DESIGNING LEARNING ACTIVITIES FROM
LEARNING OBJECTS
I.T. Hawryszkiewycz
Faculty of Information Technology
University of Technology, Sydney, AUSTRALIA
igorh@[Link]
Abstract
Learning objects have been proposed as an approach to creating and sharing
learning resources. There are further advantages to be gained if the objects can be
reused. This paper emphasizes the creation of learning environments from reusable
learning objects. It suggests that reuse can be simplified by a learning object
classification, which can be used to semantically construct learning activities. The
paper suggests three major object classes for this purpose. These are environment,
subject and learning method objects. The paper then describes how web-based
learning spaces can be composed from these object classes. An illustration is given
using teaching about technology as an example.
Keywords
Portals, Knowledge Management, Learning Objects, Customization
Introduction
Learning communities are now beginning to take many forms. There are the conventional classroom situations
that still predominate, but increasingly we are beginning to see new forms such as work based and just-in-time
learning, distance learning, and virtual universities. Web based technologies are increasingly used to support
learning in such environments. Early work in this area has illustrated the potential advantages of web-based
support. Wade and Power (1998) for example outlined a number of requirements for computer supported
learning systems and described alternate technologies for supporting learning activities. Neal (1997) has carried
out work on their use in distance teaching emphasizing the delivery of materials. It is however fair to say that
much of this early research has been in specific settings and supported limited activities. The field is now
beginning to recognize the need to develop reusable materials that can be used in different learning contexts.
Three questions appear important in solving this problem. These are:
• How to package materials so that it can be used in different contexts and learning strategies?
• What granularity of packaged material is needed to allow material to adapt to different contexts and
learning strategies? and
• What technology or implementation environment is needed to deliver the materials?
Learning objects have been proposed in answer to the first question. The goal here is to share learning objects
across the InterNet. Standards are now being developed for learning objects. Perhaps the two most quoted
standards are the Dublin core metadata initiative ([Link] and the IEEE Learning
Technology Standards ([Link] These emphasize the definition of a metadata structure that can be
used to describe any learning objects. These standards are very general and describe objects in terms of the
standard elements that can be shared across the WWW. They do not suggest any classification schemes to make
design easier. Other proposed standards are the Netherlands Open University and the Canadian Cancore
Learning Resource Metadata Application Profile ([Link]
Some writers have found the need to classify learning objects to simplify design. Fisher (2001) suggests the
need to define object classes and metadata to describe their combination. This includes separation of subject
metadata and learning metadata as well as metadata to describe the construction of questions from subject
metadata. Douglas (2001) suggests object oriented approaches to design also suggesting the need for a
classification scheme to improve reuse. Two kinds of reuse are also possible. One is the reuse of learning
objects as a whole, which is suggested by most standards. The other is to have finer objects that can be used to
construct new learning activities. As an example, should a learning object be a book or a chapter. If the latter we
can customize a book for a course. Similarly should an object be a whole subject or individual groups of
lectures, so that we can customize subjects.
This paper emphasizes the second class of reuse and defines an object classification to provide a meaningful
way to construct learning activities. It proposes that learning activities be constructed from three object classes –
subject metadata, learning method and environment. The paper proposes Nonaka’s knowledge creation process
(1994) as the underlying theory for defining learning methods. Nonaka’s activities include socialization,
developing an understanding of concepts, articulation of ideas, followed by artifact construction and evaluation.
The paper then describes a system, called LiveNet, which can be used to customize such learning models, is also
described.
This paper proposes a way to classify learning objects to provide a semantic basis for combining them into
learning activities. It concentrates on logical definition of learning activities followed by mappings to an
implementation given a technical environment. Different implementations are possible for the same learning
activity.
Developing a Classification Scheme
In this paper we categorize learning objects into a number of classes. One class learning objects is explicit
information as illustrated in Figure 1. Here information is gradually focused through appropriate subject
classification schemes on a particular learning objective and then used to create the learning object. The learning
object is often a collection of artifacts that satisfy a particular keyword set but is presented in a way suitable for
learning. There is however more to learning than accessing selected explicit knowledge. Learning must take
place in an environment that enables learners to apply explicit knowledge to reach some goal.
