shift happens
TRANSCRIPT
Shift happens
Online education as a new paradigm in learning
Linda Harasim*
TeleLearning Network of Centres of Excellence, Simon Fraser University at Harbour Centre, 7th Floor,
Time Centre, 515 West Hastings Street, Vancouver, British Columbia, Canada V6B 5K3
Received 1 August 2000; accepted 1 November 2000
Abstract
This article addresses that paradigmatic shift. It begins by presenting an overview of the history of
online education as a context and framework for understanding the state of the art today, especially the
use of network technologies for collaborative learning in post-secondary education. Beginning with
the innovations of early pioneers as contributing to the paradigmatic shift, it provides a framework for
understanding this new field. The article then focuses on the Virtual-U, a Web-based environment
especially customized to support advanced educational practices. The Virtual-U research team hosts
the largest field trials in post-secondary education in the world with empirical results and insights
generated from over 439 courses taught by 250 faculty to 15,000 students, attesting to what works in
online education. This article concludes by discussing the signposts to future advances that these data
suggest. D 2001 Elsevier Science Inc. All rights reserved.
Keywords: Paradigmatic shift; Virtual-U; Online education
The proponents of competing paradigms practice their trades in different worlds . . ..Practicing in different worlds [they] see different things when they look from the same point
in the same direction . . .. [B]efore they can hope to communicate fully, one group or the
other must experience the conversation that we have been calling a paradigm shift (Kuhn,
1970, p. 150).
1096-7516/00/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved.
PII: S1 0 9 6 - 7 5 1 6 ( 0 0 ) 0 0 0 3 2 - 4
* Tel.: +1-604-291-5395; fax: +1-604-291-3439.
E-mail address: [email protected] (L. Harasim).
Internet and Higher Education
3 (2000) 41±61
1. Introduction
The invention of the World Wide Web in 1992 made online education increasingly
accessible and allowed new pedagogical models to emerge. Because the Web is easy to use
and capable of presenting multimedia, it expanded the range of disciplines that could be
offered online. The 1980s and 1990s saw enormous innovation and expansion in online
education and networking at all levels of education. The telecommunications and knowledge
revolution enabled greater and faster human communication and collaboration and led to
fundamentally new forms of economic activity that produced the knowledge economy and
required basic changes in education.
The 21st century thus begins with a paradigm shift in attitudes towards online education.
So far the signs are subtle, but the changes will ultimately be profound. Our new under-
standing of the very nature of learning has affected the definition, design, and delivery of
education. It will alter global civilization as educators and learners worldwide adopt and adapt
networked collaborative learning.
This article addresses that paradigmatic shift. Its overview of the history of online
education provides a context for understanding this new learning paradigm. The models
we have now resulted from the efforts of many pioneers, whom this article can reference only
selectively. As all the designers, implementers, and practitioners whose achievements have
created the framework we are now using would attest, in the process of shaping virtual space
into social and learning space, a lot of shift happened!
As Table 1 shows, research, development, and field practice in online learning has a long
history. Nevertheless, many apparently think that online education is very recent, perhaps
born of the invention of the Web. In fact, online education was one of the first progeny of the
invention of e-mail, and its development is intertwined with the history of computer
networking. Remarkable as it may seem (given the exponential rate of adoption), e-mail
communication and computer conferencing began less than three decades ago. These
networked communication systems were the fruit of the research of visionaries who sought
to create broad opportunities for `̀ meetings of minds,'' participatory government, and
interconnected social and cognitive communities (Hafner & Lyon, 1996).
In its vibrant 25-year history, online education has tackled tough questions and developed
various models to try to understand how new methods of learning and teaching can be
effective, exciting, and relevant. But while developments in the 1980s and 1990s prepared for
a revolution in the field of education, most of the noise generated in the media questioned the
value and quality of online education and expressed the concerns of some faculty who felt they
would be displaced by less well-trained staff. Evidently, we need to further promote the ways
this new knowledge about online learning will affect plans for virtual university activities and
explain how online education may claim to be one of the major forces to socialize cyberspace.
Ithiel de la Sola Pool et al. (1984) recognized by the early 1980s that computer networking
would profoundly affect our world.
One could argue that computer communication is one of the perhaps four most fundamental
changes in the history of communications technology. Any such list is, of course,
judgmental, but the case can be made that writing 5,000 years ago, printing 500 years ago,
L. Harasim / Internet and Higher Education 3 (2000) 41±6142
telegraphy 150 years ago, and now computer communication were the four truly
revolutionary changes.
While e-mail remains the major networking application in education, its group commu-
nication counterpart, computer conferencing, is the core of online education as a collaborative
learning environment. Murray Turoff designed computer conferencing to be a `̀ collective
intelligence'' system that would structure group communication for information exchange and
problem solving (Hiltz & Turoff, 1978, p. 43).
Table 1
Computer networks and online education: history and overview of the field
1861 telegraph is invented
1876 telephone is invented
1969 ARPANET begins
1971 e-mail is invented
1972 computer conferencing
is invented
Mid-1970s university courses are
supplemented by e-mail
and computer conferencing
1981 first totally online course
(adult education)� The Source
1982 first online program
(executive education)� WBSI Executive Education (IEIS)
1983 networked classroom model
emerges (primary and
secondary education)
� ICLN: Research Project in four countries� RAPPI: Canada-X-Cultural Project in 5 Countries� 1985: National Geographic Society Kids Network� 1987: AT&T Learning Network� 1988: Writers in Electronic Residence (WIER)� 1989: SITP in British Columbia, Canada
1984 first online undergraduate courses � Virtual Classroom (NJIT)
1985 first online graduate courses � Nova Southeastern University� Connect-Ed (New School of Social Research)� OISE (University of Toronto)
1985 first labor education network � Solinet (Canadian Union of Public Employees)
1986 first knowledge building network � CSILE (OISE)
1986 online professional development
communities emerge� OISE Ontario Educators Online Course� 1990 Global Lab, Lab Net And Star Schools, TERC� 1992 Educators Network of Ontario
1986 first online degree program � Connect-Ed (New School of Social Research)� 1989 University of Phoenix Online
1989 Internet in launched
1989 first large scale online course � Open University (UK)
1992 World Wide Web is invented � CERN (Switzerland)
1993 first national educational networks � SchoolNet (Canada)� 1995 TL�NCE (Canada)� 1998 CL-Net (Europe)
1996 first large-scale online education
field trials� Virtual-U Research Project
2000
L. Harasim / Internet and Higher Education 3 (2000) 41±61 43
Educational adoption of computer networking began in the mid-1970s, following closely
upon the invention of packet-switched networks in 1969 and of e-mail and computer
conferencing in 1971. In the early years, academics and educators had limited access to
computer networks. Nonetheless, many of the scientific researchers involved in early
experiments with ARPANET were also academics, and by starting to link their students
with the larger knowledge community, they introduced e-mail and computer conferencing
into their courses. Coincidentally, they expanded opportunities for student communication,
interaction, and collaboration. Just as e-mail, a technological afterthought to ARPANET's
original purpose of creating a file-sharing system, immediately became the dominant
application, networking changed the means of educational communication beyond what
any had imagined.
