Penn State Teams Use Metadata to Get Digital Resources in Order

April 16, 2002

Digital technology has taken learning resources from the static confines of

the old overhead projector to easily accessible, interactive experiences that

include sound and motion. But despite the dramatic evolution in the quality

of learning resources, there has been little progress towards a system of classification

that allows educators to efficiently identify, locate, and reuse these materials.

Now, two teams, whose members include researchers and information technology

experts from Penn State, are participating in projects designed to bring order

to a bourgeoning but chaotic collection of digital educational resources.

At the heart of their projects is a common, indispensable ingredient: metadata.

The term may evoke philosophical abstraction, but metadata, or data about data,

have many practical uses. Libraries use metadata to help people find books,

programmers use metadata to store information, and scientists use metadata to

catalog research.

Several years ago, Instructional Management Systems Global Learning Consortium,

Inc., (IMS), a group of educational, corporate, and government entities dedicated

to making advances in distributed learning, began working on defining metadata

to include in electronic learning materials to help educators more easily identify

and share resources.

An IMS technical board has developed recommendations for the types of metadata

that producers of educational material should include in the resources they

create, and how they should be included. The metadata defined by IMS includes

hundreds of pieces of information from the basic, such as "author,"

to items that communicate a much greater level of pedagogical detail, like "time

spent on the activity," and "semantic density."

The Center for Institutional Cooperation (CIC), an educational consortium of

twelve major research and teaching universities including Penn State and charter

member of the IMS specification project, hopes to demonstrate the value of the

IMS specifications through the development of a prototype repository for metadata.

The CIC Educational Resource Repository (CICERO) project, whose team includes

members from a number of CIC institutions, is led by Mike Halm from Penn State’s

Center for Education Technology Services. "One of the biggest problems

we have with first generation digital resources," says Halm, "is that

there is no descriptive data associated with the objects." This makes them

hard to find.

The CICERO repository is designed to allow users to add IMS specified metadata

to materials to clearly identify them, to store the metadata, and to provide

a way to search for resources using the metadata that will yield highly specific

and accurate results. Users testing the prototype will perform these functions

using a Web-based interface.

"There are a number of benefits to users of such a repository," Halm

observes. "It exposes their digital resource to a community which might

potentially find it useful; there is the potential of revenue from its sale

or licensing; it gives developers recognition for their contribution to their

field of study; and it may encourage the development of communities in different

disciplines to review and discuss such materials."

If the prototype provides the hoped for functionality, it will have helped

to lay the groundwork for future repositories scalable to larger groups of users,

such as the twelve universities that make up the CIC.

The Penn State Visual Image User Study (VIUS), a project funded by the Mellon

Foundation, shares CICERO’s goal of helping users more efficiently identify,

store, and locate learning resources. VIUS (pronounced "views") however,

focuses specifically on the use, storage, and retrieval of digital images.

The VIUS team, which also includes Halm, will first conduct a detailed assessment

of the ways Penn State faculty and students use digital images. It will then

use the findings of the study to design a system that will hold both the metadata

that describe digital images, and the images themselves.

"Users will be able to search and find digital images using metadata,"

explains Halm. "Once the search has been narrowed, they will be able to

view thumbnails of the images, then select those they want and build collections

for a variety of academic or research purposes."

The system will be designed to provide access to digital images created by

local faculty and students, as well as those available through Penn State collections.

But the team also intends to design the system so that it may be integrated

with digital image retrieval systems at other educational institutions, museums,

and public and private collections.

"Accommodating the use of these types of collections into a single educational

experience," as Halm describes the ultimate goal of the project, will be

possible in large part due to the implementation of IMS specifications for metadata

by institutions with digital image holdings. Halm also notes that the use of

IMS specifications will make it possible in the future to integrate the repository

with Penn State’s Course Management System, ANGEL.

To learn more about IMS and its specifications, visit its Web site: http://www.imsproject.org/