata. Based on this philosophy, we agree to support modular, extensible, structured metadata. For the interchange and use of metadata, the World-Wide Web is a strategic medium, and we agree that an effective architecture must be deployable using existing web infrastructure, and can be expected to evolve along with that infrastructure.

Architecture

Features of a common architecture include a commitment to:

  • Open extensibility: Communities or applications must be able to refine or extend the semantics of the elements, to support local or community needs. On the other hand, such extensions should not compromise cross-domain interoperability. We agree that extensions should be identified in a standards-based way as to which community has created the extension and to produce guidelines which may facilitate this.
  • Modularity: The architecture should support the use of complementary groups of metadata, the characteristics and evolution of which may be managed by distinct entities or agencies. For example, DCMI addresses general characteristics of web resources, such as author and title. The IEEE LOM focuses on detailed descriptions of learning objects. Therefore, a description of a learning, education or training resource could include elements from these two (or possibly more) sources. An example of an additional group of elements in the broad 'e-learning' community could be discipline-specific extensions.
  • Refinement: Appropriate values for metadata elements can be refined through the use of specific schemes or encoding standards. Examples of such refinement include the specification of date encodings (e.g., ISO 8601) to assure unambiguous parsing rules, the specification of a classification scheme (e.g., Dewey Decimal Classification), or the definition of a controlled vocabulary (e.g., roles of contributors to a resource in IEEE LOM).

Semantics

We agree that, in order to best support the modular construction of metadata, metadata elements should not contain overlapping semantics. This means that the same concept or definition should not be present in two different groups of elements. We recognize that this type of overlap can and does occur, especially with legacy metadata collections. However, DCMI and IEEE LTSC agree that it is in our communities' best interest if our specifications are developed to avoid such semantic overlap. We further agree that metadata elements should be defined to reflect broad geographic, cultural, and linguistic requirements.

Syntax

Philosophical, architectural or semantic aspects of this Memorandum of Understanding are independent of any technology or syntax for expressing metadata. In order to build tools and applications that actually implement these aspects, some syntax is required and we agree to collaborate on interoperability in that domain as well. As of this date, there are a number of technologies that we envision using to create common specifications, including:

  • Simple representations of embedded metadata in HTML are and will continue to be important for significant segments of our communities and establishing common representations of our standards in HTML is important.
  • XML can be expected to accommodate the expression of rich metadata, and has the additional advantage of being a Web industry standard for data and document encoding.
  • The Resource Description Framework (RDF) provides a useful approach to satisfy the requirements of extensibility and modularity, and can thus lead to direct interoperability between each communities' metadata requirements.

Ongoing Collaboration

As implementation of metadata takes place within the learning, education and training communities new or refined requirements will arise. For those requirements that address general metadata (i.e. not specific to learning, education and training) the IEEE LTSC will forward those requirements into the DCMI community for discussion and specification. For those requirements that are specific to learning, education and training, DCMI, through the DC-Education Working Group, will forward those requirements to the IEEE LTSC LOM Working Group. The chair(s) of the DC-Education Working Group will have invited expert (voting) status within the LTSC LOM Working Group. Additional dialog between the groups on learning, education and training metadata will take place in public forums such as email discussion list.

Implementation of this Agreement

This document expresses our agreement on and common understanding of the aims and goals of educational metadata. The DCMI and IEEE LTSC LOM groups will make this agreement more concrete through an upcoming technical document that will document how we expect to reach common specifications, agreements and standards. Completion of the next round of technical agreements is scheduled for December 06, 2000.

Signatories

Wayne Hodgins for IEEE LTSC Stuart Weibel for DCMI

Concurring Projects

Ed Forte for ARIADNE Jon Mason for EdNA Stuart Sutton for GEM Ed Walker for IMS