Learning objects symposium special issue guest editorial.

We are very proud to introduce this special issue of the Journal of Educational Multimedia and Hypermedia. The papers included are the best papers we received for a Symposium on Learning Objects that we organized in 2003, as a pre-conference Symposium to ED-MEDIA 2003 (Duval, Hodgins, Rehak, & Robson, 2003). The use of learning objects promises to increase the effectiveness of learning by making content more readily available, by reducing the cost and effort of producing quality content, and by allowing content to be more easily shared. However, the development and deployment of learning objects also present new challenges to designers of learning experiences and technology alike. For that reason, we took the initiative to organize this symposium to explore the benefits and challenges of learning objects.

**********

THE INTRODUCTION

Initial Observations

From the original 51 submissions we received for the Symposium, this special issue includes nine reworked and extended papers. In the introduction to the Symposium proceedings (Duval, Hodgins, Rehak, & Robson, 2003), we mentioned that we were struck by a number of issues:

* There is a diversity of contexts where learning objects are being employed. Even though most submissions referred to applications in academia (no doubt influenced by the overall EdMedia focus), we also received submissions that dealt with industrial and military training. As far as the domain is concerned, submissions addressed similarly varied themes, including sign language, mathematics, poetry, and Chinese.

* Many groups seem to be grappling with issues that relate to the pedagogically sound use of learning objects. Few papers included clear guidelines or methodologies, or analyzed in any detail what had worked and how or why it worked. It seems as if there is more agreement on the nature and relevance of the questions than on approaches to making progress with answering these questions.

* Similarly, quite a few papers devoted considerable space to a discussion of the exact definition or nature of learning objects as a concept. Authors often complained about the vague or fuzzy definition of learning objects, and then proceeded to give their own definition or interpretation. Our impression is that many of these attempts for a more precise definition are somewhat simplistic and that a more promising approach would be to develop a (set of) taxonomy(ies) for different kinds of learning objects or their components. In fact, this theme could be further developed into component architecture for learning objects (Duval & Hodgins, 2003).

* There were a number of attempts to analyze, in an empirical way, how learning objects are being employed; this pleased us. Often, however, the analyses remained somewhat superficial. It is of course quite difficult to separate issues that are caused by the (not always very satisfactory) state of tools from those that relate to the context of use or from those that are inherent in the notion of learning objects itself.

* Somewhat surprisingly, the number of papers that focused on technical interoperability issues was rather low. Possibly this reflects the somewhat less technical focus of the EdMedia context, or perhaps it suggests that this is not the area where the most pressing problems reside (anymore).

Overall, as individuals who have been involved with some of the efforts related to the development of the learning objects paradigm, we are often struck and encouraged by the diversity of applications that people find for them. On the other hand, we are somewhat concerned by some of the misconceptions and over-hyped expectations that people project onto this field.

A Side Note on Presentation Versus Dialogue

One of the original goals for the Symposium was to emphasize dialogue, by putting more focus on discussion rather than presentation. This proved to be a key factor in the success of this event.

Presenters were asked to assume that attendees had read their paper; all of the papers were made available to the public a few weeks before the event. The guideline was to spend three minutes or less on a quick reminder of the essence of the paper and up to seven minutes to give demonstrations or to relate one's findings to those presented in the other papers. The remaining 20 minutes per slot were reserved for discussion with the audience and other presenters, under the coordination of a session chair.

In fact, we believe that our experience may be relevant for others as well. Many of us participate in conferences and workshops where the presenters "go through the motions," while the attendees are reading the proceedings, the program--or today's paper. Often, little or no time is left for discussion or questions. It strikes us as odd to spend so much time on presentation, whereas the full paper is supposed to present the material in more detail than can be achieved in the typical 20-30 minute presentation.

Often, little or no advantage is taken of the opportunity to demonstrate an application actively or to engage in conversation, dialogue, and discussion. We advocate a style whereby papers are made available in advance, and most of the time is reserved for precisely those aspects that are much more difficult to realize without the benefit of having all the participants at the same place at the same time.

