Understanding the Lifecycles of Network-based Learning Communities

James Levin

Raoul Cervantes

Introduction

More than half of the classrooms in the United States are wired to the Internet and the number of classrooms connected is rapidly increasing (NCES, 1999). As this network infrastructure is put in place, teachers and learners can form and participate in network-based learning communities. But for these communities to function in productive ways we need to better understand how these communities are formed, grow, function in some mature steady state, and decline and terminate. A better understanding of this "lifecycle" allows teachers and learners to better function in these network-based learning communities, and permits the development of institutional structures that more appropriately support learning and teaching in these new media.

In this chapter, we review studies of network-based learning communities, especially those communities formed around collaborative projects, and will present evidence for systematic patterns of change in these communities over time. Such communities are born, undergo growth, reach a level of mature functioning, and then undergo decline and cease to function. Like biological organisms, this lifecycle can be truncated when the community is not properly supported or when external factors intervene in some traumatic way. We describe the "lifecycle" of network-based communities by examining in depth an extended case study of a network-based learning activity. We conclude with a discussion of the kinds of support needed to encourage the growth and mature functioning of productive network-based learning communities.

There have been a number of pioneering efforts to explicate the nature of network-based learning communities. Three have described specific time sequence that network-based projects typically experience. These include Levin, Waugh, Chung, & Miyake's (1992) and Waugh, Levin, & Smith's (1994a; 1994b) description of 6 stages of organizing network-based instructional interactions, Riel's (1993) 5 steps in a Learning Circle (forming the Learning Circle, planning the Learning Circle projects, exchanging work on the projects, creating the publication, and evaluating the process) and Harris’ (1995) 8 steps in organizing telecollaborative projects (choose the curricular goal(s), choose the activity's structure, explore examples of other online projects, determine the details of your project, invite telecollaborators, form the telecollaborative group, communicate, and create closure).

How many "stages" are there in the lifecycle of network-based community? In the case study to be described here, it is clear that there is a continuum of development without discrete boundaries, where activity in one stage flows into another. The critical issue is the concept of a lifecycle, with the named stages as a convenience for keeping track of the different sections of an otherwise continuously changing process. The issue of whether there are six stages, five stages, or eight stages may turn out to be less important than the idea that learning communities go through a lifecycle, and that the nature of the interaction early on can be quite different from the nature of the interaction later.

For learners and teachers to engage productively in these activities, it is important to understand the ways that these activities systematically unfold over time. An understanding of the lifecycle of network-based communities can lead to more productive learning by communities of learners distributed across the world. This understanding can help create powerful learning environments for diverse sets of learners in ways that may lead help people become better able to deal with the challenges of today and tomorrow.

Network activity is episodic, unfolding over time through a series of different phases. The exact list of steps or stages specified varies, but there is generally some sort of initiation phase, a phase in which the educational activity is carried out, and then some sort of wrap-up phase. Following Levin et al. (1992), this study examines six stages in the network project life-cycle: proposal, refinement, organization, pursuit, wrap-up, and publication.

Levin et al. (1992) describe the proposal stage of a network activity life cycle as:

They described the refinement stage as follows:

The organizational stage is described as follows:

They described the pursuit stage as follows:

The wrap-up stage is described as follows:

They described the publication stage messages as:

The episodic nature of educational network interactions is important for several reasons. Since the nature of the interaction in these different phases varies, the roles that participants in the activity need to play for differ as well. Unless participants are aware of the ways in which these interactions unfold, they may be disappointed in their expectations about the timing or nature of interactions. In addition, knowing about the nature of these network processes allows the participants to integrate them more effectively with the other educational activities in which they are engaged.

One of the ways that educational network interactions differ from other comparable face-to-face interactions is that the network interactions can be stretched out over time. Riel describes them as "...group conversation carried over electronic mail in slow motion" (Riel, 1993). This time elongation is surprising to novices, who see electronic networks as enabling communication at the "speed of light." Compensating, at least partially, for this is the fact that networks allow one to participate in several such interactions at the same time. Even though a given learning activity is stretched out in time, the same group of learners can participate in many such activities over the same time period.