In the 1970s, e-mail made possible more generalized educational adoption of computer
networking. It was first used for academic information exchange and then to supplement
university-level courses. By the early 1980s, network communication began to be adopted
by K-12 schools. And indeed they originated an entirely new approach in online
education: the networked classroom, in which teachers and learners launched joint writing
and research projects. Some employed basic methods, such as informal electronic pen
pals, while others developed cross-cultural collaboration projects integrated with their
respective curricula.
One of the earliest examples was the Canadian ReÂseau d'Ateliers PeÂdagogique Pilote
(RAPPI) network (1985±1987), which linked schoolchildren and teachers in over 70
secondary schools in Canada, France, England, and Italy. RAPPI used the computer
conferencing system at the University of British Columbia, Canada, to facilitate information
exchange. The curricular focus was social studies and writing, and through the network
students learned about other cultures, lifestyles, and perspectives. In the process, they gained
increased knowledge about themselves and how they fit into a larger global community. It
was an ambitious launch of network classroom activity, at both national and global levels.
Another approach that emerged in the early 1980s has matured into the Learning Circles
(Riel, 1996). Launched by the InterCultural Learning Network (ICLN) in 1983, the project
first used `̀ sneaker mail'' (Riel's term for mailing diskettes) and then e-mail to link
schoolchildren in San Diego, CA, with peers in Alaska and other American states and
eventually Japan, Mexico, Puerto Rico, and Israel (Levin, Kim, & Riel, 1990). The purpose
of the project was to facilitate joint writing of classroom-based newsletters. What was
especially important, however, was the research component, which explored the effect of
networking on curricular activities. The research question was whether writing to real
audiences on the network improved writing. Controlled studies of cross-classroom collabora-
tion showed an increase in student writing skills (Riel, 1996), and having an audience was
also found to be more motivational than writing for assignments only (Cohen & Reil, 1989).
The first totally online courses began in 1981 with noncredit `̀ mini-courses'' and
executive training programs. One of the first was the online Executive Education program
launched in 1982 by the Western Behavioral Sciences Institute (WBSI). Feenberg (1993)
describes the struggles experienced by the first WBSI-associated faculty who sought to
master the online group environment. No one had ever tried teaching a course totally online,
and, of course, no one had ever studied in an online course. All were proceeding blind,
L. Harasim / Internet and Higher Education 3 (2000) 41±6144
without context, base of knowledge, or precedents. Feenberg describes the faculty process,
based largely on trial and error but with feedback from folks such as Peter and Trudy
Johnson-Lenz, who had in 1978 coined the term `̀ groupware.'' WBSI faculty, some of them
famous orators, were surprised by startling failures. The failures yielded important dis-
coveries that remain relevant: long textual `̀ lectures'' do not work online, nor do Q&A
approaches on their own (1993). Students would not participate, and long virtual silences
ensued. Eventually, the faculty adopted group learning activities, such as discussions, and
history was made.
As access to computers and networks continued to grow, educators recognized that
cyberspace could be shaped for a wide range of uses. They also realized that online
education was a separate field and began to explore how they could enable students to
socialize in this new space. Their activities led to path breaking new approaches for
networked collaboration that helped to define both online education and online collaboration.
They were early precursors to subsequent offspring such as the Computer-Supported
Cooperative Work movement.
From the early 1980s, creative applications of computer conferencing in university
courses have contributed towards the development of a powerful new mode of learning:
online collaborative learning (Bradsher, 1996; Feenberg, 1993; Harasim, Hiltz, Teles, &
Turoff, 1995; Hiltz, 1994; Hiltz & Wellman, 1997; Khan, 1997; Mason & Kaye, 1989). In
1989, the Open University launched the first application of computer conferencing in a
large-scale distance education course; 1500 students had access to tutors and peer discus-
sions online.
The mid-1980s witnessed many other experimental applications online: Solinet emerged as
one of the first wide-scale online labor education programs; a variety of professional
development networks were launched, which created a base for online learning communities
and efforts at online professional continuing education began.
The 1990s were a time of other significant `̀ firsts'' in online educational activities:
national educational networking programs, such as Canada's SchoolNet, a first step in
networking all schools, museums, libraries, and aboriginal communities in the country, is one
example. And major online research programs, such as the TeleLearning Network of Centres
of Excellence (TL�NCE) in 1995, and field testing, such as Virtual-U field trials in 1996,
were also initiated.
The approaching Silver Anniversary of online education is an auspicious time to note its
outcomes as well as its history. To explore the features of the new paradigm of learning, the
article examines how online education has resulted in:
� new modes of educational delivery,� new learning domains,� new principles of learning,� new learning processes and outcomes,� new educational roles and entities.
The first two describe the improved opportunities to participate afforded by online
education. The remaining three emphasize the improved quality of learning.
L. Harasim / Internet and Higher Education 3 (2000) 41±61 45
2. New modes of educational delivery
Three modes of delivery distinguish online education:
� Adjunct mode uses networking to enhance traditional face-to-face or distance education.� Mixed mode employs networking as significant portion of a traditional classroom or
distance course.� Totally online mode relies on networking as the primary teaching medium for an entire
course or program.
2.1. Adjunct mode
Adjunct mode use of computer networks provides the earliest examples of online
education. It originated with the scientists and academics who introduced e-mail and
computer conferencing to their students to expand opportunities for class discussion (Quinn,
Mehan, Levin, & Black, 1983). Adjunct mode typically refers to use of the network as an
enhancement but not a required or graded component of course activity.
