A new trend is shaping the future of educational technology. In the new e-learning economy, buyers and sellers in education and training markets are economically motivated to pursue global commerce opportunities involved in distributed learning over the Internet. International working groups, such as the Aviation Industry CBT Committee (AICC), the Instructional Management Systems Global Learning Consortium (IMS), and the Institute of Electronics and Electrical Engineers (IEEE), are developing standards for web- and computer-based courseware architecture and packaging. The implications of these standards for educational institutions and organizations developing courseware should not be dismissed (Barron, 2000b). For instance, the US Department of Defense (DOD) has selected some of the AICC, IMS and IEEE components for the Sharable Content Object Reference Model (SCORM) that may soon be adopted as the standard for all DOD courseware (http://www.adlnet.org/).

Software standards for digital media and instruction are intended to ensure that courseware will be developed, organized, and distributed in a uniform manner. Some of the reasons why standards are needed are termed the "ilities." They include interoperability among different systems connected by the Internet, accessibility of the content anytime from another location, reusability of content by other developers to save time and money, discoverability of content located in repositories using metadata, extensibility of existing courses due to their modular construction, affordability due to reduced development costs, and manageability of the content by allowing easy changes and updates to small chunks (Computer Education Management Association, 2001; Singh, 2000).

One concept underlying the evolving courseware standards is that the same instructional content may be usable in different instructional contexts. Therefore, instructional content designed as context-independent chunks in an object-oriented programming environment can be shared with other users, recombined with other objects, or redesigned by other instructional developers with reasonable expectations of cost savings (Merrill, 1998; Reye, 1996).

The vision of object-oriented instruction is that an educational object economy will be created. Learning objects will be placed in public repositories for free reuse or in commercial repositories for sale, and these objects can be used as needed by instructors and instructional developers (Koutlis, Roschelle & Reppening, 1999). In the most futuristic vision of this economy, objects will be assembled from repositories "on the fly" to create personalized instruction that satisfies an organization's or learner's specific needs and desires (Fletcher & Dodds, 2000). But, if this vision is to happen, instructional content must be developed as reusable, stand-alone learning objects, and these objects must be tagged with metadata so that they can be located.

What Are Learning Objects?

Learning objects have been described in the open literature in many different ways, depending upon the background of the author and the context of the description. This literature uses a number of terms, including learning objects, information objects, instructional objects, educational objects, content objects, media objects, knowledge objects, and knowledge bits to describe similar concepts. Standards groups very often use a broad description of learning objects because their role is in creating standards and not in describing learning. For example, the Learning Technology Standards Committee (LTSC) of the IEEE uses the term "learning object," to refer to a large collection of various kinds of programming objects. The LTSC defines a learning object as "any entity, digital or non-digital, that can be used, reused or referenced during technology-supported learning...." (http://ltsc.ieee.org).

The IEEE definition of learning objects includes a wide range of content and applications that only have to be object-oriented pieces of a course to be considered to be learning objects. Hill (1997) defines a learning object as a "sub-class of Internet component objects used to construct any interactive program or application...learning objects are identified as special units based on the intention of the author for encapsulating the data or programming object." Examples of learning objects in the IEEE context include media (graphics, audio, animations), instructional components (lessons, modules, units), and instructional software tools, such a chat room or a discussion forum, that are referenced during technology supported learning. In the widest sense, the IEEE definition of learning objects could include learning objectives, persons, organizations, or events." (http://ltsc.ieee.org).

The use of learning objects in this paper is not as broad as the definition provided by IEEE. In our view, learning objects should be small but pedagogically complete segments of instructional content that can be assembled as needed to create larger units of instruction, such as lessons, modules and courses. Learning objects should be stand-alone, and be built upon a single learning objective, or a single concept (South & Monson, 2000). They are not simple information objects, such as a definition or a graph, because they do more than present information--they have a pedagogical strategy and attempt to teach. The design of learning objects in courseware is not new, and many commercial companies such as NetG and Cisco Systems have been using this design model for several years (Barron, 2000a). A review of the best practices of these companies, and the ideas presented in other publications (Downes, 2000; Longmire, 2000) reveals a tentative set of principles and guidelines for designing instruction with learning objects.

Learning Object Design Guidelines

Emerging courseware standards require instructional designers to focus as much on the design of the learning objects composing a course as the overall design of the course. Designers will be asked to create new content or convert existing learning content into small, stand-alone learning objects with metadata so that they can be stored in a learning object repository. It is these learning objects that will eventually be assembled into lessons, modules, or courses by an instructional designer as needed (Centre for Learning Technologies, 2000).