Introducing the Issues

In the remainder of this introduction, we would like to reflect upon how the field of learning objects has evolved and advanced over the last year, i.e., since the Symposium took place. We present a number of themes that we have identified below.

Issue 1: Proliferation of Repositories

Although some educational repositories such as the ARIADNE Knowledge Pool (Forte et al., 1997) started in the 1990s, a significant change--the proliferation and maturation of digital repositories--occurred in 2003 and 2004. As an example, at Ed-Media 2004, there were 19 talks that included digital repositories in their title or abstract; in contrast, there were 14 at EdMedia 2003 and a maximum of seven in any previous year. Factors that have led to the rise of repositories include the following:

* In 2002 and 2003, leaders from organizations such as the Coalition for Networked Information (www.cni.org) (Lynch, 2003) and Association of Research Libraries (Crow, 2002) started to identify institutional repositories as significant development in the management of information and knowledge.

* Concomitantly, projects such as DSpace (www.dspace.org), Zope (www.zope.org), and Fedora (www.fedora.info) started to provide open source solutions and architectures that enable the implementation of digital repositories.

* De facto standards such as harvesting protocols from the Open Archive Initiative (www.openarchive.org) (Lagoze et al., 2002) have started to gain acceptance. These standards allow repositories to enjoy a network effect, thereby adding value to these repositories.

* Initiatives such as Open CourseWare (http://ocw.mit.edu/index.html) have received considerable attention and have helped to promote the very concept of making content available beyond its original context.

* Early federated digital repository projects, such as those launched by the European Schoolnet (www.eun.org), EdNA (http://www.edna.edu.au/), and the US National Science, Mathematics, Engineering and Technology Digital Library (www.nsdl.org) have grown and matured. They now contain hundreds of individual collections of digital educational resources; they also provide "out-of-the-box" solutions that make it easy to add new repositories to the federation.

The growth of educational digital repositories has important implications for the learning object world. One of the questions about learning objects, raised in various forms at the Ed-Media 2003 Learning Object Symposium, is "where are they in the educational and research communities?" The extensive learning object collections that exist in corporations and government agencies owe their existence in part to content and document management systems that already support metadata-enabled organization and searching. The emergence of repositories in the academic and research space provides similar infrastructure, as is reflected in the papers by Kang, Fletcher, and Bradley in this volume. These papers assume that learning objects will be stored and retrieved from a repository. The continued growth of repositories in the year since the Ed-Media 2003 symposium can be viewed as laying the foundation for the development and application of learning objects.

Issue 2: Web Services 

Although Web services were certainly around in mid-2003, they have since become more of a cornerstone to global interoperability. This is most likely true in a general sense for Web technologies, facilitating single logon over different toolsets as an example. It is certainly also true for the technical learning infrastructure.

Early implementation work took place at the Learning Systems Architecture Lab in Carnegie Mellon, where an overall stack of such services was designed (Carnegie-Mellon Systems Architecture Lab, 2004a).

Currently, most of the developments for Web services-based interoperability are framed in the context of the work on a Simple Query Interface (SQI) for learning repository interoperability (LORInteroperability, 2004). A first goal is to enable federated searches across a wide variety of approaches, from client-server to peer-to-peer.

Development of this specification has led to a number of cross-links between existing repositories, including ARIADNE (http://www.ariadneeu.org/), MERLOT (http://www.merlot.org/), EDUCANEXT (http://www.educanext.org/), CELEBRATE (http://celebrate.eun.org/), and others. The explicit aim is that this should help bootstrap the development of a global open infrastructure for learning. Alliances between organizations that have set up and manage learning infrastructures are beginning to be formed on a truly global scale, thereby vastly increasing the number of learning objects available to their communities. Early work on expanding this development beyond search and into the realm of replication is in process.