Today, the exponential growth in use of the Internet by universities around the world has
resulted in the integration of the adjunct mode into all levels of academic activities. Faculty
and students use the Internet to find resources for their research, to connect with peers and
experts in various specializations, and to remain current with their field through online
journals and newsletters. Participation may be integrated as a formal part of the course
curriculum, or it may simply be encouraged in order that students learn to tap the rich
resources on the Internet.
Computer conferencing systems have also been found valuable for such adjunct mode
activities as electronic office hours and general discussion. Many questions are of interest and
relevance to the whole class, and using a conference for open class discussion avoids
duplication of effort. Often students offer assistance to one another, thus expanding the level
of response to a question; students also use conferences to provide course-related information
or tips for their classmates. Private questions for the professor can be handled by e-mail.
Instructors use e-mail and computer conferencing for administrative procedures, such as
distributing class outlines, supplementary notes, instructions, and assignments and, in some
cases, for administering tests and quizzes. They also use personal e-mail messages to post
grades and provide individual feedback. Many instructors allow and some require students to
submit assignments over the network. Classroom management tools may also be available on
the network.
2.2. Mixed mode
Mixed mode is distinguished from adjunct mode by the fact that networking is fully
integrated into the curriculum. The networking activities constitute a regular part of the
course and form part of the course grade. Mixed mode delivery has many variations. It may
L. Harasim / Internet and Higher Education 3 (2000) 41±6146
be used for one or more major activities in a traditional face-to-face or distance mode course,
such as small group discussions, seminars, and group projects.
The use of online seminars in a face-to-face class is a common application of mixed mode
delivery. An example is an undergraduate course in Communication at Simon Fraser
University, Canada, which uses 6 weeks of a 13-week course for online student-led seminars;
the other 7 weeks are held as face-to-face lectures and tutorials. Online, students participate in
seminars and play two roles: as seminar leader for 1 week and as seminar discussant for 5
weeks. As seminar leaders, students work in small teams to prepare and present the topic,
moderate the discussion (synthesize, focus, and advance the interaction), and summarize that
week's activities and key discussion issues. While face-to-face seminars are often restricted to
small classes of advanced students, the online approach has been valuable even with second
year students, encouraging active engagement in formulating positions, exploring arguments,
and converging on common understanding. Online seminars enable all students to participate,
something not possible in a face-to-face classroom, even with small numbers, and certainly
impossible in large undergraduate classes.
The volume of student participation is quite distinct from traditional classroom courses
where the instructor dominates most of the time. The reverse is true when the activity is
online. The overall volume of student messaging is very high, but it is also fairly evenly
distributed. Most students participate most of the time, each sending several messages each
week. A few students are less regular contributors, sending few or no messages some weeks.
However, rarely does a student not participate at all during an online course.
Another variant of mixed mode delivery, especially in primary and secondary education, is
the networked classroom in which classes in different geographical locations (local or global)
share information and resources, joint projects, and social interaction. Many notable examples
of school-level networks emerged in the 1980s, including the National Geographic Kids
Network or KIDSNet, the AT&T Learning Network, and Canada's Southern Interior
Telecommunication's Project (Teles & Duxbury, 1992). University examples included the
European Campus 2000 (Mason, 1993), and BESTnet and AFRINet that linked university
students and courses in the US, Canada, Latin America, and Africa (Bellman, Tindimubona,
& Arias, 1993).
2.3. Totally online mode
Online courses use the Web or computer networks as the primary environment for course
discussion and interaction. Course activities such as presentation of information, discussion,
and group work are undertaken online (although other media such as textbooks and course
readings in hard copy, audio, or video cassettes, telephone calls, or audio or video
conferencing may be integrated as part of the instructional design). The face-to-face meetings
once considered essential have increasingly been found not to be Ð especially considering
the time constraints and travel costs for students.
The use of computer networking for course delivery is growing for both university and
training courses. Post-secondary institutions began experimenting with the use of computer
conferencing for undergraduate course delivery (either total or partial) in the mid-1980s. In
one of the most prominent of these early experiments, the Virtual Classroom project launched
L. Harasim / Internet and Higher Education 3 (2000) 41±61 47
by Roxanne Hiltz in the late 1980s, computer conferencing was proven to be a viable option
for course delivery. In matched classes in subjects as different as Sociology, Computer
Science, Management, and Statistics, there was generally no difference in outcomes as
measured by exam scores, but students reported that they learned more online (Hiltz, 1994). I
believe Hiltz's project was a key research undertaking that significantly benefited the field in
general. She applied scientific quasi-experimental design processes to yield powerful
quantitative evidence.
In 1985, two graduate schools (the Ontario Institute for Studies in Education (OISE),
University of Toronto) and Connected Education (affiliated with the New School for Social
Research, New York), unbeknownst to one another, each began offering graduate-level
courses online, using computer conferencing as the principal mode of delivery (Harasim,
1987; Levinson, 1990). Students would logon to designated computer conferences to
participate in class discussions, debates, seminars, individual and group assignments, and
virtual `̀ cafes'' and libraries. The online courses that I taught at OISE from the mid- to late
1980s were distinguished by the focus on collaborative learning approaches and a fairly
elaborate conference design to support a variety of group learning approaches.
The design was somewhat daring, especially at 150- to 300-baud modem speeds and
poor connectivity to the remote parts of Canada where some of the students were located.
While the challenges could be the stuff of legends, so were the results Ð at least for an
untested educational application. The collaborative learning approach pioneered in these
courses was highly successful, generating extraordinarily high levels of active and
equitably distributed learner participation, high completion rates, quality work, and strong
user satisfaction.
This and subsequent online courses also yielded results that greatly surprised me. Chief
among these was the fact that students who had a choice (i.e., nondistance education students
living in Toronto, where the courses were offered) might prefer online to face-to-face
delivery. These students reported online education to be more social and convivial than
face-to-face at a time when computer-mediated communications were being dismissed as
`̀ cold'' and `̀ inhuman.'' They also noted superior learning opportunities.
The online courses that I designed and implemented at OISE were graduate level, with a
limit of about 25 students to a course. They began and ended with a set of plenary activities
intended to build a sense of group identity and community. Seminars, small group
discussions, and small group assignments comprised the core curriculum. Each lasted for
1 week or for a set number of weeks. The topics in the first plenary group discussions
included conferences for self-introductions, setting personal and class learning objectives,
and engaging in a `̀ great debate.'' They also employed the following types of group learning
activities: seminars (plenary and small group), dyads (learning partnerships), and project
teams. The 12-week courses were organized into 4 weeks of seminar activity, followed by 2
weeks of a dyad assignment, then there were 4 weeks of project work and class
presentations, and the course concluded with 2 weeks of debates structured around dyad
interaction. Between 35 and 60 computer conferences and sub-conferences were used to
create the environment.
