Lynn Davie & Jason Nolan

The Ontario Institute for Studies in Education

University of Toronto

© 1999



This paper is about learning constructionist skills in a collaborative virtual environment (CVE). One example of a CVE is the text-based MOO environment; a Multi-user-dimension, Object Oriented. We feel that the MOO is a theatrical space that allows the students and instructors to construct and live out microworlds with educational significance. This paper describes our work with graduate students in education, as we seek to develop a constructionist way of looking at learning and a set of constructionist skills, in order to design information rich collaborative learning environments. We describe the setting, the roles, the performance, and our learnings from working with a CVE for the past three years.


We work with graduate students in a faculty of education. We help them learn to think from a constructionist framework in the design of computer mediated learning environments. The experiences that we would like to share with you in this paper arise from several years of working with students at various levels of background knowledge and experience (Davie et al, 1998). In the past year, we have worked with classes of students whose skills in the use of computers vary from rudimentary, through intermediate classes with some basic skills, primarily in coding HTML, to advanced classes working at various coding levels.

We work in a MOO, called MOOkti, that is specially designed for educators to provide them with a wide sense of what can be done in a CVEs. In this setting, we work to help our students learn to communicate within a text-based virtual environment, to work with the concept of navigable spaces, and finally to build MOO objects, some of stunning complexity. The educational goals are to think about the cognitive and affective educational impact of virtual environments and to learn to construct information rich learning environments. Accordingly, while we are dealing with a synchronous environment, we are going beyond chatting and navigating. And our use of the CVE moves beyond simple discussion to the enactment of constructioNist learning. Papert (1991) defined constructionist learning as adding a social component, a public performative element to constructivism:

Constructionism - the N word as opposed to the V word - shares constructivism's connotation of learning as `building knowledge structures' irrespective of the circumstances of the learning. It then adds that this happens especially felicitously in a context where the learner is consciously engaged in constructing a public entity, whether it's a sandcastle on the beach or a theory of the universe... If one eschews pipeline models of transmitting knowledge in talking among ourselves as well as in theorizing about classrooms, then one must expect that I will not be able to tell you about my idea of constructionism. Doing so is bound to trivialize it. Instead, I must confine myself to engage you in experiences (including verbal ones) liable to encourage your own personal construction of something in some sense like it. Only in this way will there be something rich enough in your mind to be worth talking about.


As we have worked with our students in the past three years, we have come to understand the collaborative virtual environment as a symbolic space (Turkle, 1995), and more importantly, a theatrical space. This theatricality has been noted before by Davie and Inskip (1992). It is a place where we can construct and play out microworlds that allow students and instructors to examine what is and what might be. We invent and learn conventions of social discourse. We capitalize on the ambiguity of the environment to probe into our understandings and our learnings.

Some find our learning environment compelling, most find it difficult at some point in the experience, but all find it has broadened their ideas of what it means to teach online and in real life. Our environment has the characteristics of play and imagination--of invention and exploration. Many of our students find the environment freeing and exhilarating. Some find it confusing and trivial. Yet, we feel that we have been successful in reaching objectives not easily achieved in any other environment, and this sense is something rare and worth nurturing in the development of learning settings.

This paper involves the exploration of this theatrical metaphor, or theatre of metaphor. This prolog sets the stage, developing an understanding of the background and generating a sense of context. Next we present the script, the course activities and challenges we set for the students. Then we will continue by taking up issues around the the roles inherent in our theatre. In performance, our experiences as co-educators will be considered. Finally, we wrap up our theatrical presentation with a curtain call, perhaps leading to an encore. So let us sit back and look at the script.



The metaphor of the script includes such elements as plot, sequence, dialog, action, challenges and scene setting. The script metaphor helps us to discuss our course activities and requirements. First, let's begin with scene setting.