Stapleton (1991) conducted a study of 18 different network-based projects in order to map out the stages that they went through. He found that the further along a project reached before ending, the more successful it was judged by the participants. In cases where a project was proposed but never received responses, the proposer was left very unsatisfied and very little learning occurred. Even for projects that were initiated and conducted, the participants judged the project as less successful if there was no closure on the project than if there was closure. Because projects are stretched out over such a long time, participants often have the impression that not much has happened until they write or read a "wrap-up" summary of the project. At that point they then realize that quite a bit of learning occurred, stretched out over weeks or months of the project’s duration.

Following up on this research, Cervantes (1993) conducted an in-depth study of one network-based project, the "Zero-g World Design Project," to explore these processes in detail and to describe the relation of network activity to face-to-face classroom activity. This project was conducted during the 1991-1992 school year. Many of the pre-college participants used the FrEdMail network, a low cost electronic network with email and electronic bulletin board features that allowed precollege schools to participate in educational network activities before the Internet was widely available to them.

The Zero-g Project was a year-long project in which participants designed activities for a zero-g environment, such as exists in the Space Shuttle or in current space stations. Zero-g Project activities included the Collision Course Challenge, in which students generated solutions for a problem involving two people on a collision course in zero-g; the Design Challenge, in which students addressed the problems of recreation and food in zero-g; the School Design Challenge, in which students designed a school for zero-g; and the Five Same and Different Challenge, in which students proposed five differences and similarities between conventional and zero-g schools.

The designers and coordinators of the Zero-g Project attempted to create functional learning environments to provide a wide range of instructional opportunities for the participants in this study. The Zero-g Project and the activities it generated then provided a context for mediation. Within this context, experts assisted novices to achieve goals specified by the project.

The message marking the Proposal stage of the Zero-g Project was posted on an electronic bulletin board in the summer prior to the school year in which the project was conducted. The project director sent the proposal message, which provided a context and purpose for the project, its goals, the diversity of participants, and applications for instructional settings.

The following message excerpt informs potential participants of the context and purpose of the project.

This opening excerpt introduces ideas that were central to the Zero-g Project. First, the project directed participants to create designs for space station environments, like SkyLab. And second, the fact that people lived in "freefall", where objects did not fall, was central to the participants' thinking throughout the project and provided a constant constraint for participants when solving problems and creating designs. These two concepts included in the proposal message influenced work throughout the entire project.

This section of the message also relates the notion of zero-g to a potential real world condition. The concepts of gravity and zero-g are made less abstract by situating them in the context of environments where people will live. This notion that zero-g is a condition that affects people and objects in life supporting environments also influenced work throughout the course of the project.

A crucial component of the Proposal message is the statement of project goals. The message focuses on project goals in the task of designing or redesigning aspects of everyday life to be consistent with zero-g conditions. The linking of everyday life aspects to the novel environment of a space station gives the participants a point of reference and requires them to use what they know to work with it. Throughout the project, which eventually issued a variety of problems and challenges to the participants, tackling everyday aspects of life in zero-g remained consistent. While this project has its parallels in the Cognition and Technology group at Vanderbilt's Jasper project, it differs from that by engaging learners in interaction with other people from both within the educational system and outside, instead of just engaging learners in a synthetically constructed learning environment.

To achieve continued and successful participation in the Zero-g Project, participants, particularly teachers, required a range of resources. The project director anticipated this and included information on resources in the Proposal:

This section of the proposal mentions the participation of experts and availability of resources for use in the classroom. It also sets up the future interactions on-line by naming the players and proposing information to be shared. These resources were crucial to the project and proved influential throughout. The teachers in this study had limited knowledge of life supporting space environments and had never taught in this particular content area. All of them expressed reliance on outside experts to assist in providing knowledge to be used by their students and by the teachers themselves.