Unfortunately, the emerging standards do not yet provide specific guidance for designers on how to plan for or to create learning objects. But, there are some principles and guidelines available from the open literature that can be use to aid in the design process. These principles and guidelines suggest how to determine the size and instructional content of learning objects, ways they might be given context within an instructional track, and how they should be stored in a repository for future use.

Principle 1: Learning Objects Must Be Units of Instruction That Stand Alone

According to Merrill (1997), providing instruction is different than providing information. Instructional content objects must have appropriate student practice, feedback, learner guidance, and organization of knowledge elements. Their format should allow for more operations than the simple transfer of information. Learning objects should stand alone, and stand-alone learning objects must be independent of instructional context, and use generic information as much as possible. It is in the assembly of learning objects into an instructional sequence that gives context to the instruction itself. Some of the guidelines for designing stand-alone learning objects include:

• Create the content of a learning object to be similar in scope and nature to the content of a typical "lesson" so as to create instruction, not merely information (Downes, 2000).
• Develop the instructional content of a learning object so that it is able to satisfy a single learning objective (Longmire, 2000).
• Develop content objects that are independent of other content, which can be recombined for various outcomes. Each learning object must be able to stand alone so that confusion is not caused by references to previous topics not visibly seen (Centre for Learning Technologies, 2000; Eaton, 1996; Quinn & Hobbs, 2000).
• Use language and content appropriate for a broad audience. Regional terminology or audience-specific humor is not appropriate. It is best to add colorful language or humor later when the objects are assembled in an instructional context by the developer (Longmire, 2000).
• Use consistent language and terminology within a topic area. In the event of adaptive assembly of the learning objects, consistency will prevent learner confusion (Longmire, 2000).

Principle 2: Learning Objects Should Follow a Standard Instructional Format

The current standards movement does not require the use of standard instructional formats.  They are suggested because they facilitate the creation of uniform, quality learning objects with understandable pedagogical strategy, making it more likely that the objects will be accepted and used by a majority of instructional designers. For instance, Cisco Systems has developed specifications that provide standard formats, or templates, for teaching various types of learning objects (Barritt, Lewis, & Weiseler, 1999). Building upon Merrill's ideas about knowledge objects (Merrill, 1997, 1998), and Clark's concepts about recycling knowledge (Clark, 1998), Cisco Systems requires that their learning objects contain content, practice, and assessment items. Cisco's objects vary according to the type of knowledge being taught and the objective of the instruction.

Many instructional design textbooks (e.g., Crowl, Kaminsky, & Podell, 1997) describe how instruction strategy is tailored to what is being taught, and standard formats, or templates, are proposed as a way to ensure uniform design quality. Molenda (2000) defines twelve types of learning, and their corresponding instructional strategies and online delivery methods. For example, Molenda and others (Thiagarajan, 1993) prescribe a commonly accepted format for teaching procedures that includes: a) presenting an overview of the entire procedure, b) demonstrating each step and identifying its critical elements, c) coaching of the trainees as they practice each step, d) requiring the trainees to demonstrate the mastery of each step, e) integrating all steps, and f) providing systematic practice toward fluent application This procedural learning template is based upon sound, research-based learning principles, and more templates for other types of learning are appearing as research and practice progress.

In addition to format, instructional designers must consider the formal architecture of a learning object. Although architectural standards are still evolving, there are trends that can be used by designers now to prepare for the future. For example, the Customized Learning Experiences Online (CLEO) Project (http://www.cleolab.org) is developing specifications for learning objects and their sequencing rules that may eventually be incorporated into the SCORM standard. In the current CLEO model, a learning object is built around a single enabling objective. Content is structured with an instructional template. Using the SCORM model, there can be no interrelationships between learning objects, but CLEO will allow for complex rules for interactions between objects. Still, it will be up to the instructional designer to determine on a case-by-case basis the appropriate template, sequencing, and interrelationships of objects. Some of the guidelines that pertain to the instructional format of learning objects include:

• Use a taxonomy to determine the types of knowledge or competencies that will be trained. These types of learning can be used to standardize types of learning objects (Weisler, undated).
• Select instructional strategies that have been shown to enhance the types of learning and create standard templates for each (Barritt, Lewis, & Wieseler, 1999; Thiagarajan, 1993).
• To design an online course as an ordered collection of related learning objects, build a hierarchy of learning objects based on a set of well-defined learning objectives. If learning objects are to be composed of smaller objects, the instructional design should reflect a hierarchy of learning objects (Downes, 2000).