The ADL initiative has recently started work on discovery and access services, referred to as the CORDRA initiative, for Content Object Repository Discovery and Resolution (or Registration) Architecture. The current CORDRA work relies heavily on the Handle system (Carnegie-Mellon Systems Architecture Lab, 2004b).

Beyond Web services for repositories, some research is being done on the use of Web services to access learning objects directly. This would lead to a shift in focus from the delivery of content to the integration of services. Whereas the former may be more appropriate in the context of passive "documents," the latter may offer more flexibility for building large-scale active components, such as simulations or role playing tools.

Issue 3: De facto not De fad!

Predictions and patterns that emerged from the presentations, papers, and dialogues have continued to play out in reality since the Symposium ended. In particular, and as many of the examples in the papers demonstrated or projected, the number, scale, and quality of Learning Objects has continued to grow substantially. There is now a plethora of repositories of not only Learning Objects but also repositories of metadata to support them; the ability to join this into virtual federated collections and to search across these also exists.

Learning Objects have emerged from their initial phases of definition and experimentation into their current phase of real implementation. The type and depth of questions have changed to reflect this implementation phase and include such issues as the following:

* How does an organization develop a content hierarchy to support Learning Objects?

* How can the policies and reference models for metadata that are relevant to a specific industry, domain, or organization be developed?

* What are the roles, skills, and personnel organizations necessary to work with content in this new modular model of Learning Objects?

* How can learning objectives (terminal and enabling) be developed such that they can be used to help determine the right content to assemble into an effective Learning Object?

* How can Learning objects be designed such that they can be maximized not only for reuse, but also for repurposing and referencing?

* How can the various aspects of a Learning Object, such as the raw content of the assets, metadata, presentation, navigation, and sequencing, be separated and managed?

* How can repositories of Learning Objects and their content assets be created and managed?

* How do Learning Management Systems (LMS) and Learning Content Management Systems (LCMS) work with a true Learning Object approach?

* How do Learning Objects and their management connect with other parts of the organization such as Human Relations, competencies, and technical documentation?

To meet the demand for assistance with these implementation issues, we are seeing many more guidelines. Important early examples include the documents to come out of the Learning Systems Architecture Lab (Carnegie-Mellon Systems Architecture Lab, 2004a) on how to develop Learning Object content and metadata. On a related note, early guidelines are appearing on the new sequencing and navigation functionality provided in SCORM 2004. We are also seeing more implementation assistance, such as templates for design, rules, sequencing, and layout.

Perhaps most pragmatically important is the emergence of tools and technology for implementing Learning Objects that are equally more mature. Just as tools for creating HTML documents matured and moved from raw text editors to true HTML editors and then to their current complete integration (and disappearance) into authoring tools, so too are the tools for authoring, assembling, and managing Learning Objects maturing and improving. This can be seen by the greater development of new and different tools for authoring than those for assembly and management of Learning Objects. This includes technology that directly addresses the instructionally sound design of the assembly and sequencing of information assets to create Learning Objects and larger collections thereof.

Issue 4: Can We Standardize the Standards?

As many papers and discussions noted, the role of standards has been critical to the advancement of Learning Objects. Learning Object Metadata (LORInteroperability, 2004) became a fully accredited IEEE standard; other standards are now emerging from accredited standards bodies such as the IEEE Learning Technology Standards Committee (http://ltsc.ieee.org/) and the CEN/ISSS Learning Technologies Workshop (http://www.cenorm.be/isss/Workshop/lt/).

Perhaps the greatest evidence of the evolution and phase change surrounding Learning Objects was the recent completion of the ADL Shareable Content Object Reference Model (SCORM) with the final release of SCORM 2004. With the inclusion of sequencing and navigation and the updating of the other major components of SCORM, such as IEEE LOM metadata and the run time environment, the technical requirements that SCORM set out to achieve back in 1997 have been completed. SCORM 2004 is thus ready for production level use and implementations have taken off accordingly. As the presentations and papers at the Symposium indicated, Learning Objects are now being used in academic as well as military, government, and commercial settings. They are being applied to everything from manufacturing to medical to metaphysics.