We gathered system-generated usage statistics on the number of messages sent to the
conferences. Analysis of the online credit courses at OISE showed that graduate students
L. Harasim / Internet and Higher Education 3 (2000) 41±6148
posted an average of between 5 and 10 conference messages per week per person (e-mail
notes were not tracked, but these were also prolific).
The OISE online courses were also distinguished by active peer-to-peer discussion and
exchange. Students contributed 85±90% of messages, a level of participation and interaction
high even for face-to-face graduate seminars. The collaborative nature of the conferences is
illustrated not only by the quantity of participation but also by the quality of the interaction.
Analysis of selected contents of the online courses indicates that learners formulated positions
and responded to their peers with active questioning, elaboration, and/or debate. Transcript
analysis of online seminars and small group activities showed that students build on one
another's ideas by posing and answering questions, clarifying ideas, and expanding on or
debating points raised by others (Harasim, 1991). Message map analysis of interaction
patterns in selected discussions demonstrated that students refer to one another's messages,
adding on to the ideas posed and building from them (Winkelmans, 1988). Peer interaction, in
which students are exposed to multiple perspectives on particular topics as well as challenged
by collegial queries about their own ideas, is a valuable opportunity for knowledge building
and developing critical thinking skills (Webb, 1989).
Online interaction thus displayed fewer of the extremes typical of face-to-face class
activity such as excessive or dominating input by a few and little or no participation by
everyone else in the class. Online environments do not entirely eliminate the more vocal
participants. What is new and different is that conferencing ensured that dominance by a few
does not prevent others from having their say.
3. New learning domains
My work in online education led me to realize that this is a new learning domain,
characterized by a unique combination of attributes (Harasim, 1989, 1990). Computer
conferencing and networking enable communication that is best described as a form of
discourse-in-writing. Five attributes distinguish communication in online educational
environments and provide a conceptual framework to guide design and implementation
of online courses:
� many-to-many (group communication),� any place (place-independence),� any time (asynchronicity, time-independence),� text-based (enhanced by multiple media), and� computer-mediated messaging.
E-mail systems, even group mail services, make possible the first four of these
characteristics; however, e-mail systems do not organize many-to-many communication.
Computer conferencing was expressly invented for this purpose (Hiltz & Turoff, 1978) and it
excels as a groupware and enabler of collaborative learning.
Online education is not the same as distance education, although it shares some of the same
attributes. Both are any place, any time, and largely text-based. However, the critical
L. Harasim / Internet and Higher Education 3 (2000) 41±61 49
differentiating factor is that online education is fundamentally a group communication
phenomenon. In this respect, it is far closer to face-to-face seminar-type courses.
Together, the five attributes make online education a unique environment for educa-
tion. In combination, they enable augmented learning environments that can enhance
cognitive activities. Below, we highlight some of the benefits of each of the attributes of
this new environment.
Many-to-many (group communication) enables:� Motivational (socio-affective) benefits of working through problems with peers.� Active exchange: rich information environment.� Identification of new perspectives, multiplicity.� Opportunity to compare, discuss, modify, and/or replace concepts (conceptual
change).� Encouragement to work through differences and arrive at intellectual convergence.
Time independence supports:� Twenty-four-hour access: Users can respond immediately or reflect and compose a
response.� Unlimited air time/access.� Student participation all week, and thereby ongoing knowledge building.� Participation by users at their best learning readiness time.
Place independence allows:� Access to the wealth of Web resources (as well as peers, experts).� Shared interests, not just shared locations among participants.
Text-based/media-enriched messaging encourages and contributes to:� Verbalization and articulation of ideas.� Focus on message rather than on the messenger (reduced socio-physical discrimina-
tion).� Clear expression of ideas.� Rich database/web of ideas.
Computer-mediated environments enable:� Searchable, transmissible, and modifiable archived database.� Multiple passes through conference (discourse) transcript.� Building tools to exchange and organize ideas and support collaborative learning.� Building templates, scaffolds, and educational supports for advanced pedagogy.� Customizing learning environments for all disciplines and evidenced-based
pedagogies.
These five attributes appear to offer almost unlimited methods of presenting online
courses, but they also place certain constraints on designing and managing the online
L. Harasim / Internet and Higher Education 3 (2000) 41±6150
environment. Computer conferencing provides some obvious examples of both qualities.
Presently, it remains the `̀ heart and soul'' of online education, and presumably a discourse
focus will always exist, for education is essentially about interaction, conceptual change, and
collaborative convergence.
Of all networking media available to educators in the 1980s and 1990s, conferencing
systems were the most amenable to instructional design. Conferences can be thought of as
`̀ spaces'' that can be shaped to create an educational forum. Structuring and sequencing the
conferences into an environment that promotes educational interaction demands significant
input by the teacher. Just as in classroom education, the instructor must organize the learning
events according to topic, task, group, and timeline to support group discussions, activities,
and assignments. However, the design requirements of the computer conferencing medium
are different, in important ways, from face-to-face communication.
For example, seminars, discussions, debates, and group projects had to be reconceptualized
to fit within the attributes of the computer conferencing environment. Designs to structure and
organize student input are clearly needed because the asynchronous and place-independent
characteristics create problems of coordination and convergence. Its asynchronous qualities
increase access and expand opportunities for discussion, interaction, and reflection. These
additional opportunities and the motivation of working with a group can also stimulate very
active input by learners. On the other hand, managing group tasks among team members located
in different locations working at different times and possibly even in different time zones and
coordinating the workflows in a compressed time period is a complex organizational challenge.
Some participants may lag behind others or even `̀ disappear'' for a time. Moreover, text-based
discussion can be voluminous, and so it may soon overwhelm even an enthusiastic reader.
Despite the powerful new learning opportunities that became evident in the first decade of
online education, it also became obvious that the networking technology available lacked
elements needed for educational application. Generic networking tools imposed major
overhead costs on the user because they were not specifically designed to support educational
activities (Feenberg, 1993; Harasim, 1993). Instructors had to reformulate their traditional
classroom activities for the asynchronous communication spaces and then also `̀ shape'' the
text-based computer conferences into `̀ virtual'' learning environments. This process involved
administrative, organizational, and pedagogical challenges and costs. Many experiments
failed, and early enthusiasts were discouraged.