Information (Material for education)
Learning Focusing on
object learning
Figure 1 – From information to learning object
Our proposal is to construct learning environments from the object classes shown in Figure 2. Figure 1 is one of
these classes – the subject class. Here the subject metadata describes the material that is to be taught. This sets a
framework for discovery and is usually implemented as links within the subject metadata structure. Thus
teaching for example about databases may place it within the context of businesses or applications. The second
class of object class is the environment, which is made up of the learning goal and support services. The
learning goal defines what is to be achieved. Thus in a University the goal may define assessment procedures
whereas in a project environment the emphasis is on project goals. The learning method is another object class
that defines what learners actually do. The learning activity defines the methods to be used to achieve the goal.
The methods will use any support services provided by the environment. Hence the semantics of building
systems follow the idea “Use the goal to select metadata of subject material and define the learning activities
that define the methods to be followed in leaning about the subject to reach the context goal.”
This classification differs from the structure proposed by Fisher (2001), who includes subject metadata and
learning metadata but includes a metadata definition for generating test questions. Koper (2000) on the other
hand defines units of study composed from subject and learning models. The context in both of these is thus
limited to questions about subject matter. We thus differ in specifically requiring an environment, especially the
learning goal, to be included in the system.
A University
teaching courses
Learning
environment
Course goals and Learning steps Pedagogy
assessment
Support Learning
Learning
system activity
goal
l
goa t for discovery
Tutorials for moderation for emen
iev
Computer support ach Subject
uses Learning metadata
method
Design will require Book
building a model Web content
Figure 2 – Objects classes
It should be noted here that a learning goal can require more that one learning activity. Thus a University
subject assessment scheme can require many activities as can a project task. Learning activities can
themselves be composed of smaller activities. Similarly contexts can be gradually narrowed into smaller
contexts. The paper now explains the three classes of object in more detail.
An example of a subject metadata structure
As an example the paper describes the metadata used as an example here – teaching information technology
to business students. It emphasizes what is commonly known as e-business (Kalakota, 2000) and also includes
a design methodology.
BUSINESS
ROAD MAPS ANALYSIS PRACTICES ORGANIZATION
Process for using Ways of doing
techniques business
STEPS
TOOLS AND SET UP
STRUCTURES
METHODS AND USE
Best practice
examples
DESIGN SEL-SERV TECHNOLOGY COMMERCE
How to setup
technology
EXAMPLES
TECHNIQUES GUIDELINES CLASSES AND BENEFITS
For choosing What to Technology
technologies use to support
METHODS structures
OF USE
Bring activity and How technology
technology together was used
Figure 3 – Structuring the Knowledge
The metadata structure is illustrated in Figure 3. It divides knowledge into seven categories:
• Business practices used in electronic commerce including customer relationship management, supply
chains and so on,
• Analysis to describe ways to analyze new systems and define requirements,
• Design approaches that define the steps needed to design new systems,
• Commercial applications,
• Technologies used in electronic commerce,
• Business services and how to select technologies to provide them, and
• Organizational relationships needed within electronic commerce.
The metadata also contains relationships between these categories to facilitate knowledge discovery. A
learner can begin at one concept and then follow links to related concepts. Thus it is possible to start with a
business practice and then follow links to technology to see what technology can be used to support the
practice. The same objects can be used in different learning activities and contexts.
Apart from the ontology of concepts the body of knowledge also includes exercises and solutions, exams,
case studies and other study material. It can include previous experiences and suggested actions in a business
process step. It can also include guidelines for filling in forms and check-lists for deciding on actions.
Defining the environment
The environment itself is composed of objects and there is the possibility of generilizing contexts for different
kinds of enterprises. The objects may be different depending on the enterprise. In a University they may be
subjects, whereas in a business enterprise they may be project tasks.
Defining the learning method
The third class of object are learning method objects. To indicate what is meant by learning method object in
this paper, we identify number of generic learning goals here called Type A, Type B and Type C for
convenience. Each of these may require different learning methods. These are:
• Type A - Understanding technologies and their application, which is primarily the teaching about the
capabilities of different technologies. This primarily requires application of socialization for explanations
and elaborations about subject material through discussion and exercises,
• Type B – Articulation on the use of technologies in business contexts by developing topic reports that make
hypothesis about how technologies can be used to realize business goals. It primarily involves interpretation
of the way technologies can be used and framing them (Boland and Tenkasi, 1995) in ways that facilitate
interpretation in terms of business objectives and submitting them as topic reports, and
• Type C Learning to use groupware technology, which primarily requires a case study that leads to the
construction and evaluation of a small system.
Our goal is to ensure that we include a complete set of learning methods to support learning activities that
realize these goals. To do this we use Nonaka as the underlying theory to provide guidelines for learning
objects.