The videotape offered in the message proved to be useful to the participants in this study throughout the project, and was perhaps the most robust and useful single teaching tool. By viewing the video, the participants learned about the physical appearance of space environments, and how objects and people are affected by zero-g.

Teachers involved in this study expressed concern about how the Zero-g Project would support their curriculum. This concern was also expressed by teachers who responded to the Proposal message. This message excerpt addresses this issue:

At first glance, the Zero-g Project would give the impression of a narrowly defined science project. However, as the project director stated in this message, the project engaged students in a range of skills and knowledge domains, giving the teacher the flexibility to focus on pertinent areas.

By definition, the Proposal phase occurs when the network project is introduced to the potential participants, primarily through a network message. The Proposal phase, although consisting of a single message, is critical to any project. Only some proposals attract participants and continue on to the following lifecycle stages (Levin et al., 1992). The Zero-g Proposal message initiated a project that extended through all of the lifecycle stages. Also, the initiatives in this message, goals, purposes, resources, and curriculum relevance, affected project work through the entire lifespan of the project.

Of central importance for the success of a proposed project is the readiness of participants to engage in the project. Of the many teachers that volunteered to participate in the Zero-g Project, four were selected for intensive study. All four teachers had some instructional telecommunications experience. The high school teacher, with the most experience among the four, had taken a course in networking at the University and had participated in network projects for four years. The second grade teacher also had taken a course in telecommunications, had one year experience with network projects, and had been a systems operator of a K-12 electronic network for a year. Our research team had worked with the fifth grade teacher the year prior to this study, participating in a network project. The middle school teacher had taken a course in instructional telecommunications seven years prior to this study, participated in an e-mail pen-pal project at that time, but had not used telecommunications since then with her students. Even though they expressed uncertainties about the likelihood of success in their interviews, these doubts did not outweigh their motivation and beliefs that the project would clearly benefit their students.

For several weeks after the project director posted the Zero-g Project proposal message, interested teachers and other network users sent messages inquiring about the project. A primary concern was whether the project was appropriate for the teachers' students, as this message states:

Another message expressed a similar concern, inquiring whether their participation could be a useful contribution to the project:

The project director made clear that the Zero-g Project was not limited to experts or even science classes, but was appropriate for participants who possessed a wide range of interests and abilities. This message excerpt in particular underscores this point:

The project director was attempting to recruit not only teachers and students, but also experts working in business or government institutions. Because collaboration between outside experts and teachers and students was new to most of the participants, their roles were a matter of uncertainty. This message exchange, between a NASA scientist and the project director, reflect this concern:

This message exchange illustrates qualities of the Zero-g project that may have appealed to experts working in fields outside education who have been concerned about their role, expectations, and demands of their time. The project director assures the NASA researcher that any commitment would be voluntary, collaborative, low pressure, and flexible.

In the four classrooms that were part of our study, the communication concerning the Zero-g Project during the Refinement stage was conducted through e-mail, as well as over the phone and in person. Two considerations were crucial for the teachers in our study: 1) would they be able to carry out technical tasks using their available computer equipment, and 2) how would they integrate the Zero-g project into their curriculum.

Of these concerns, the curriculum integration issue was resolved with each teacher incorporating the Zero-g Project into their existing curriculum and teaching practices. As stated by the project director, the project could be applied to language arts, science, social studies, or computer literacy. Although this was true, what emerged fairly early into the project was that the participation in the project would spread into knowledge areas outside of their curriculum. For example, in the middle school computer literacy class, students were required to address the issue of the nature of gravity, what caused it, and could it be produced artificially.

The problems of technical expertise and equipment were addressed throughout the project as difficulties arose. Prior to the project, the teachers were reassured that they would have full support of our research team, including trouble shooting, training, software, and equipment.