Customized educational network environments were urgently needed with tools to:
� provide easy access and navigation,� assist instructors in designing the curriculum for the online environment,� manage the course (calendars, grade books, easy uploading and downloading of
multiple media files),� support cognitive activities that employed message representation, visualization,
annotation, and concept mapping, and� provide group and personal workspaces.
Discipline-specific tools were also needed, especially in those fields that required lab
and studio instruction, such as computer and engineering sciences, health, medicine, and
L. Harasim / Internet and Higher Education 3 (2000) 41±61 51
the cultural and fine arts. In addition to circumscribing the disciplines that could
participate, the lack of multimedia capacity had limited development of tools to support
new discourse patterns. The advent of the Web fuelled an explosive growth in educational
adoption of the Internet, both to augment face-to-face courses and to deliver courses
totally online.
Ironically, the technological solutions provided by the Web also introduced new problems
or exacerbated existing ones. While the invention of e-mail and computer conferencing had
launched an unprecedented level of social interaction, communication, and collaboration, the
invention of the Web led to a phenomenal amount of self-publishing. Two basic models of
online courses thus emerged: one based on collaborative learning and interaction, and the
other based on publishing information online (course materials, lecture notes, student
assignments, etc.).
The second, based on the old model of transmission of information or lecture mode seemed
to flourish during the late 1990s, but then its weaknesses became evident. At the same time,
new tools and environments customized for education based on educational interaction and
collaboration were emerging.
One of the first Web-based environments was Virtual-U hhttp:/www.vu.vlei.comi at Simon
Fraser University. An online environment tailored for a course-oriented approach, it provided
an integrated set of instructor and learner tools to support educational delivery based upon
principles designed to support group learning and teamwork in constructing knowledge
innovation, and solving problems. Software development was research-based and evidence-
driven, and it was iteratively modified according to findings from its field trials.
In 1995, the Virtual-U became one of the key research applications within the TL�NCE
hhttp://www.telelearn.cai. This national program links Canadian researchers and client
communities involved in the development, application, and evaluation of advanced educa-
tion technologies. In particular, the TL�NCE, like Virtual-U, focuses on principle-based
design, that is, educational methods and tools that support collaborative learning and
knowledge building.
4. Principle-based design: the key
I believe the most profound aspect of the current paradigmatic shift in education is the
possibility of shaping that shift. What is distinctive about the communication revolution
going on today is that we are able to debate different perspectives on a major civilizational
change and so have an opportunity to influence our collective future (Feenberg, 1999;
TL�NCE, 1995). Humans have experienced several paradigmatic shifts, but they have
never intentionally shaped them. Today, we have the unique opportunity and responsibility
to engage in designing, at least to some degree, the world that we, and future generations,
will inhabit.
The importance of design, especially principle-based design, is the recognition that
technologies are not just useful for this or that purpose; they construct our worlds (Feenberg,
1999). Virtual space is profoundly social space. `̀ We shape our buildings, and afterward our
buidings shape our lives,'' as Churchill observed in 1960.
L. Harasim / Internet and Higher Education 3 (2000) 41±6152
Some of the features of educational principles that should influence the design, use, and
study of new telelearning environments are discussed below.
4.1. Principle of collaboration in learning
The principle of collaborative learning may be the single most important concept for online
networked learning, since this principle addresses the strong socio-affective and cognitive
power of learning on the Web. The Web's asynchronous nature both enables and requires
collaborative learning: collaboration provides the social glue of a community that engages
learners and motivates them to participate.
Collaboration or co-laboring means working together to accomplish shared goals;
individuals seek outcomes beneficial to themselves and to the other members of the group.
Collaborative learning is the instructional use of small groups so that students maximize their
own and one another's learning. In collaborative learning activities, there is a positive
interdependence among students' goal attainment; learners perceive that they can reach their
goals best if others in the group also reach their goals (Johnson & Johnson, 1989).
Collaborative learning is based on a well-formulated theory that has been validated by
numerous research studies and translated into practical procedures that can be employed at
any level of education.
Discourse is a fundamental aspect of learning. `̀ Learning is enhanced by articulation,
abstraction, and commitment on the part of the learner: instruction should provide opportu-
nities for learners to articulate their newly acquired knowledge'' (Koschmann, Myers,
Feltovich, & Barrows, 1994). Articulation is a cognitive act in which the student presents,
defends, develops, and refines ideas. To articulate their ideas, students must organize their
thoughts and information into knowledge structures. Active learner participation leads to
multiple perspectives on issues, a divergence of ideas, and positions that students must sort
through to find meaning and convergence. Cognitive growth and development of problem-
solving skills depend on epistemic conflict, that is, the collision of adverse opinion (Johnson
& Johnson, 1996). Students encounter opportunities to experience and resolve academic
controversies in the online discourse environment.
Instructional models where faculty `̀ present'' or publish information on the Web are less
engaging and result in higher drop-out rates, as was shown in the earliest WBSI experiments
using well-known speakers whose discourse was presented in textual lectures. Engaging
learners in a cooperative pursuit of knowledge requires new instructor roles. They must learn
to create courses that are constructional or conversational where discourse and teamwork
create a sense of commitment. They must learn to moderate, mediate, and facilitate discussion.
4.2. Principle of access
Lifelong education must enshrine the principle of access so people may be linked
together. As we became aware that we could reinvent many aspects of education by using
the potential access networking provides, we also realized that access is a multifaceted
term. Online education could transcend traditional geographic obstacles to where one could
learn (and where one might teach), but place-based institutions also presented physical
L. Harasim / Internet and Higher Education 3 (2000) 41±61 53
barriers to learners with disabilities. These barriers began to fall swiftly as network access
grew, and by the 1990s, rural areas and many Third World regions began to participate.
Intergovernmental recognition of what many call the digital divide between rich and poor
has initiated calls for action, which new inventions that do not require telephone lines may
help circumvent.
Temporal access was another evident feature, although again we did not immediately
appreciate the remarkable educational benefits offered by asynchronous collaborative learn-
ing. Not only does it enable access to learners with family or employment commitments, the
24/7 access expanded air time for discussion and reflection, allowing everyone to have a
voice, overcoming challenges, and traditional discrimination factors, such as ageism, sexism,
and racism.