Some Underlying Theories For Learning
Learning stresses the importance of interaction within a community. This in turn requires ways to describe
communities and governance structures within them. It also requires ways to describe the learning process itself.
To do this we draw on the ideas of communities of practice and from the work of Nonaka (1994).
Learning Communities
It is now general for most e-learning standards to define roles and their responsibilities in teaching activities
For example, the IEEE Learning Standards Committee defines a variety of roles for this purpose, especially
learners and coaches with different interaction styles for different learning environments. These can be
expanded to include tutors or assistants that work together with the teacher. In more elaborate environments,
there can be owners, experts, novices or apprentices as well as a variety of users. Communities of practice can
also include a variety of experts such as subject specialists to discover, classify and distribute knowledge. In
business environments, they can also be people responsible for specific business process steps. These roles
can be allocated responsibilities and provided with appropriate services to carry out these responsibilities.
Thus the responsibility of the owners may be to create and update the body of knowledge. They can also give
permissions to users to access the portals. Our description of communities is based in a collaborative
metamodel (Hawryszkiewycz, 2002), which predominantly specifies activities in terms of roles with
workspaces with given responsibilities for actions and artifacts.
Nonaka’s Knowledge Creation Process
Our approach is to develop a framework for learning process objects using the work of Nonaka (1994) as
underlying theory. Nonaka sees knowledge sharing and creation following the process shown in Figure 4. These
identify the kind of activities that are fundamental to knowledge management.
Externalization
Socialization Co
un nve
in stru rtin
nd ex form ctur g
n ga n e pli at ed
n i o dg
p lai ting owle cit ion
Ex bora kn s tr in
uc to
ela isting tur
ex es
Ev
a
cre luatin red
ate sto to
d e g n ew in g in
xpl
icit ly m bin data
t
dat
a Co plici rms
ex w fo
ne
Internalization Combination
Figure 4 – Nonaka’s knowledge creation process
Nonaka’s process includes four phases together with their associated environments. These are:
• socialization to bring people together and share their experiences and insights in an area. For example,
this may be a description of subject material in lectures and in informal ways to exchange experiences,
develop trust, share values, particularly applicable in Type A activities,
• externalization, is where some of this captured expertise is framed and interpreted into a form that can
lead to further actions. For example, tacit knowledge about subject material is externalized through
examples and experiments. This requires articulation on ideas using agreed upon terminologies to
externalize them to familiar contexts, usually assignments. The emphasis is on dialoguing, sharing of
mental models, articulation of concepts, development of common terms and requires ways to visualize
interactions, construct artifacts, particularly applicable in Type B and C activities,
• Combination is where ideas generated in externalization are compared to earlier outcomes, as for
example previous assignment solutions, to relate newly created artefacts to previous stored ones,
• The ideas are then combined where necessary with existing information and then applied in practice
during internalization in completing assignments followed by reflection on the outcomes after
assessment.
Any outcomes of any actions evaluated in further socialization and the cycle is repeated. Our goal is to
provide such generic services as reusable objects and provide ways to customise them to particular
application needs.
Creating Learning Activities
We now describe how learning activities can be created using the various classes of objects. The ideas of
Nonaka give a guideline on what is needed in a learning space. The most important paradigm here is that
learning takes place in a space that allow a variety of interactions as well as the ability to reflect on outcomes.
For our example, we need to define a different space for each of the three goals described earlier and yet
integrate these spaces within the same context.
Objects needed to create the learning space
The spaces for learning activities are created from a combination of learning objects Figure 5 illustrates this
combination. It also shows the two main requirements of customization, namely:
• Providing ways to combine the standard subject and the environment and thus relate learning to the
context goal, and choosing the activities suitable for the subject and pedagogy.
Learning Activity
Method Subject
Environment
(Goal, Services)
Socializ Evaluation
ation
Relate learning process to Relate subject to
goal using services provided goal
by the environment
Figure 5 – Object Classification
Components in the learning space
The next question is what should be available in a space. The abstract object structure proposed for learning is
illustrated in Figure 6. It combines Nonaka’s framework and contains components that support the aspects of
Nonaka’s process. These for example, are stories, discussions for socialization, and experiments and
assessments for learning within the environment context. The explicit knowledge is predominantly derived
from standard learning objects.
Environment Subject
Stories
Stories Discussion Explicit Explicit Discussion
Goal about learning
about about Environment Subject about
about the
goal goal Knowledge Knowledge subject
subject
Self Group
learning learning
Discovery Experimen-
Socialization Evaluation
services tation
Learning method
Figure 6 – Generic Structure
There is of course the explicit knowledge to describe the environment goal and the subject matter. Stories in
the diagram predominantly result from the capture of tacit knowledge. A variety of services are then provided
to support learning. Hiltz and Turoff (2002) describe the importance of providing services that match the
learning environment.