Principle 3: Learning Objects Should Be Relatively Small

Once the various types of learning objects are defined, another important issue centers on how learning requirements will determine the size or granularity of a learning object. Instructional designers continually ask for guidance on how to create learning objects, and are concerned about their size and their content. While there is no optimal size for a learning object, it has been suggested that they be kept relatively small to increase the potential for reuse (Quinn & Hobbs, 2000), and to facilitate an adaptive, competency-based approach to training (Longmire, 2000). If each learning object is based upon a single enabling objective, and the granularity is small enough, then each learning object will be "appropriately" small.

Stand-alone learning objects could potentially become job aids for an electronic performance support system (EPSS), or branches in a tree structure for an intelligent tutoring system. No courseware standards exist so far that specify the physical size of a learning object, other than the idea that an object address a single enabling objective. Standards bodies are not suggesting specific limits, such as they should be no more than 5 minutes long, or no more than 500 words long. One author suggests that the size of a learning object should be based on a meaningful division of learning that can be accomplished by a student in a continuous effort, that is, one sitting (Jordan, Mann, & Regalado, 2000).

Keeping the size of learning objects small supports flexible instructional design.  For instance, in problem-based learning it is sometimes best to provide problems first, before conceptual material, while other applications suggests conceptual material should be first.  If the instructional material is broken down into learning objects, either path could be programmed (Quinn & Hobbs, 2000).

Size decisions are dependent on the instructional format used to create the object's content, the delivery system, and maybe some other factors. For some instructional designers, design decisions about size and content are driven by specific training requirements. These instructional designers are designing content objects for their own training needs, and reuse and sharing are auxiliary goals. In the near future, however, instructional designers may design generic content that is independent of training context for sale or reuse by other designers. In these cases, reuse, re-purposing, and profit are the primary goals, and minimal learning object size may be more likely to support maximum reuse. Some of the guidelines for the size of learning objects include:

• In the design phase, create specifications for content development. The size of a learning object is partly determined by instructional design specifications (Longmire, 2000).
• Keep the size of content objects small to support flexible, individualized learning. If an instructional design provides alternative sequences of content objects that are instructionally equivalent, a learner will be able to choose a preferred learning path, while the design could preserve a default path that followed a safe and standard approach (Quinn & Hobbs, 2000).
• Some learning objects will become the "assignable units" for a learning management system (LMS). These objects are the smallest units that the LMS assigns and tracks (Jordan, Mann, & Regalado, 2000).

Principle 4: A Sequence of Learning Objects Must Have a Context

When stand-alone learning objects are assembled into a larger unit of instruction, such as a module or course, the sequencing of the objects is a real issue. Context is created by the act of assembly. Normally, an instructional designer develops units of instruction that easily transition from one to the next, and provides a context that threads the units into a coherent, larger whole that makes it meaningful (Mealy & Reeser, 2000). If one or more of these objects are reused by an instructional designer for another purpose, most likely the units will be out of context for the new instructional context if left as is. The instructional designer will have to provide context or give learners the opportunity to provide their own context.

In the future, it may be possible for a learning management system to assemble learning objects dynamically to adapt course materials to the learner's needs. It may be possible to use tailored context wrappers. The LMS may be able to generate multiple wrappers, so that when a learner accesses the learning object, the context that appears will be tailored based on learner attributes and content object attributes (Longmire, 2000). Some of the guidelines concerning the context of learning objects include:

• When assembling several stand-alone learning objects to create a piece of instruction, provide ways for the learner to contextualize the information (Longmire, 2000).
• Orient objects to their original or most likely contexts, and provide cues for learners to apply their own meanings and contexts to the information (Longmire, 2000).
• Add links to the learning object that point to outside context. This way, developers may spend very little time changing the objct and the learner can choose or not (Longmire, 2000).
• When it is possible to tailor the framing of objects in an instructional context, make sure the context frames are divorced from the object (Longmire, 2000).
• Learning object content should not be embedded within formatting, so that it can be re-purposed within a different visual schema without losing the essential value or meaning of the text, data, or images (Longmire, 2000).
• Use a structured markup language with an XSL file to define output format such as font styles, background colors, etc. (Downes, 2000; Quinn & Hobbs, 2000).

