Imagine trying to find a book in the library without the aid of the Dewey Decimal system or the card catalogue system (online or manual). Or imagine walking aimlessly through stacks and stacks of books that have been organized in no seemingly apparent manner. This is the feeling many people have as they search the World Wide Web for the valuable materials they know are there, but they just can't find. In order to access relevant, high quality learning materials, faculty, learners, employers and parents have all identified the need for better search strategies and mechanisms. This need will be answered by the Educom NLII Instructional Management System (IMS) Project specifications and Java-based software for metadata.
Metadata is a formal description of resource materials. The author, title and subject information in card catalogs and digital library databases are examples of metadata describing books in library collections.
The primary purpose of metadata is to provide more helpful information about a work than can be obtained by inspecting the contents of the work, e.g., a Web page may be designed to teach mathematics skills to a third grade audience, but the terms "third grade" and "mathematics" may not appear in any of the text of the Web page. Therefore, traditional Web search engines, which often utilize full-text search indexes, would not return the page if "third grade and mathematics" were used as the search criteria.
Standards for metadata allow information and materials to be easily and consistently located. Unfortunately, where metadata solutions exist today, they are not consistent and are often proprietary. This has created an administrative nightmare for organizations that own or manage large collections of Web-based materials. It is these administrative challenges and the potential benefits to users that are driving the Internet industry to solve this metadata problem. The NLII IMS is building upon the industry's technology efforts by defining the necessary metadata elements to support widespread reuse, discovery and sharing of learning materials via the Internet.
Based on evolving metadata technology standards from the World Wide Web Consortium and input from many organizations, the IMS specifications and software greatly simplify the process of locating learning materials on the Web. IMS partners and other organizations are "tagging" their content so that it is compliant with the IMS Metadata specifications.
The development of the IMS Metadata specifications has been a collaborative effort in soliciting and consolidating input from many individuals and organizations. Faculty, students, publishers and administrators established requirements for the IMS project that were then reflected in the selection of metadata fields. Representatives from the National Institute of Standards and Technology, Pennsylvania State University, University of North Carolina, International Thompson Publishing, California State University - Sonoma, and the National Science Foundation-funded digital library project at University of Michigan attended IMS metadata meetings. IMS staff have actively participated in IEEE learning object metadata discussions, and they have consulted with members of the ANSI X3L8/ISO JC1/SC14 groups, Apple's Educational Object Economy (EOE) project, Microsoft and Netscape. Existing work in the field, such as the "Dublin Core" and the ISO 11179 standard, has also informed the developme
nt of the IMS Metadata specifications.
IMS Metadata can be included in Web pages, programmed into Javabeans or ActiveX objects, or stored as separate files. At a high level, metadata is represented as fields with labels and values, e.g., AUTHOR = John Doe; DATE = 08/15/1997. In some cases, a field can have multiple values, such as FORM = Simulation, Exercise. The specific syntax of how to represent metadata in HTML files had not been released by the World Wide Web Consortium at the time of this writing. The IMS Web site will include examples by the time this article is published.
Some major Web search engines have already adopted early metadata technology. It is expected that all the major search engines will support the new metadata technology standards released by the World Wide Web Consortium. Therefore, users will be able to find material "tagged" with IMS Metadata by using their favorite Web search engine.
Several factors affected the design of the IMS Metadata specifications. First, IMS Metadata has to be simple to fill out. For this reason, the IMS Metadata includes a small set of mandatory fields and a secondary set of optional fields. Second, IMS Metadata has to be extensible in order to allow authors and publishers to add new keys and values, some of which may be specific to certain collections of information or disciplines. Third, the IMS Metadata has to support internationalization. All human-readable IMS Metadata fields can be represented as multiple pairs of values and the language in which the value is written, e.g., TITLE = The Boat, English; Das Boot, Deutsch.
The IMS Project has developed a Web-accessible tool for visitors to create IMS Metadata and to experience the benefits of searching with IMS Metadata. Simple forms and selection lists are used to speed entry of IMS Metadata. The Metadata tool then automatically generates IMS Metadata in the proper format. The IMS Metadata tool is currently accessible via the IMS Web site, and it will be part of the IMS software to be made freely available in March of 1998.
The IMS Project is working with Apple Computer, an IMS Corporate Partner, so that the Apple Educational Object Economy Project will use IMS Metadata to describe the Java-based materials in its collection. See (http://trp.research.apple.com/).
A partial list of the IMS Metadata fields follows. At the time this article was written, the IMS Metadata Specifications were in a public review phase. Therefore, the final specification may or may not include the following fields.
Abstract - An overview of the work.
Format - For digital works the format is the mimetype. For non-digital works,
the format is a term for that work, such as book, videotape, etc.
Keywords - Terms used to describe the work.
Language - The human language in which the work is presented.
Location - Where the work can be found. Typically this will be a URL.
Steward - The organization or individual to contact regarding any
licensing/purchase of the work.
Title - The label for the work.
Author - The person who created the work.
Credits - Lists of contributors to the work.
Expiration Date - The last date that the work can be used.
Learning Level - The difficulty of the work.
Objectives - Lists the learning objectives met by the work.
Pedagogy - The teaching method used in the work.
Platform - Hardware and software requirements.
Prerequisites - Course and/or capabilities required to use the work.
Presentation - How the work is presented, such as text, images, video etc.
Price Code - The cost of the work. Price code can be 0.
Relation - Other works related to this work.
Source - The work, if any, that this work was derived from.
Subject - The knowledge domain of the work.
Use Rights - Defines the permitted uses for the work, such as whether or not the work is allowed to be incorporated into other works or whether parts of the work can be used in other works.
Use Time - The approximate time required for a student to use the work.
Additional information about the IMS Metadata specifications can be found at http://www.imsproject.org/metadata/.
Testimonials
"Educom's IMS Project is creating the learning object metadata specifications that will be adopted by the Apple-hosted Educational Object Economy (http://trp.research.apple.com/EdEconomy). The EOE provides the world's largest directory to Java-based educational applets, and community members are welcome to try out and comment on any applet in the directory.
"We know metadata can help people more easily find and make appropriate use of educational objects, and we are excited about the opportunity to provide an early testbed for the IMS metadata specification. NIST and other government agencies, industry and universities all have made important contributions to this work. We'll be encouraging others in the education community to adopt, comment on, and evolve the specification as together we gain experience with this excellent and much needed starting point."
Jim Spohrer, Distinguished Scientist, Apple's Learning Communities Group
"The Department of Defense is investigating more efficient and effective means of developing and conducting military training. Of particular interest is the creation of an object-oriented open architecture environment for advanced distributed learning which will enable military personnel to access information needed to support their individual missions and tasks regardless of time or place. The IMS metadata specification is an important part of a wide-scale collaborative effort with industry and academia to create the tools and content needed for this new learning environment."
Michael Kendall
Director of Training
Readiness and Training Office
Office of the Secretary of Defense
Steve Griffin is director of technology at the University of North Carolina at Chapel Hill. [email protected]
Tom Wason is with the Institute for Academic Technology at the University of North Carolina at Chapel Hill. [email protected]