Scene Setting

We work in MOOkti, an educationally focused MOO designed and maintained by an unfunded group of volunteer programmers, independent of direct faculty control. MOOkti is hosted on Noisey, a patchwork computer made of odds and ends connected to the Internet at OISE/UT. As a virtual environment it contains characters and room descriptions of a wide variety of settings, as with any MOO. However, our classes are not the only occupants of MOOkti. In contrast to most collaborative virtual environments, it was intended from the start as a semi-public (as opposed to a semiprivate) environment where our graduate education students must learn amid communities of writers, undergraduates, wanderers, and programmers. However, we have constructed a number of virtual rooms that serve as the home base of our educational experiences, a couple of rooms of our own, in which we can contemplate virtual pedagogy in relative peace. Within our educational wing, called Lynn's Learning Lounge, we have a learning lounge, a lecture hall, and breakout rooms. We also have galleries, sculpture gardens, museums, pubs, and a wide variety of public and private spaces. In addition, we make provision for each student to construct a personal space that they can use for their own educational projects.

Our room descriptions seek to convey a safe space, welcoming and inviting. Descriptions often include humour and subtly that seek to set people at ease. We want to engage the creative, to unlock the exploring child within us. Or for those who do not wish to return to that period in time, the open-minded maturing adult within us. For example, the description of Lynn's Learning Lounge reads:

Lynn's Learning Lounge

The lounge is a pleasant, comfortable room. In the Northwest corner is a stack of pillows that can be used to sit on the floor. Along the west wall is a conference table with comfortable chairs, and in the southeast corner is a cluster of three couches. The beer cooler is quietly masquerading as a filing cabinet.

Obvious exits: out to Chateau MOOkti Mezzanine, down to Exhibition Space, up to Lynn's Lecture Hall, staff to L^3 Staff Wing, east to L^3 elevator, and north to Visual Literacy Gallery Reception


People can access MOOkti from any Java capable browser, using our MOOca and MOOcaLite clients, and our web interface. In addition, one can telnet to MOOkti or use a variety of other MOO clients. At present MOOkti has stabilized at a constant size, as this table shows from the output of the @moostats query command entered at Sun Feb 7 17:53:26 1999 EST.

MOOkti was created 3 years, 10 months, 10 days, 7 hours, 15 minutes, and 12 seconds ago.
A new player is created here every 6 days, 4 hours, 20 minutes, and 51 seconds.
A new object is created here every 9 hours, 4 minutes, and 39 seconds.
MOOkti is home to 228 players and 3,726 objects.


We have a relatively consistent turnover of students, and a fairly stabile core population of community members. Depending on the number of OISE courses offered in any one term, the population fluctuates between about 225 and 350 users.



Our communal plot varies according to the level and goals for each course. However each plot usually contains three levels of constructionist skills that confront students as a series of challenges to ensure that they have the opportunity to develop a shared body of experience in the short 12 week period that students will be working together. Before we can walk, we need to learn to communicate with other actors in the MOO. So the first challenge is always designed to help students learn to speak, to communicate non-verbally, and to use the paging system in the MOO. We learn to see who is online at the same time as we are, to join them and to talk with them. We learn to shake hands and to communicate using a wide variety of social verbs, such as waving, hugging, nodding, etc. We learn to speak and to whisper. We learn to make sense of a synchronous conversation with its relentless scroll of messages. We learn social conventions of when to speak and when not to speak. Some of these lessons are easy and some are hard.

These skills take time to learn. One of the things that we have learned in working with students is to leave enough time for them to develop communication skills in order to provide the foundation that will enable students to move swiftly onto more sophisticated tasks. We generally meet in class once a week, and we schedule a number of challenges in the form of discussion topics, generally over a period of three weeks. At the end of this time, most students feel comfortable in the MOO and can move and communicate with some dexterity. We break classes down into groups of 4-5 students and direct these small groups to discuss topics appropriate to the specific course. We ask recorders to summarize these discussions and to place a report of the discussions on an asynchronous conferencing system.