Principle 5: Learning Objects Must Be Tagged and Managed

Metadata is structured data about content, and tagging is the creation of the metadata file that is to be placed within a repository. In order to be discoverable and ultimately reusable, learning objects must be tagged with metadata that provide the important and descriptive information about the object. This may include information such as format, size, delivery requirements, authorship, ownership, version number, instructional role, instructional characteristics, and type of interactivity. Within any predetermined metadata schema, a limited amount of metadata can capture the main idea or essence of the learning object in a coherent and unitary fashion (Longmire, 2000). This is usually called the "core metadata" (Quinn & Hobbs, 2000).

Metadata is the key to discovery of existing content in a content repository. Metadata is for the instructional designer who wants to search repositories for existing learning objects, and it is usually written by instructional designers to describe the learning objects they have created. A standard metadata schema utilizing a standard vocabulary for filling in the descriptors or fields within a metadata file makes it possible for a large number of users to understand what a learning object is about without ever seeing it. Metadata may eventually be used by a learning management system to assemble learning objects, and high-quality metadata will be absolutely required in order to assemble the objects dynamically and adapt course materials to the learner's needs (Mealy & Reeser, 2000).

Each organization will have to adopt a metadata schema and tagging rules that are appropriate for the kinds of information that it uses. For example, the DOD is in the process of developing or adopting a standard metadata schema for learning objects that can be used across as many organizations and institutions as possible (Quinn & Hobbs, 2000). DOD is incorporating the best practices from other standards organizations such as the IMS Global Learning Consortium and the Dublin Core to create its own set of mandatory and optional metadata elements. The IMS Best Practice Core set of learning object metadata represents a set of elements that are considered fundamental by the broader learning community for describing learning resources (IMS, June 2000). A second very broad set of metadata is included in the Dublin Core Meta-data Element set (Dublin Core, 1999). The Dublin Core working group has 15 super categories that make up their metadata scheme: Creator, Subject, Description, Publisher, Contributor, Date, Format, Identifier, Source, Language, Relations, Coverage, and Rights (Schatz, 2000).

In some cases, instructional designers may want to create a unique metadata scheme for their own organization that goes beyond the limited number required by the standards. Schatz (2000) emphasizes the importance of using a systematic procedure to produce a metadata scheme that is both usable and within budget. It takes time and therefore money to tag content. Schatz's procedure begins with choosing the training domain that is expected to be addressed by the metadata schema, and listing the people, jobs, locations and skills that make up the domain. Some of the guidelines that relate to the development and use of metadata include:

• Create content objects as segments of knowledge or interactions that can be stored in a database and that can be presented as units of instruction or information (Centre for Learning Technologies, 2000).
• Create the metadata describing the learning object. For learning objects to be used intelligently, they must be labeled as to what they contain, what they teach, and what requirements exist for using them. Metadata is most easily authored by using a form that is appropriate for the type of data being described, e.g., test or multimedia object. (Downes, 2000; IMS Global Consortium, 2000; Quinn & Hobbs, 2000).
• Use a structured markup language such as XML or SGML to represent an object hierarchy. All components of an online course object can then be structured using a system of tags. Tags have a syntax that indicates the name of the field or domain of the tag, and the value attached to that label (Quinn & Hobbs, 2000).
• Use a pre-existing standardized set of metadata elements for tagging your learning objects (Quinn & Hobbs, 2000).
• If creating your own metadata scheme for uses internal to an organization, it is important to develop a set of tags that is not too big. The greatest expense and delay in developing learning objects is assessing, cataloging and tagging the objects (Schatz, 2000).
• Metadata schemas should not be applied to training areas that are constantly evolving, as the time required to build and tag bits may render them obsolete before they come on line (Schatz, 2000).

Conclusions

Architectural and packaging standards are being developed by several industry groups for web- and computer-based courseware, and these standards may be implemented worldwide in the near future. It is important that course designers begin to understand the implications of standards for instructional design, and begin to design their courses in a manner that is consistent with the emerging technical standards. Two of the more important design components of these standards will be learning objects and learning object metadata. Instructional developers can design courseware with these two components today, and be prepared for their mandatory use tomorrow.