4.3. Constructivism and knowledge work
Constructivism as it has come to be known mainly through the works of Jean Piaget asserts
that knowledge acquired by a process of mental construction. Constructivism has become a
synonym for `̀ learning by doing'' (Bereiter, in press).
Knowledge work is work that creates or adds value to conceptual artifacts (Bereiter, in press).
The concept of producing knowledge by collaborating in groups is essential in the new
learning paradigm.
5. New learning processes and outcomes
The emergence of online learning environments and Web-based course activities has
significantly contributed to the recognition that we need a better understanding of the
conditions and means for achieving effective learning. Brown (1990) viewed new educational
environments as part of `̀ the shift from seeing technology as a cognitive delivery system to
seeing it as a means to support collaborative conversations about a topic and the ensuing
construction of understanding.'' A critical question emerged: How can the online environ-
ment support the conversations and shared explorations that form part of the user's active
(co-)production of knowledge?
Studies have suggested a number of new theoretical and methodological ways to under-
stand learning effectiveness in online environments. These are integrally linked to under-
standing how collaboration contributes to learning and under what circumstances.
In 1990, Harasim considered the strengths and limitations of computer conferencing for
online collaborative knowledge construction, noting the need to support three educational
processes from divergence to convergence of knowledge: idea generating (and gathering);
idea linking, organization, and intellectual progress; and idea structuring (and convergence).
The first process implies divergent thinking, while the second and especially the third involve
convergent thinking and information management.
Roschelle (1996, p. 245) presents a complementary perspective, arguing that the `̀ crux of
learning by collaboration is convergence,'' a process of mutual construction of knowledge.
L. Harasim / Internet and Higher Education 3 (2000) 41±6154
`̀ Democratic participation, intellectual progress, and gradual convergence are base attributes
of social inquiry practices that enable scientists to undergo conceptual change. A convergent
account alone suggests the attractive possibility that students develop their concepts in the
course of learning to participate in the practices of inquiry that scientists themselves use to
develop scientific concepts.''
Discourse is key. Bruffee (1999) argues that knowledge is a construct of the community's
form of discourse, maintained by local consensus and subject to endless conversation and that
learning is a social, negotiated, consensual process. Thomas Kuhn (1970, pp. 199±203 and
209±210) similarly argued that scientific knowledge changes as scientists revise the
conversation among themselves and reorganize their relations.
These theoretical formulations on the role of collaborative discourse in conceptual change
and intellectual consensus or convergence provide valuable frameworks for the study of
online learning. Their application to the analysis of data generated by the Virtual-U field
trials may illuminate issues of effective learning processes, outcomes, and conditions. Data
from over 15,000 students who have taken Virtual-U courses from more than 220 instructors
and faculty (involving over 450 courses) are being studied within this conceptual lens.
Below is a synthesis of the new learning processes and outcomes evident in the Virtual-U
research. This research involved collaborative learning approaches and assessment used in
all courses studied:
� consistently high levels of active learning,� high levels of learner engagement and interaction, and� evidence of conceptual change and convergence.
5.1. Collaborative learning designs and assessment strategies
Detailed analysis of 100 Virtual-U courses revealed that 100% incorporated some form
of collaborative learning activity as a significant portion of the course: 26% of the courses
used only discussions, 30% employed discussions plus individual work, 11% used
discussions and group work activities, and 33% employed discussions, group work, and
individual assignments.
Data on assessment strategies similarly reflect the growing significance of collaborative
learning in online courses, as well as the variety of approaches faculty still employ: 48% of
the courses included individual assignments, 21% used group assignments, 25% graded
online participation (involving both quality and quantity of student input), a further 13%
graded participation only in terms of quality of input, and 12% employed exams.
5.2. Active learning/participation
Detailed analysis of the usage statistics of over 50 courses demonstrates a pattern of highly
active students regularly reading and contributing to the online discourse. Analysis of 32
totally online courses shows that 75% of the students logged in at least 10 times per week and
88% logged in at least 5 times per week to read and write messages or to access resources.
L. Harasim / Internet and Higher Education 3 (2000) 41±61 55
Seventy-seven percent of all students in the courses studied posted at least three messages per
week. Distribution of communications is measured by percentage of the total messages
generated by the 50% of students who posted the most. The average distribution of
communications is 72%. Peer-responsiveness is measured by the percentage of total messages
that are reply messages. The average peer-responsiveness was 44%.
Even in mixed mode, student participation and interaction is significant. Analysis of usage
data from 19 mixed mode courses with 28 instructors and 206 students shows that half of the
students logged on 5 or more times per week and that 25% logged on 10 or more times per
week. On average, students logged in 4.5 times and posted two messages per week. The
average distribution of communications for the 19 courses is 71%. The average peer-
responsiveness of the 19 courses was 49%.
The high level of user activity suggests that environments such as the Virtual-U motivate
students to actively engage in discourse.
5.3. New patterns of engagement
Online course activity based upon asynchronous communication yields an entirely
new learning pattern: highly active engagement. In all the courses studied, learners
logged on at all hours of the day and night, and there was no hour when the system
was not being accessed.
Peer interaction is high: typically over 30% of all messages are sent as replies (rather than
as new messages). And approximately 80% or more of the messages are sent by students. The
voice of the professor is not lost, for faculty involvement remains important to students and to
the quality of the discourse. However, what is new is that there is a multiplicity of voices or
perspectives, and students are exposed to a variety of possible interpretations or solutions,
rather than just the `̀ right'' or the `̀ textbook'' answer.
Moreover, the distribution of communications in the online course activities is fairly
evenly spread. Most students are participating most of the time, unlike traditional classroom
situations in which the instructor dominates the airtime with only a few (usually the same)
students having an opportunity to ask questions or add comments.
5.4. Towards a theoretical and methodological framework
The three processes outlined by Roschelle (1996) (democratic participation, intellectual
progress, and gradual convergence) complement those identified by Harasim (1990) in
describing progress from divergent to convergent thinking in online education.
Both resonate with Bruffee's (1999) theoretical position that intellectual convergence
through collaborative discourse is key, and in combination suggest a methodological frame-
work for understanding discourse in online seminars (based on Roschelle and Harasim).