In the Refinement stage of the Zero-g project, the project director's role was to clarify the demands of the participants, emphasize their possible contributions, and offer what support if any is available. The potential participants decided whether the project benefits their students and is appropriate in their respective situations. The commitments that were crucial to the success and completion of the projects were declared in the Refinement stage.

The Refinement, Organization, and Pursuit stages overlapped in time during the lifespan of the Zero-g Project. During the course of the project, participants joined and left, new challenges and activities were proposed, and participants sent messages containing their designs, questions, and solutions to problems. Although this overlap was evident, there was an intensified effort to plan procedures, establish a timeline, and distribute resources in August and early September, prior to the beginning of the school year. A message sent by the project director on August 30th detailed a timeline and procedures for the entire school year. In the initial plan stated in this message, in September, participants would view the Zero-g video tape and send their solutions to a problem of what two people in Zero-g who were moving toward each other could do to avoid colliding with each other. In October, participants would select a challenge from a list of design challenges. In November they would submit a progress report, and in December they would post a final design that would be evaluated by a group of experts. From January to April, this process would be repeated with the addition that participants would generate their own design challenges. In May participants would integrate their design challenges into a single report which would be submitted to NASA.

Throughout the school year, specific design challenges and activities were posted on the network. In October, participants received specific design challenges, and were asked to choose either a food problem or a recreation problem. On February 26, the project director posted another message, outlining procedures for another challenge in which students would describe a typical day in Zero-g. This message also asked participants to design a school, addressing the differences they noted in a Zero-g environment. Finally, on March 4, the project director posted another message, asking participants to list five differences and similarities between Zero-g and gravity environments.

Once the teachers had decided to commit to the Zero-g Project, they began making arrangements in their respective instructional situations to carry out technical requirements. Each teacher required at least one computer, a modem, and a phone line to carry out the project with their students. Combining resources from their classroom equipment and funds, their schools, district, and the research team, the equipment was obtained. Computer facilities and on-line access varied across classrooms, with the middle and high school having a computer for each student but only one phone line in their computer lab. The elementary school classes shared one or two computers among twenty to thirty students and performed on-line tasks outside their classroom.

Each site also received technical support including training, software, and in some cases hardware from our research team. This support was essential, both to the initial and continued commitment of the teachers in this study. The second grade teacher said about support from the research team:

During the Organization Stage, which for the teachers occurred immediately prior to the school year, instructional logistics were planned. At this time, teachers determined how often class time would be devoted to network activity, where the work would occur, what tasks the students would perform, and what social groupings would perform those tasks.

In our study, the Pursuit Stage was observed to be the most active both inside the classrooms and on the network. Both network and classroom activity was organized round the goals and tasks set forth in messages in the Proposal and Organization stages of the project. The focus of the network messages during the Pursuit stage served to carry out the project challenges. As mentioned in the description of the Pursuit stage above, other messages involved management, logistics, and relationships between participants during the course of the project.

Messages addressing the challenges in the project can be grouped into six categories: (1) designs and solutions, (2) feedback to designs and solutions, (3) questions, (4) replies to questions, (5) discussions and, (6) referring to resources.

In the classrooms we observed, students spent the greatest proportion of time engaged in tasks devoted to design challenges, set forth in the Zero-g Project. This message excerpt sent by the fifth grade students includes solutions addressing the Collision Course Challenge.

These students interpreted the problems as a school in zero-g, rather than a space station. This context shaped their first solution, in which students would grab on to school lockers. The second and third answers illustrated a common misconception found in participant's messages, that people could propel themselves in zero-g through swimming or flying motions.

In response to the Zero-g Design Challenge, one of the middle school students in our study created a game to be played in a zero-g setting. This game is typical of many of the student designs, in that human movement in zero-g needed to be incorporated in the game design. The students learned, over the course of the project, that moving from one point to another in zero-g, required the individual to push off a stationary object, aiming in the intended direction. This game design addresses this problem.