While standards will make it possible to locate and reuse digitized instructional material, will “sharing” occur in practice?  How will educational institutions and corporate organizations interpret intellectual property issues?  Today, we can look at the unique approaches of individual organizations to help us predict where we will be tomorrow.  For instance, university environments typically do not have a history of resource sharing, even within departments, yet Brigham Young University has developed a university-wide system for sharing learning objects that is showing increasing participation by faculty (South & Monson, 2000).  The University provides funds to faculty to produce one or more learning objects for their classes and for a university repository.  Faculty receive development tools, are given assistance by a team of professionals in instructional and media design, and are required to adhere to university technical standards.  Funding priority is given to development projects where the resulting media would be useful to many instructors and learners at the university, e.g., high enrollment, general education courses.  However, funding is sometimes given to innovators who are given the freedom to explore while still contributing useful objects to the common digital library. Faculty are eager to contribute their materials knowing that they will have access to the materials of others, eliminating their need to maintain a personal media library.

When learning object systems are viewed as traditionally designed learning objects running on standardized delivery systems, a concern quickly arises that these systems stifle alternative, creative instructional strategies.  Yet just the opposite could be argued.  Digital learning objects are often dynamic and contain highly engaging content, and they can be selected and arranged in a myriad of ways by an instructional designer, or in some cases, by the learner.  Brigham Young University has created technical standards that provide learning objects that can be used in a conventional classroom or as part of an online course (South & Monson, 2000).  Bannan-Ritland, Dabbagh, & Murphy (2000) argue that select learning object systems on the market today can be used as constructivist learning environments and have the potential to capitalize on the goal-oriented nature of human learning processes.  The potential uses and benefits of learning objects are endless and should outweigh their limitations. 

Whether instructors, instructional designers and their home institutions offer their authored materials to a limited collaborative group or, at the other extreme, to the general public, and whether the software is freely shared or licensed for a fee are decisions currently made on a case-by-case basis at individual institutions. Copyright and fair use issues are still under discussion.  In a recent Federal report (Web-Based Education Commission, 2000), commissioners endorsed the U.S. Copyright Office proposal to convene educators and publishers to build greater consensus and understanding of the “fair use” doctrine and its application to online learning.

Instead of being an impediment to good instruction, technical standards can help the designer to create higher quality instructional software that can easily be reused, reassembled, or re-purposed. This will create more available material for faculty and for designers, generate new uses of instructional content, expand the economy, and make the promise of anytime, anywhere learning a reality.

References

Bannan-Ritland, B., Dabbagh, N. & Murphy, K. (2000). Learning object systems as constructivist learning environments: Related assumptions, theories, and applications. In D. A. Wiley (Ed.), The Instructional Use of Learning Objects: Online Version. Retrieved May 2, 2002, from the World Wide Web: http://reusability.org/read/chapters/bannan-ritland.doc

            Barritt, C., Lewis, D., & Wieseler, W. (1999). Cisco systems reusable information object strategy. Version 3.0 Cisco Systems, Inc. Retrieved October 20, 2000 from the World Wide Web: http://www.peer3.com/text/knowledge/ke_whitepapers.html 

            Barron, T. (2000a). Learning object pioneers. ASTD Learning Circuits, March 2000. Retrieved May 7, 2001 from the World Wide Web: http://www.learningcircuits.org/mar2000/barron.html  

            Barron, T. (2000b). Standards: The vision and the hype. ASTD Learning Circuits, November 2000. Retrieved May 7, 2001 from the World Wide Web: http://www.learningcircuits.org/nov2000/standards.html

            Centre for Learning Technologies (2000). The Design, Development and Delivery of Internet Based Training and Education. Fredericton, NB: New Brunswick Distance Education, Inc. Retrieved May 12, 2000 from the World Wide Web: http://teleeducation.nb.ca/content/media/03.2000/ddd-ibte/index.html

            Clark, R. (1998). Recycling knowledge with learning objects. Training and Development, 52, 60.

            Computer Education Management Association (2001). Learning Architecture Learning Objects Overview. Retrieved May 7, 2001 from the World Wide Web: http://learnativity.com/lalo.html

            Crowl, T. K., Kaminsky, S., & Podell, D.M. (1997). Educational Psychology. Dubuque, IA: Brown & Benchmark.