Creating the Learning Space
Given the object classification, the following steps needed to build the learning space:
• Identify the environment goal,
• Select the subject matter from the subject metadata,
• Select the context and develop links to the subject matter,
• Specify the pedagogy in terms of learning activities to be followed.
• Specify the method to be used in each learning activity,
• Select the environment services t support the method.
Subject
Goal
object
Define Learning Select Learning Select
Padagogy Services
process activity method method services
(use cases)
Figure 7 – The Design Process
The paper now illustrates these steps using a technical implementation.
Technology
Figure 6 indicates the kind of services that must be provided in learning spaces. The environment includes
technologies that provide a variety of generic services to support the construction of learning objects and
support a chosen pedagogy. Some choices here are described by Benbunan-Fich (2002), who discusses the
general options of supporting synchronous or asynchronous work. These technical choices include discussion,
chatboxes, terminology builders and so on. The technical infrastructure must contain generic functions that
can be easily combined to provide these services. Such generic structures include:
• the community governance structures through its roles,
• supports interaction through discussions,
• provides ways to manage explicit documents, and
• supports group formation for joint study.
Combining the services
The basic components are now used to create a variety of learning services. Many contemporary technologies
often simply automate existing processes but do not provide frameworks for defining learning activities that
support the discovery and contextual services implied by learning objects. Technology in this case should
provide a way to select object structures and place them in context. A number of reference models have been
proposed for technical implementations. We use the model shown in Figure 8, which is closely related to
Koper’s model. Here there are three classes of service namely:
• Creation of subject metadata that is often left to subject matter experts,
• Creation of learning spaces, or what is termed publishing, and
• Delivery, which can include the evolution of the workspaces as learning evolves.
Each set of services supports different roles. Here the subject metadata is developed by experts, and is
published as learning activities by teachers who are free to choose their own pedagogy. It is then delivered to
students who can in change the learning activity. The amount of change permitted can depend on the activity.
Thus considerably more change will be allowed for case study groups than to activities that concentrate on
presentation of material.
Users
Pedagogy
Teachers
Experts Novices
Development Publishing Delivery
Initial Evolving
Components
learning earning
Owners actitity actiitity
Subject Learning
metadata process
metadata
Figure 8 – The portal approach
An Example System
Currently we have been using a system, LiveNet, for this purpose. The approach is to emphasize
collaboration through an entry interface that emphasizes collaboration while providing access to the body
of knowledge. Figure 9 illustrates the basic structure of this interface.
dy F
st u dis ocuse
se To cus d
Ca pic sio
stu ns
die
s
G
man roup
age
men
t
Figure 9 - A LiveNet collaborative services interface
Our mapping here is for a workspace like that shown in Figure 9 to present a learning environment with
individual activities defined by folders that contain both subject matter as well as interaction support.
Roles can be given responsibilities to both change the workspace and the folder. Other folders lead to
different topics for in-depth topic studies. Discussion is also focused on individual learning activities.
Students can also create their own spaces for the case study activities in Figure 4. The workspace also
provides awareness and notification features to alert members to events important to them as needed in
emergent processes. Some details can be found in Hawryszkiewycz (2000).
The interface shows all the information in the subject. It also provides different roles with different views.
Thus for example there is a folder named ‘information-to-tutors’ can only be seen by tutors thus reducing the
need for meetings and saving peoples time.
Accessing Subject Metadata
The body of knowledge is accessed through a knowledge map. Knowledge maps show the concepts and
relationships between them and can be accessed from any system. Thus they can be accessed from a
community space like that in Figure 9. Figure 10 is a simple knowledge map used in this system and useful
for Type A situations. It is a linear list of terms, each of which leads to a concept screen that describes the
concept and a self-assessment screen.
Term
Select topic for
self assessment
Metadata A subject object
Self=1assessment
Figure 10 – Using the subject metadata
On selecting a concept or process step the user is presented with a description and can then follow up with
some self-learning services. With concepts that refer to process steps, they can add to the concept by
recording their experiences and interpretation of step guidelines. It is of course possible at any time to post a
question for further explanation by experts within the community. These can be followed up with questions
and discussions for further interpretation.