Issue 5: Empirical Validation 

If we want to move the domain of learning technologies from a collection of opinions and beliefs to a real scientific endeavor, then empirical validation of the effects of tools and methodologies is crucial. Early studies are beginning to appear in this area, although they are often faced with severe methodological challenges. Indeed, it is not easy to distinguish between the following:

* The effects of concepts and methodologies (e.g., learning objects reuse);

* The influences of specifications that operate on these concepts (e.g., XML binding of learning object metadata);

* Functionalities made available to the end user in an interoperable way, based on these specifications (e.g., federated search across learning object repositories);

* The end user interface (e.g., an electronic form or some form of information visualization).

Moreover, many of these early studies are somewhat idiosyncratic and are based on a limited number of participants and learning objects. Therefore, it remains unclear how appropriate it is to extrapolate from these early findings. What is certain, however, is that this new focus on end user issues will help to make the learning object paradigm "cross the chasm" and break through into mainstream learning.

Issue 6: The Definition of LOs: No More Questions!

Since the introduction of the term, conversations about learning objects have often started with the observation that there is no universally accepted definition. Gradually, however, this lack of a universally accepted definition. Gradually, however, this lack of a universally accepted definition has become less of an issue. Papers and presentations may still devote a sentence or two to the precise definition of learning object that they use, but it is neither a major discussion point nor impediment. Apparently, there are two reasons for this.

First, there is a fuzzy definition that is becoming increasingly accepted. As Baruque points out in one of the papers that follows, many authors consider a learning object to be "a small 'chunk' of learning content that focuses on a specific learning objective." This is in line with the Learnativity model of a learning object as a resource with a single instructional objective (Wagner, 2002) and is not far from definitions given in other papers in this volume, such as those by Bradley and Koppi. Each of these authors takes time to explain what is meant by "learning object," and the definitions they give are not very different from each other.

Second, there is more recognition that the important aspects of learning objects are how they are implemented and used, not how they are defined. Quoting from the paper that follows by Lusakiak:

     While debate continues about the exact nature and definition of
     learning objects (Hodgins, 2002) and appropriate instructional
     models for their construction (Longmire, 2000; Boyle, 2003), the
     attention of some theorists and researchers has turned to the
     problem of how learning objects can be implemented to create high
     quality educational experiences (Bannan-Ritland, Dabbagh & Murphy,
     2002).

Although it might be comforting to have a crisp definition of a learning object, it is even more comforting to know that researchers are focusing on the practicalities of learning objects rather than the definition. This brings us to the papers themselves.

The Papers

The papers that follow deal with the issues mentioned above in a diverse set of contexts and domains:

* The paper "Learning Theory and Instructional Design Using Learning Objects" by Baruque and Melo presents a methodology for designing courses. The authors advocate the adoption of an eclectic approach to learning theory in the design of instruction using LOs, rather than the choice of one particular learning theory. They present a methodology that incorporates principles from different learning theories. This approach demonstrates the flexibility and adaptability of the learning object approach.

* The paper "The Design, Development, and Use of Multimedia Learning Objects" by Bradley and Boyle analyzes practical experiences with the development and deployment of learning objects in a blended learning context. A pragmatic and simple approach for "useful learning objects" is presented in some detail.

* Dodds and Fletcher present "Opportunities for New 'Smart' Learning Environments Enabled by Next-Generation Web Capabilities." They explore the relationship between learning objects and the Semantic Web. The aim is to enable new learning environments that build on earlier work on Intelligent Tutoring Systems.

* The paper on "Indexing Learning Objects: Vocabularies and Empirical Investigation of Consistency" by Kabel et al. investigates consistency of metadata produced by different indexers, and how that is improved by structured (rather than flat) value lists. The authors advocate an ontology-based approach to increase consistency and improve search effectiveness.

* In "Learning Designer[TM]: A Theory-Based SCORM-Complaint Content Development Tool," Kang et al. present a tool for the production of SCORM-compliant learning objects that are a combination of content and activities.