This student's design addresses two issues addressed in the recreation design challenge. First, participants were asked to consider how people would control their movement in zero-g. Also, game logistics, including the physical setting, equipment and rules also needed to be addressed. These elements of the game would later be addressed by experts and other participants when critiquing the design.

The third challenge in the Zero-g Project asked students to consider how everyday life would be different in a zero-g environment and write a short story based on the theme, "A Day in Zero-g." A middle school student wrote:

A fourth challenge sent to the Zero-g Project asked students to identify five similarities and differences between normal gravity and zero-g environments, particularly school and instructional environments. The participants who tackled this project sent responses similar to the following message excerpt sent by a group of second grade students:

The Five Same and Different Challenge was the last problem solving activity in the Pursuit Stage. The second grade class sent their solutions in April, and the middle school students sent their messages in mid May, near the end of the school term. The "Day in Zero-g" and the "Five Same and Different" challenges were particularly effective in allowing students to apply their knowledge of Zero-g gained in the previous challenges to situations familiar to them. These everyday situations, including going to school, getting dressed, showering, and attending classes, provided opportunities for students to test their personal theories of how Zero-g affects the motion of objects and people. In the second grade and middle school classes, which participated in these challenges, students completed their work with fewer obstacles and interruptions than with the previous activities.

Qualitative analysis of the pursuit of the four challenges in the Zero-g Project revealed a common set of steps. These Pursuit steps include orientation, problem solving, writing and graphic construction, sending messages, feedback from net participants, network discussions, class discussions of feedback and network discussions, and written responses to feedback and network discussions.

The orientation phase introduces the topic to the students, cultivates background knowledge, and asserts or interprets the problem or task.

Before beginning work on the Zero-g Project at each of the four sites studied, students and teachers viewed the Free Fall video prepared for the project, followed by a discussion or question and answer session. Teachers asked students to think about living in zero-g and the problems people would encounter. The video included no verbal commentary, and offered no explanations about living in zero-g. The video motivated questions from teachers and students on how people went to the toilet, slept, ate, and controlled their movement. These questions provided a basis for the challenges to follow, and reoccurred during the course of the project.

To introduce the challenges to the students, teachers either read or paraphrased the message sent by the project director that contained the challenge description. After the teacher described the challenge and clarified its goal, she led the class in a discussion of the task and the particular aspect or problems of a zero-g environment addressed by the task. When introducing the project to the students, teachers explained that the work they produced for the project would be considered by NASA scientists and possibly be used in the future. In addition, the teacher explained the network community, and how messages were sent and received.

If some messages had already been posted on the network which addressed issues relating to the challenge, as in the case of the middle school that joined the Zero-g Project at mid-year, the teacher stimulated discussion by reading these to the students or by paraphrasing their contents.

For each challenge, each site spent some amount of time considering the problem, generating solutions, and expressing that solution in verbal and graphic forms. Although there was variation across sites and within each site depending upon the challenge, there was an identifiable sequence in the way the task was engaged. After the orientation stage, students engaged in brainstorming and problem solving. They carried out their work in small groups for the most part, although they also worked in small groups, dyads, as whole class, and as individuals. The process often entailed a series of subtasks including discussion, writing, writing conferences, and revision. However, the particular subtasks varied across and within classes depending upon the skills of the students, the complexity and difficulty of the task, and the time required and available for the task. Some students brainstormed and composed simultaneously, in other instances students discussed for hours, wrote, discussed their written work, then composed a final draft. In some instances written work served as the basis for further discussion, brainstorming, and problem solving. A message was then typed on the computer. This was a separate stage in the elementary school classes where one computer served twenty students. In the middle school and high school, where there was a one-to-one student-to-computer ratio, this took place while students discussed the challenge.