After three weeks, we feel that it is time to work on skills related to the spatial metaphor of the MOO. In a MOO you move from one space to another. You go places. When you construct your own personal space it is "somewhere" and you go to it and from it. This geographical or spatial metaphor is as essential to the design of a synchronous system as it is important to be able to separate discussions into manageable chunks. If a small group of people could not meet by themselves, the resulting class-sized discussion would have too many participants. We subdivide the online participants into separate rooms in which they can communicate without distraction. If you and I are not in the same room, we cannot see each other's contributions. In order to talk with you, I must move into the same virtual space that you are inhabiting. Most students, in fact, most people in general, are not used to educational computer applications that use a spatial metaphor, so it is important to provide them with activities that help them to develop skills in negotiating not just virtual communication but also virtual topology.

Students learn the commands for moving their characters without too much difficulty. What is difficult is navigating, of knowing where there is to go, why you might want to go there, and how to get there. At present there are 655 rooms in MOOkti. At first glance a simple solution would be to provide a map of MOOkti with indications of how to get from here to there. And in fact, there are maps created by some students and placed in the MOO that can be helpful. However, MOOkti is a user extensible collaborative virtual environment. At any given moment, anyone connected to the moo can decide to be a builder, creating new spaces, extending, modifying changing MOOkti. A map is out of date before it can be completed and posted.

More importantly, our learning objectives for the students are to encourage exploration, of meeting the unknown. We think it is hard to build, to construct your new knowledge without knowing something of what already exists. So we provide hints of where to go, things to see, things to do. But primarily we encourage self-exploration by setting the challenge of a treasure hunt.

First, we divide the students into teams. This is important since we need to develop the social support system that encourages exploration. We see this treasure hunt as a kind of outward bound experience encouraging teams to help each other and to go beyond where anyone in particular would have the time or interest to explore. We ask the teams to find characters (people and bots--programmable robots (Leonard, 1997), to locate spaces, to fetch virtual things. We ask them to map their own virtual universe. This treasure hunt lasts for a couple of weeks, and it is often high energy. We deliberately don't specify whether the treasure hunt is to be competitive among teams or not. We encourage each team to work collaboratively, but leave inter-team dynamics to chance. In some of our courses, the teams have been competitive and in some they have helped each other. We see no difference between the collaborative and competitive teams in terms of the learnings achieved. Both kinds of teams developed a high energy, a high degree of bonding among team members, and not incidentally a good grasp of the virtual terrain.


Finally, we come to the collaborative skills of building and extending our virtual environment, a fundamental aspect of educational CVEs. Our understanding of virtual environments is that they can stimulate learning. Rooms can suggest puzzles, or provide information, or set social conventions that allow people to work collaboratively. Objects can be bots (interactive programs with which you can converse), or serve as slide projectors, or serve as recorders, or be virtual galleries. Objects are linked with web pages allowing for text, graphics, sound, or even video clips. Whatever that can be coded to a web page, can be a part of a MOO object.

For the final part of each course, we set the challenge for the students to construct educationally useful collaborative micro-environments. In the beginning courses, these might be museum displays or static collections of objects that can serve as stimuli to learning. In the intermediate courses, we focus on tutorial constructions. Some of these tutorials have taken the form of galleries linked to hypertextual material with self-testing activities. In the advanced courses, we have built full blown educational simulations. The challenge asks the design teams to work together to create objects that stimulate learning. Note that we focus away from presentational activities toward objects that intrigue, that interest, that engage the learner. It is this switch from the teacher as presenter to the teacher as guide and designer that we hold so important. This is the reason for the MOO.

These learning micro-environments are added permanently to MOOkti. In this way, the work in the courses is not just temporary constructions designed for an instructor, but rather collaborative efforts that add to our mutual collaborative virtual environment.