1. Divergent thinking: democratic participation and idea generating� quantitative indicators: usage reports� volume of participation� density of participation
L. Harasim / Internet and Higher Education 3 (2000) 41±6156
� distribution of participation� qualitative indicators: discourse analysis
2. Intellectual progress and idea linking� quantitative indicators: level of interactivity� qualitative/quantitative indicators: name mentioning and transcript analysis
3. Convergent thinking� quantitative indicators: density of (1) and (2)� qualitative indicators: conclusive supported position statements (agree or disagree)
Currently, we are developing and testing a taxonomy for discourse analysis to be applied to
online seminars. Early findings are promising.
Transcript analysis of the discourse in online seminars suggests important new directions
and opportunities for the study of learning processes. The online environment is especially
powerful because it creates a text-based archived transcript of the group discourse, which can
be subsequently studied using quantitative and qualitative research methods.
6. New educational roles and entities
The Virtual-U field trials also yield a tentative profile of some of the new roles and
organizational forms being generated: the virtual professor, the virtual learner, the virtual
course, and the virtual university. Below, we synthesize selected data analyses.
6.1. The virtual professor
The virtual professor is an educator who chooses to teach online instead of or in addition to
teaching traditional classes. Contrary to the general image of online instructors as young and
male, we found that three-quarters of the professors adopting the Virtual-U were senior
faculty and over one-third were female. And the majority taught in the social sciences,
humanities, and health sectors, with a minority in engineering and computing science.
Virtual professors report that teaching online reinvigorated their enthusiasm for teaching.
Instructors said they felt more intellectually stimulated and motivated because their online
students were more engaged with learning and developed a sense of group. Several
instructors reported better outcomes than in face-to-face classes. In their opinion, students
produced better-written and more reflective assignments because others provided a variety of
perspectives. They felt that the conferences helped students grasp more concepts and engage
in more meaningful analyses. In addition, instructors believe that experience online improved
their face-to-face teaching style.
Instructors report major changes in their instructional roles as they move from face-to-face
to online instruction. In particular, instructors and students become more interactive, and
instructors had to learn how to facilitate and engage rather than lecture (and/or entertain).
L. Harasim / Internet and Higher Education 3 (2000) 41±61 57
Therefore, the instructor becomes less of a provider and more of a participant as the students
assume more responsibility for generating input.
The instructors' major complaint was the high initial workload. A typical novice to expert
learning curve was evident as instructors learned to deal with new technology and
expectations. With experience, they became more comfortable. Both classroom instructors
and distance educators reported that the second offering was somewhat easier and that the
third offering required a `̀ similar workload.''
6.2. The virtual learner
Another new entity is the virtual learner, the student who prefers to take all or some
courses totally online. Some choose this mode because they have family or job responsi-
bilities, but many also prefer the enhanced interaction and educational quality that an online
course can offer.
A large majority of the students surveyed reported satisfaction, and they particularly
appreciated the improved opportunities for peer interaction and the increased control over
their time. The major problems students encountered were not related to their workload but
rather to technical difficulties and slow network times. Many also experienced communica-
tion anxiety initially; that is, they were insecure about the appropriateness of the messages
they sent and whether they were sent to the `̀ right'' conference. Feedback and explicit user
expectations helped alleviate student concerns as they gained skills and confidence in
navigating the online classroom.
6.3. The virtual course
An analysis of 439 Virtual-U courses shows that 28% were delivered entirely online
(with no significant classroom or face-to-face component), 69% were delivered in mixed
mode (in which the online component was a significant component), and 3% were
adjunct mode.
These results are interesting because of they contrast with generic Internet tool use.
Whereas the major use of the Internet for classroom use is for adjunct mode, i.e., informal use
of the Internet for research, seeking information or additional resources, the Virtual-U field
trials suggest that the use of a customized educational environment encourages more formal
educational activities online.
Course completion rates in the 64 courses studied are high, with 90% completing the
course and only 10% dropping out. The rate of student completion in the Virtual-U field trials
has to date far surpassed that of traditional distance education offerings and that of online
offerings by traditional distance education institutions.
6.4. The virtual university
A virtual university is not unlike a physical university in its mission or educational
processes. A virtual university offers a wide variety of credit courses that lead to a particular
area of specialization and degree or diploma. It might offer only undergraduate programs or
L. Harasim / Internet and Higher Education 3 (2000) 41±6158
both undergraduate and graduate. The courses are taught by full-time academic faculty who
engage with students to develop their expertise in specific bodies of knowledge and employ
curricula and assessment of the learning outcomes in order to provide the basis for
credentialing. Typically, participation in activities such as discussions, seminars, debates,
simulations, and individual and/or group projects, all mediated through the online learning
environment, characterize a virtual course. Hence, the activities are similar to those in a
physical university course, although they are mediated differently. Students in a virtual
university typically interact with faculty, peers, and curricula through a computer conferen-
cing system, using tools and software and resources, which may be online (readings, video
clips, animations, graphs, spreadsheets, etc.) or in a CD-ROM or simulation software or
offline, such as a course textbook.
The principal difference is student access. Students in the virtual university can gain access
to the course(s) from anywhere, at a time convenient to them and appropriate to the course.
They can engage in the class discourse every day or at least three times a week. Students in
the physical university engage in similar activities and assignments taking part in the class by
traveling to a room in a building by public or private transportation at a specific time and for a
specific time each week.
Nonetheless, the success of virtual universities requires real and perceived academic
quality as well as access. The virtual university should not be developed as a `̀ second class''
or `̀ last resort'' learning opportunity. To meet the criterion of quality programming, a virtual
university must employ top quality faculty and instructional staff, produce high-level
academic curriculum and resources, use group discussion and project activities for most if
not all the coursework, produce demonstrable results, and provide integrated coherent and
cohesive degree/diploma programs, not just assorted courses.
7. Conclusion: Shift happens: online learning matures
The computer communications revolution of the mid-20th century affected all social and
economic realms. Educators were among the first to embrace the revolution, and the
increased educational opportunities and especially the new learning models that have
emerged are now influencing education and society as a whole. At the turn of the 21st
century, public discourse is beginning to recognize the implications of this educational
transformation. There has been a sea change in attitudes, a phenomenal level of investment,
and a frenzy of (often unrealistic) expectations despite the (often glacial) progress in changing
institutional and pedagogic strategies. Online learning is no longer peripheral or supplemen-
tary; it has become an integral part of mainstream society.
The convergence of the computer network revolution with profound social and economic
changes has lead to a transformation of education at all levels. The new paradigm of
collaborative networked learning is evident in the new modes of course delivery being
offered, in the educational principles that frame the educational offerings, the new attributes
that shape both the pedagogies and the environments that support them and that yield new
educational processes and outcomes.