* Koppi, Bogle, and Lavitt analyze three years of experience with the LRC Learning Objects Repository in "Institutional Use of Learning Objects: Lessons Learned and Future Directions." Their main finding is that the cultures of academia, including the lack of recognition for learning rather than research, seem to be the main impediments to the quick uptake of the learning objects paradigm.

* In "A Framework for the Flexible Content Packaging of Learning Objects and Learning Designs," Lukasiak et al. present their experiences with a tool that augments MPEG-21 content metadata with LOM pedagogical metadata to enable the packaging of learning designs with learning objects.

* In "Integrating Concept Mapping into Designing a Course Management System" by Shyu, Hsieh, and Chou, the authors propose a new algorithm for course design based on the concept mapping technique. Their approach relies on concept maps for the personalization of learning paths.

* Wiley et al. briefly list typical dangers with the deployment of learning objects in "Overcoming the Limitations of Learning Objects." They have developed a methodology to employ learning objects in project-based learning that avoids these dangers.

References

Duval, E., Hodgins, W., Rehak, D, & Robson, R. (2003), Learning Objects 2003 Symposium: Lessons Learned, Questions Asked, June 24, 2003; Honolulu, Hawaii, USA [Online]. Available: http://www.cs.kuleuven.ac.be/~erikd/PRES/2003/LO2003/.

Duval, E & Hodgins, W. (2003). A LOM Research Agenda, WWW2003 Conference, 20-24 May 2003, Budapest, Hungary [Online]. Available: http://www2003.org/cdrom/papers/alternate/P659/p659-duval.html.html.

Carnegie-Mellon Systems Architecture Lab (2004a). Learning and Web services [Online]. Available: http://www.lsal.cmu.edu/lsal/expertise/technologies/learningservices/.

Carnegie-Mellon Systems Architecture Lab (2004b). CORDRA: Technical Introduction and Overview [Online]. Available: http://www.lsal.cmu.edu/lsal/expertise/projects/cordra/intro/intro-v1p00.html.

LORInteroperability (2004) Learning Object Repository Interoperability [Online]. Available: http://www.prolearn-project.org/lori.

Crow, R. (2002). The Case for Institutional Repositories: A SPARC Position Paper. The Scholarly Publishing & Academic Resources Coalition. August 7, 2002 [Online]. Available: http://www.arl.org/sparc/IR/ir.html.

Lynch, C. (2003). Institutional Repositories: Essential Infrastructure for Scholarship in the Digital Age. Association of Research Libraries, February 2003 [Online]. Available: http://www.arl.org/newsltr/226/ir.html.

Forte, E., Wentland, M., Duval, E. (1997). The ARIADNE Project Part i: Knowledge Pools for Computer-based and Telematics-supported Classical, Open and Distance Education. European Journal of Engineering Education, Vol. 22, No. 1. 1997. (Similarly, Part ii, in the same volume).

Lagoze, C., Van de Sompel, H., Nelson, M. & Warner, S., Editors. (2002). The Open Archives Initiative Protocol for Metadata Harvesting, Protocol Version 2.0 [Online]. Available: http://www.openarchives.org/OAI/2.0/openarchivesprotocol.htm.

Wagner, E. (2002). Steps to Creating a Content Strategy for Your Organization. eLearning Developers' Journal. eLearning Guild. October 29, 2002 [Online]. Available: http://www.elearningguild.com/pdf/2/102902MGTH.pdf.

IEEE Standard for Learning Object Metadata (LOM 1484.12.1-2002) [Online]. Available: http://ltsc.ieee.org/wg12.

ERIK DUVAL

Katholieke Universiteit Leuven & Ariadne Foundation

Belgium

[e-mail]

WAYNE HODGINS

Autodesk & Learnativity

USA

[e-mail]

DAN REHAK

Carnegie Mellon University 

USA

[e-mail]

ROBBY ROBSON

Eduworks

USA

[e-mail]