These three challenges are carefully sequenced. Each challenge is at a level that builds on the skills achieved in the previous level. There is a challenge and effort to expand within each level. And we provide the support for meeting the challenge. We try to help the students know that they are not alone, but that help is a collaborative activity in our MOO. We ask students to help each other, to teach each other and to learn from each other. In addition, there are many different kinds of people on the MOO who can help. You might meet a Wizard (one of the the volunteer programmers) to help you with a particularly difficult problem, or you will almost certainly find one of the authors online and in the MOO to help.



Maintaining our theatrical metaphor, we can specify a number of roles, the cast, that support our production. We have roles equivalent to the director, technical director, drama coach, and actors. As with the roles in a real theatre, each of our roles contribute significantly to the activity and learning of the collaborative virtual environment. Although the names of the roles are familiar, there are special meanings for each role in our theatre.

The Director (aka Lynn)

We can look at the instructor for the course as a kind of director. His role is not a technical blocking of where everyone will stand and how they will deliver their lines. Choices of where to go are mostly the responsibility of the actors as is how the lines will be delivered. Our director looks after the overall setting of the experience. He provides the challenges to the class, encourages, sometimes disciplines, but mostly talks to the actors about motivation. He works with the actors to help them understand the goals behind the activities. He links the experiences to the real world and helps the actors reflect on the meaning of the activity, the depth of their learning.

Our director leads by example. He models behaviour. He sits with those having difficulty. He praises those who have made significant progress. He helps integrate the behaviour in the collaborative virtual environment with our understanding of how adults learn, with the research related to knowledge building communities and other forms of collaborative learning. He is also responsible, finally, for the product, the performance, and the evaluation of that performance. In this sense, he is also perhaps the critic.

The Technical Director (aka Jason)

The technical director is the expert in education in CVEs. He maintains the MOO. He knows how to find things in the MOO. He collaborates with the actors as they try to solve technical problems. But his role is not a technician, for the director of the technical must first be able to direct and provide some level of the actions for which the director is responsible, but primarily when these functions are tied to technical issues. He also has a deep understanding of constructionist learning and theory. He provides links to thinking about the use of collaborative virtual environments. He is instrumental in maintaining the team of volunteer programmers (stagehands, in one sense, and the creators of the theatre itself, in another).

The Acting Coach

We were extremely fortunate to have had several different coaches during our work with the MOO. These individuals worked with students in the laboratory. They helped people learn to use the software, and they provided knowledge and skills about collaborative learning. They were essential emotional and intellectual supports for people challenged by the new environment. However, the role of acting coach goes beyond the specific teaching assistants. Any actor worth her salt will sooner or later accept the mantle of acting coach. This is perhaps the essence of this theatre, that we are all actors, but everyone who stays throughout the performance will end up with a coaching duty to perform.

The Actors

Anyone who takes part in MOOkti is an actor in our metaphor, just as 'All the MOOs a Stage'. Certainly, the students are actors, and the rest of the participants cannot have any efficacious interactions without acting at some point. They have their own motivations for participating. They perform actions, their characters develop over time. Each actor's character has a description and optional special functions. The actors meet each other in the role of citizens of MOOkti. We walk (or navigate) through the the set, talking, hugging, waving, bowing, helping. We try to model the behaviour we wish to stimulate, the roles we have chosen to act out. And finally, the many other people who join our community everyday are unwitting actors in our production; though we may also be unwitting actors in theirs as well. Sometimes these other actors are simply background, but at any moment, they can become a principal actor in our ongoing drama. They might stop to help you learn how to do something having switched into the role of an acting coach, or they might ask you for help. This constant interaction is an important part of making the MOO a dynamic virtual microworld instead of a sterile designed educational simulation. We feel that the variety of actors brings a sense of involvement to each of us who work in MOOkti.



What have we learned through our own experience?

Time in a Bottle

We have learned from our experiences that the virtual experience of time and how we sense time varies from one person to another. This is more than time having a different meaning. Time has a different emotional loading for different people. The constructionist learning we undertake in the MOO requires a relaxation of the stranglehold time has on most of us.