            Downes, S. (2000). Learning objects. Retrieved August 18, 2000 from the World Wide Web: http://www.atl.ualberta.ca/downes/naweb/Learning_Objects.htm

            Dublin Core Metadata Element Set, Version 1.1: Reference Description (July, 1999). Retrieved from the World Wide Web: http://purl.org/dc/

            Eaton, M. (1996). Interactive features for HTML-based tutorials in distance learning programs. Retrieved May 2, 2000 from the World Wide Web: http://www.scu.edu.au/sponsored/ausweb/ausweb96/educn/eaton/paper.html

            Fletcher, J., & Dodds, P. (2000). All about ADL. ASTD Learning Circuits, May 2000. Retrieved May 7, 2001 from the World Wide Web: http://www.learningcircuits.org/may2000/fletcher.html

            Hill, T. (1997). Learning object metadata framework: Requirements Specification, Version 1.1. http://www.geocities.com/ResearchTriangle/6568/metadata.htm

IMS Meta-Data Best Practice and Implementation Guide, Final Specification, Version 1.1. (June, 2000). IMS Global Learning Consortium, Inc. Retrieved September 29, 2000 from the World Wide Web: http://www.imsproject.org/metadata/index.html

            Jordan, D. H., Mann, J., & Regalado, U. (2000). An instructional design perspective: Implementing the shareable courseware object reference model. White Paper, Booz-Allen & Hamilton Inc. Retrieved October 20, 2000 from the World Wide Web: http://www.peer3.com/text/knowledge/ke_whitepapers.html

            Koutlis, M., Roschelle, J., & Reppening, A. (2000). Developing educational software components. IEEE Computer, 32, 50-58.

            Longmire, W. (2000). A primer on learning objects. ASTD Learning Circuits, March 2000. Retrieved August 1, 2000 from the World Wide Web: http://www.learningcircuits.org/mar2000/primer.html

            Mealy, M., & Reeser, S. (2000). <XML>Your Course Here</XML>: The implications of extensible markup language for course development and design. In Proceedings of the 16^th Annual Conference on Distance Teaching and Learning, August 2-4, 2000. Madison, WI: University of Wisconsin System, 277-282.

            Merrill, M. (1997). Instructional strategies that teach, CBT Solutions, Nov/Dec, 1-11. Retrieved August 31, 2000 from the World Wide Web: http://www.coe.usu.edu/it/id2/constncy.htm

            Merrill, M. (1998). Knowledge analysis for effective instruction. CBT Solutions, Mar/Apr, 1-11. Retrieved June 12, 2000 from the World Wide Web: http://www.coe.usu.edu/it/ide/cbtko.htm

Molenda, M. (2000). Selecting pedagogical methods based on learning requirements. In Proceedings of the 16^th Annual Conference on Distance Teaching and Learning, August 2-4, 2000. Madison, WI: University of Wisconsin System, 277-282.

            Quinn, C. & Hobbs, S. (2000). Learning objects and instructional components. Educational Technology and Society, 3(2). Retrieved August 10, 2000 from the World Wide Web: http://ifets.ieee.org/periodical/vol_2_2000/discuss_summary_0200.html

            Reye, J. (1996). Reusability via formal modelling. Position Paper for ITS'96 Workshop on Architecture and Methods for Designing Cost-Effective and Reusable ITSs, Montreal, June 10, 1996. Retrieved May 6, 2000 from the World Wide Web: http://advlearn.lrdc.pitt.edu/its-arch/papers/reye.html

            Schatz, S. (2000). Paradigm shifts and challenges for instructional designers: An introduction to meta tags and knowledge bits. Retrieved September 18, 2000 from the World Wide Web: http://www.imsproject.org/feature/kb/index.html

            Singh, H. (2000). Achieving interoperability in e-learning. ASTD Learning Circuits, March 2000. Retrieved May 7, 2001 from the World Wide Web: http://www.learningcircuits.org/mar2000/singh.html

            South, J. B. & Monson, D. W. (2000). A university-wide system for creating, capturing, and delivering learning objects. In D. A. Wiley (Ed.), The Instructional Use of Learning Objects: Online Version. Retrieved May 2, 2002, from the World Wide Web: http://reusability.org/read/chapters/south.doc

            Thiagarajan, S. (1993). Rapid instructional design. In G. M. Piskurich, (Ed.). ASTD handbook of instructional technology. New York: McGraw-Hill, Inc. Retrieved June 22, 2000 from the World Wide Web: http://www.thiagi.com/article-rid.html

            Web-Based Education Commission. (2000).  The Power of the Internet for Learning: Moving From Promise to Practice. Final report of the Web-Based Education Commission to the President and Congress of the United States, Washington, D.C. Retrieved May 3, 2002 from the World Wide Web: http://www.ed.gov/offices/AC/WBEC/FinalReport/WBECReport.pdf

            Wieseler, W. (1999). RIO: A standards-based approach for reusable information objects. Cisco Systems White Paper. Retrieved July 31, 2000 from the World Wide Web: http://www.peer3.com/text/knowledge/ke_whitepapers.html