It was during the problem solving and expression stage that the teachers faced most difficulties particularly with time pressures. One problem was that students, when tackling the design challenge, typically faced at least one impasse, which would stall their work for two or three class periods. This would often require the assistance of the teacher or one of the adults working with the class. Another obstacle was caused by the shortage of computers for typing messages. In the second grade classroom, several class periods were required for students to type their messages on a single computer. In the fifth grade class the teacher, frustrated by computer problems and delays, typed some of the designs on her home computer. The elementary school teachers expressed the most anxiety over time pressures, reporting that they were neglecting other curriculum areas, and felt they lagged behind other project participants.

One episode illustrates their concerns about time. The fifth grade teacher had decided to end her class's participation in the Zero-g project after they had completed the Design Challenge. In fact, she had decided not to send all of her students designs to the other Zero-g project participants. When asked about this, she commented, "I don't have the expertise to upload the files, and I don't see anymore things being sent to Zero-g." One of the authors then offered to send the text files for her and her students. The teacher declined the offer, saying that it was "late in the year, and I haven't finished other curriculum units."

During the Pursuit stage at each site, students and teachers read and discussed network messages. This occurred during the pursuit stage, often after the teacher had found time to download a batch of messages. The messages served a number of purposes: they were models for student designs, a basis for discussing the characteristics of a zero-g environment, and feedback for student work. In several cases, when time was available, students read the feedback from other network participants, and proceeded through the problem solving and expression stage reported above.

In the second grade and middle school classes there were a few students who wanted to continue working on the project after other students had stopped. One second grade student enjoyed writing stories in a zero-g setting. A middle school student continued a discussion with one of the professors over the nature of gravity and life in outer space. This was typical of the final phase of the pursuit stage. In the classrooms we observed, the final phase came to a close gradually, as students moved on to other interests and teachers decided to address other curriculum areas.

Two patterns of message exchange were observed during the course of the Zero-g Project. The more common pattern involved an initial message, typically a question or a student-generated design or solution (which itself was a response to the original proposal message), followed by one or two responses, giving feedback or offering information. In this type of exchange, there were no replies from participants, responding to the original exchange. The exchange did not lead to an extended network discussion or debate. The two message excerpts reported below exemplify this type of exchange in which a high school student asked for feedback for her idea of using a conveyer belt to move in zero-g. The project director replies to her suggestion, after which no other participants responded to the exchange.

The second type of exchange, much less typical, began with a student question or design, continued with responses offering information or feedback, but then receiving responses from other participants, who engaged each other in an extended discussion or debate. This occurred twice during the course of the Zero-g Project, once at the beginning of the project, and again near mid-year.

The following message excerpts exemplify this type of extended exchange. A message sent by a high school student which suggested using magnetic shoes initiated the exchange.

This message received several message responses which commented on the practicality and efficacy of using magnetic shoes on spacecraft as a means of compensating for lack of gravity. Excerpts from two of these responses are reported below.

Following this exchange, another high school student responded by inquiring about a means to block the force of magnets in zero-g, rendering them harmless to sensitive equipment.

This message received a few responses commenting on the ability of lead to block magnetic force and described experiments that the student could conduct to arrive at a solution to his question. One of these responses is reported below:

The discussion concludes when one of the on-line expert participants, a university professor, questions whether magnets would be an optimal solution.

This set of messages illustrates flexibility of identifying problems. The discussion branches from whether magnetic shoes would work, to the feasibility of using magnets on spacecraft, and finally to finding alternatives to magnetic shoes, identifying their function and providing an alternative.

Also the discussion and points of view are distributed among several participants, including students. Participants whose background knowledge differs will find some aspects of problems more salient and critical than others. This gives the student more flexibility to choose the problem he or she wishes to address. In the case reported above, the student decided to experiment for himself whether lead would block magnetic force. This set of messages illustrates how the network community provided rich possibilities for problem solving and learning.