Though learning to make and script a 'bot' may take many hours, the successful animation of your imagination can be extremely satisfying--time well spent. Yet the problem rests with our understanding as a society that time is a commodity. We say that we keep time, we spend time, we waste time. Too seldom do we say that we share time or that we invest time. We find that to truly learn you need to relax the boundaries of our time keeping. If we allow only a short time, we will be able to learn only shallow things. Deep learning and deep meaning require us to suspend our constant accounting of time to follow our puzzle, to seek our understanding.

The commodification of time in education is a function of how learning is governed and regulated by administration, rather than a function of how learners make meaning (Foucault, 1991). The notion of sticking to a task until it is complete is fundamental to a CVE in which you construct knowledge and meaning, as well as in an environment where the learner is having to negotiate the learning curve of a scripted and programmable environment. But our own self governance does not allow us the flexibility to engage in a learning activity that does not have externally observable temporal limitations. We cannot just learn for as long as it takes to understand what we want to understand. For the k-12 student, the length of the period is the limiting factor. With the graduate student the education, career, family, and just being an adult in the modern world leaves little time to embrace a learning moment.

Yet working in a collaborative virtual environment (CVE), especially a user-extensible one such as a MOO, requires almost a meditative, reflective, engagement with the environment. You must spend time, or as our director says, "Invest Time", in order to gain a sense of place, and later purpose, and finally practice. The CVE must be muddled through, puzzled through, and pondered. This requirement is nothing that reality hasn't thrust upon us, but reality gives us time, from the age of 0-5 when school's administration of school begins, to figure out what's going on in reality. But of course, the administration of learning and the self-administration of learning by adults makes no such allowances. MOO/CVEs are no better than reality.

Pacing and patience

Our experience has sensitized us to the need for attention to pacing. In our enthusiasm, at the beginning, we went too fast. What seemed simple for us to do, took more time for our students. While we were eager to get beyond the beginning steps to the things that seem far more interesting, we too often moved to quickly. It took us some time to understand just what a radical shift in perception working in the MOO was for some of the students. If we moved too quickly, anxiety was increased and a sense of powerlessness set in. If we moved too slowly, a sense of frustration was too prevalent. Our solutions are two: First, to design open ended problems so that the design problem could be solved at a number of levels. For the beginner, there was a solution at an elementary level. For the person who had advanced skills the solution could be at a more complex level.. The second solution is to have patience and to monitor what was happening with the students. While we had a broad timetable for the experience, we did modify it as necessary, based on the progress of the students.

Role of the Technical Director

During our work,the need for a technical director (TD) has become clear. We make an important distinction between our concept of technical director and the more usual roles of help desk, or technician, or programmer. There may be a need for all of those functions, but the role of TD has some unique features. First, he or she is the nexus for conceptual development, the intersection between the educational world and the world of programming. Just as the TD in the theatre can bring a visual and spatial reality to a play based on an interpretation of the script, our TD brings a vision of what can be done in the MOO. He engages the actors in an imagining of possibilities not yet seen, and supports the actors as they develop MOO structures that accomplish their creative imaginations. A TD in the theatre has mastered the tools of the theatre, i.e. lights, scenery design, sound systems and can construct an illusion of a reality based on these tools. Our TD also has mastered the tools of the MOO (or knows where to find people who have the mastery) and he can work with the director to bring the illusion of the MOO reality to fruition.

What have we learned from students?

Techn'educators: Oh strange new world, that has such people in it!

We recognized, as did our students, that the primary purpose of human-computer interaction in the 21st century is putting people in touch with other people. Technology is therefore primarily social and a communicative act, not a communications technology. People come first, their fears, joys, learning curves, plans and journeys. The TD's only purpose it to facilitate the manifestation of these personal and social events. Neither a slave to the technology, nor the whim of the user, the TD is first teacher, a director of collaborative virtual learning in this instance, and second a technician whose sole mission is to overcome the limitations that the requirements of technology place on what needs to be done.