Introduction Into Teaching
One of our goals has been to use the technology in ways where it easily diffuses into the process and ensure
that use of technology does not become dominant and detract from the learning process. For this reason the
computer technology was introduced in a gradual way to improve diffusion for the different learning
activities. Following the ideas of Salmon (2000) we first set up community pages where students accessed
information needed in Type A activities to familiarize students with the technology. The next step was to
introduce the design process. Students are organized into groups to discuss design alternatives and make
design choices that dominate Type C activities. Correspondingly students create a project space where such
alternatives can be considered and final decisions posted. Finally there is the prototype development where
students choose technology to implement the design. In the case study students were given a number of
milestones to aim for, starting with analysis, through design specification to setting up a prototype LiveNet
system. Generally, these were successful in the sense that students understood the basic LiveNet interface and
functionality and workspace description and set up prototypes with little effort. The social effect of this is to
require students to pace their work according to the process rather, as is often the case, leaving it to the last
minute. This has an obvious learning benefit although it is perceived as a nuisance by some students in that it
requires them to follow a process.
Summary
The paper described ways of using the ideas of learning objects to create learning activities. These centered
on creating spaces where learning takes place through the gradual construction of a body of knowledge. Three
classes of objects were identified to provide a semantic basis for creating learning activities. The paper
described the structure of such portals and illustrated this with an example.
References
Benbunan-Fich, R. (2002): “Improving Education and Training with IT”, Communications of the ACM,
Vol. 45, No. 6, June 2002, pp. 94-99.
Boland, R.J. and Tenkasi, R.V. (1995): “Perspective Making and Perspective Taking in Communities of
Knowing” Organizational Science, Vol. 6, No. 4, July-August, 1995, pp. 350-372.
CanCore – Canadian Core Learning Resource Metadata Application Profile – [Link]
[September 15, 2002]
Douglas, I. (2001): “Instructional Design based on reusable learning objects: applying lessons of object-
oriented software engineering to learning system design” Proceedings of the 31st. Frontier of
Education Conference, 2001. Vol.3, pp. F4E-1-5.
Dublin Core Metadata Initiative – [Link] [September 15, 2002]
Fisher, S. (2001): “Course and Exercise Sequencing Using Metadata in Adaptive Hypermedia Learning
Systems” ACM Journal of Educational Resources in Computing Vol. 1, No. 1, Spring 2001
Murphy, L.D. (1998): “Digital document metadata in organizations: Roles, analytical approaches, and
future research directions” Proceedings of the Thirty-First Hawaiian Conference on System Sciences,
Hawaii, 1998, pp. 267-276.
Hawryszkiewycz, I.T. (2000): “Knowledge Networks in Administrative Systems“ Working Conference
on Advances in Electronic Government, Zarazoga, Spain, February 2000, pp. 59-75.
Hawryszkiewycz, I.T. (2201) “LiveNet” –
[Link] [15 September 2002]
Hiltz, R. and Turoff, M. (2002): “What makes learning networks effective?” Communications of the
ACM, Vol. 45, No. 4, April, 2002, pp. 56-59.
IEEE Learning Technology Standards Committee – [Link] - [September 15, 2002]
Kalakota, R. and Robinson, M. (1999): e-Business: Roadmap for Success Addison-Wesley, 1999.
Koper, R. (2001): :Modeling units of study from a pedagocical perspective: the pedagogical model behind
EML http:/[Link]/articles [15 September 2002]
Neal, L. (1997): “Virtual Classrooms and Communities” Group 97, Phoenix, Arizona, pp. 81-90.
Nonaka, I. (1994): “A Dynamic Theory of Organizational Knowledge Creation” Organization Science,
Vol. 5, No. 1, February 1994, pp. 14-37.
Salmon, G. (2000): E-Moderating: The Key to Teaching and Learning Online Stylus Publishing, Sterling,
VA.
Wade, V.P. and Power, C. (1998): “Evaluating the Design and Delivery of WWW Based Educational
Environments and Courseware” Proceedings of the 6th. Annual Conference on the Teaching of
Computing, August 1998, Ireland, pp, 243-248.
Copyright 2002 I.T. Hawryszkiewycz.
The author assigns to ASCILITE and educational non-profit institutions a non-exclusive licence to use this document for
personal use and in courses of instruction provided that the article is used in full and this copyright statement is
reproduced. The author also grants a non-exclusive licence to ASCILITE to publish this document in full on the World Wide
Web (prime sites and mirrors) and in printed form within the ASCILITE 2002 conference proceedings. Any other usage is
prohibited without the express permission of the author(s).