L. Harasim / Internet and Higher Education 3 (2000) 41±61 59
Not all shifts along the tectonic plates create earthquakes of great magnitude, but
eventually they cause the global landmass to change shape. Shift in educational models is
also definitely happening in the same way.
Thus, paraphrasing Kuhn, we can say that `̀ [educators] are looking at the world and what
they look at has not changed. But in some areas they see different things, and they see them in
different relations one to the other. That is why, before they can hope to communicate fully,
one group or the other must experience the conversation that we have been calling a paradigm
shift'' (Kuhn, 1970, p. 150).
References
Bellman, B., Tindimubona, A., & Arias, A. Jr. (1993). Technology transfer in global networking: capacity building
in Africa and Latin America. In: L. Harasim (Ed.), Global networks: computers and international commu-
nication (pp. 237±254). Cambridge, MA: MIT Press.
Bradsher, M. (1996). Making friends in the global village: tips on international collaborations. Learning and
Leading with Technology, 23 (6), 48±50.
Brown, A. L. (1990). Domain-specific principles affect learning and transfer in children. Cognitive Science, 14 (1),
107±133.
Bruffee, K. A. (1999). Collaborative learning: higher education, interdependence, and the authority of knowledge
(2nd ed., pp. 154 and 231). Baltimore, MD: Johns Hopkins Univ. Press.
Churchill, W. (1960). Speech to the House of Commons, meeting in the House of Lords, October 28, 1943. Time
Magazine (September 12).Cohen, M., & Reil, M. (1989). The effect of distant audiences on student's writing. American Educational
Research Journal, 26 (2), 143±159.
de la Sola Pool, I. (1984). Communications flows: a census in the United States and Japan. Amsterdam:
University of Tokyo Press.
Feenberg, A. (1993). Building a global network: the WBSI executive education experience. In: L. Harasim (Ed.),
Global networks: computers and international communication (pp. 185±197). Cambridge, MA: MIT Press.
Feenberg, A. (1999). Questioning technology. New York: Routledge.
Hafner, K., & Lyon, M. (1996). Where wizards stay up late: the origins of the Internet. New York: Simon & Shuster.
Harasim, L. (1987). Teaching and learning online: issues in designing computer-mediated graduate courses.
Canadian Journal of Educational Communications, 16 (2), 117±135.
Harasim, L. (1989). Online education: a new domain. In: R. Mason, & T. Kaye (Eds.), Mindweave: computers,
communications and distance education (pp. 50±62). Oxford: Pergamon.
Harasim, L. (Ed.) (1990). Online education: an environment for collaboration and intellectual amplification. In:
Online education: perspectives on a new environment (pp. 39±66). New York: Praeger.
Harasim, L. (1991). Designs and tools to augment collaboration in computerized conferencing systems. Proceed-
ings of the Hawaiian International Conference on Systems Science, 5, 379±385.
Harasim, L. (Ed.) (1993). Global networks: computers and communication. Cambridge, MA: MIT Press.
Harasim, L., Hiltz, R., Teles, L., & Turoff, M. (1995). Learning networks: a field guide to teaching and learning
online. Cambridge, MA: MIT Press.
Hiltz, R. (1994). The virtual classroom: learning without limits via computer networks. Norwood, NJ: Ablex
Publishing.
Hiltz, R., & Turoff, M. (1978). The network nation: human communication via computer. Reading, MA: Addison-
Wesley Advanced Book Program.
Hiltz, R., & Wellman, B. (1997). Asynchronous learning networks as a virtual classroom. Communications of the
ACM, 40 (9), 44±49.
Johnson, D., & Johnson, R. (1989). Cooperation and competition. Edina: Interaction.
L. Harasim / Internet and Higher Education 3 (2000) 41±6160
Khan, B. (1997). Web-based instruction. Englewood Cliffs, NJ: Educational Technology Publications.
Koschmann, T., Myers, A., Feltovich, P., & Barrows, H. (1994). Using technology to assist in realizing effective
learning and instruction: a principled approach to the use of computers in collaborative learning. Journal of the
Learning Sciences, 3 (3), 227±264.
Kuhn, T. (1970). The structure of scientific revolutions (2nd ed., pp. 150, 199±203, and 209±210). Chicago, IL:
University of Chicago Press.
Levin, J. A., Kim, H., & Riel, M. (1990). Analyzing instructional interactions on electronic message networks. In:
L. Harasim (Ed.), Online education (pp. 185±214). New York: Praeger.
Levinson, P. (1990). Computer conferencing in the context of the evolution of media. In: L. Harasim (Ed.), Online
education: perspectives of a new environment (pp. 3±14). New York: Praeger.
Mason, R. (1993). Computer conferencing and the New Europe. In: L. Harasim (Ed.), Global networks: compu-
ters and international communication. Cambridge, MA: MIT Press.
Mason, R., & Kaye, A. (1989). Mindweave: communication, computers and distance education. Oxford: Pergamon.
Quinn, C. N., Mehan, H., Levin, J. A., & Black, S. D. (1983). Real education in non-real time: the use of
electronic message systems for instruction. Instructional Science, 11, 313±327.
Riel, M. (1996). Cross-classroom-collaboration: communication and education. In: T. Koschmann (Ed.), CSCL:
theory and practice of an emerging paradigm (pp. 187±207). Mahwah, NJ: Lawrence Erlbaum Associates.
Roschelle, J. (1996). Learning by collaborating: convergent conceptual change. In: T. Koschmann (Ed.), CSCL:
theory and practice of an emerging paradigm (pp. 209±248). Mahwah, NJ: Lawrence Erlbaum Associates.
TeleLearning Network of Centres of Excellence (1995). TeleLearning NCE: building a knowledge society. Ap-
plication proposal to the Government of Canada, Networks of Centres of Excellence Program, Ottawa, Canada.
Teles, L., & Duxbury, N. (1992). The networked classroom: creating an on-line environment for K-12 education.
Burnaby, BC: Faculty of Education, Simon Fraser University.
Webb, J. (1989). Peer interaction and learning in small groups. International Journal of Educational Research, 13
(1), 21±29.
Winkelmans, L. C. (1988). Educational computer conferencing: an application of analysis methodologies to a
structured small group activity. Unpublished MA thesis, University of Toronto.
L. Harasim / Internet and Higher Education 3 (2000) 41±61 61