The TD must prefer people to machines. This is something that we thought was true, but something that is less visible in the realm of online learning. We players on the virtual stage, as we are performing only for our collective selves, must become our own Technical Directors. We must be able to push the limits of the technology so that we may bring forth the ideas and learnings of our students.

This is not something left in the hands of Technical Support personnel. They should not be dictating what can and can't be done, what software will be installed, when and how. Our students quickly realized that we were rewriting software based on their previous week's comments. That we were not merely users with more experience than they, but actually willing to spend endless hours with volunteer programmers to bring forth the new props and stage implements on which their pageants would stand or fall on opening night. This is, of course, what we had planned, but we had not realized how they would perceive it. What they saw was a new model for learning using technology. Through our agency, the technology was modifying itself to meet their needs. The educators were forcing the learning environment to meet the needs of the user, rather than fulfilling the traditional role of showing the student how to modify his or herself to adapt to the requirements of the technology. We did become co-educators and co-learners, as we learned how to ensure that our technology reflected the needs of our students while students themselves learned how to navigate through the new forms of learning and expression our theatre offered.



As cyberspace and virtual reality moves beyond Hollywood's purview, we need to relocalize it within our notions of learning for the individual and the community. And as computer technology gains in sophistication, we need to stretch our sense of what it means to communicate beyond the duality of synchronous and asynchronous into a polysynchronous experience that more closely reflects real life (Nolan, 1998). The only way in which educators, researchers, and most importantly students write their own acts into the ongoing productions of cyberspacial stage is for educators to camp out on the edges and immerse themselves in what is coming up from the abyss. If the future slips through their grasp and into the grasp of corporate agendas, the virtual community agenda will become a one act template for interaction rather than a fringe festival of learnings and interactions.

We hope that three years of looking at how our students, who are themselves educators, learn and work in MOOs has readied us for the next move forward. In partnership with Industry Canada, Canada's Schoolnet, and the Division of the Environment of the University of Toronto, we are embarking on Project Achieve, a three year experiment and research project that will make a MOO available to secondary level students across Canada. Our goals are to foster a sense of purpose and community through student developed, centred and implemented projects. Our requirements are that students be able to use one of our templates to design any project of their choice, and then we will provide the infrastructural support to help them complete it. But it is up to them to set their own timeline and determine when they have completed the phases of their project. It is our hope that some participants will want to participate in a project that involves learning what it takes to run a MOO of their own.

As participants set up their own theatres of interest, action and invention we hope to learn more about the choices that people make and the goals they set for living, learning and working virtually.



Davie, L., Abeygunawardena, H., Davidson, K. and J. Nolan (1998). "Universities, Communities, and Site Building: Exploring Three Online Learning Systems Virtual University, WebCSILE & MOOkti". Presentation at the 1998 ECOO (Educational Computing Organization of Ontario) conference. Toronto.

Davie, L.and R. Inskip (1992). "Fantasy and Structure in Computer Mediated Courses." Journal of Distance Education VII(2): 31-50.

Foucault, M. (1991). "Governmentality." The Foucault Effect: Studies in governmental rationality. Burchell, G., Gordon, C. & Miller, P. eds. Hertfordshire: Harvester Wheatsheaf.

Harel, I., & Papert, S. Eds. (1991). Constructionism: research reports and essays, 1985-1990 / by the Epistemology & Learning Research Group, the Media Laboratory, Massachusetts Institute of Technology. Norwood, N.J. : Ablex Pub. Corp.

Kent, Phillip (1995). Computers - Constructionism - Constructivism: A brief reading study. http://

Leonard, A. (1997). Bots: the origin of new species, the strange and wild saga of cyberspace's software robots. Toronto: Penguin.

Nolan, J. (1998). "Educators in MOOkti: A Polysynchronous Collaborative Virtual Learning Environment."

Turkle, S. (1995). Life on the Screen. New York: Shuster.