What can we say about the relationship between work in the classrooms and activity on the network during the Pursuit stage? The overall relationship may be best described as a network of loosely coupled communities, each committed to similar goals, but ultimately achieving those goals largely independently. Each class worked toward reaching their own designs and solutions to Zero-g challenges. However, during the problem solving process, participants both contributed and added to the network resources. Participants posted questions and designs on the network, then, sometimes during the same on-line session, downloaded questions, designs, and feedback from other participants.

The pace and intensity of activity over the network and inside the classrooms was dissimilar. When engaged in problem solving discussions or composing their designs, classroom work was at its highest intensity. Network activity only intensified when a message stimulated discussion or debate as in the case of the discussion about wearing magnetic shoes. These peaks in network activity seemed asynchronous with classroom activity.

Among the Zero-g project challenges, the Design challenge proved the most difficult, required the most time, and resulted in more complex social activity than the other challenges. At the four sites we studied, particularly where small groups collaborated on the designs, a distinctive pattern of activity was observed. This pattern consisted of five stages: (1) group and task assignment, (2) initiating problem solving, (3) group discord and work obstacles, (4) reorganization, and (5) revival.

Once groups were formed and the teacher informed them of their goal, to design either recreation or food facilities for a Zero-g environment, the students tackled the problem. Activity in the initial stages varied among groups, with some students brainstorming ideas while others pondered the problem in silence.

After a period of time, ranging from minutes to days, the group entered a group discord and obstacle stage, facing both social and problem solving obstacles. Group members disagreed on a game design or the details of a design. In the fifth grade class, the students initially decided to design a tennis game in which participants would roller skate on tracks. As students began to disagree on the design of the game, they started to divide, in different locations. At one point, we suggested that they could consider using the properties of zero-g, rather than trying to overcome them. Immediately the group generated ideas for a new game. One of the students was offended that his roller tennis game was discarded, and quit the group. He then worked individually designing a new game. The remaining students divided into two groups, one working on basketball, another on a different tennis game.

This pattern, where students encountered task obstacles which coincided with social turbulence, was observed at all sites in various forms. In some cases, students decided to leave their group to work individually. In the high school computer club, students entered into an argument which was never resolved. Eventually the students abandoned the project and gave their ideas to the teacher who completed the design. In the second grade class, it was common for one or two students of a larger group to take over the design task, while the remaining students ceased active participation. This is the reorganization stage.

Once new social groupings were reformed, the students entered the revival stage, resuming work on the design. However, after reorganization, there were fewer problems, and the design was completed. In the fifth grade class, each of the groups designing games completed their designs over a period of several classes. Although problems were encountered, they did not lead to group discord, and were worked out in a short period of time.

Student learning was affected by the different challenges over the lifecycle of the Zero-g project. Each challenge confronted students with similar problems and the application of conceptual knowledge in various contexts. The different challenges allowed students to discuss and write through a variety of genres including lists, narratives, and expository discourse. Over time, the solving of similar problems in different contexts allowed students to apply knowledge gained in earlier challenges to problems posed in later challenges. The most common example of this was the tendency for students to equate Zero-g conditions with that of water. In earlier challenges students often wrote that people could swim or float in Zero-g and that objects would "float up" if released. Through discussions and feedback from other network participants, students began to understand that Zero-g conditions were quite different than that of water. They wrote less about swimming and floatation, and devised other means of control of movement in Zero-g.

Following the Pursuit Stage, there comes a time when the participants end their project work. This is often indicated through the exchange of messages containing expressions of gratitude or a semi-formal announcement that their work is finished.

All of the wrap-up messages in the Zero-g Project expressed gratitude for being included in the project. One message expressed an intention to be more involved in the future. The second grade class included a summary report, which although was an attachment to the wrap-up message, is more applicable to the publication stage. The project director's message offered thanks to the participants, and also suggested that students view the free-fall video once again to realize their change in understanding the concepts discussed during the course of the project.

Wrap-up activities took place inside the classrooms. These were sometimes reported on the network and sometimes they were not. Among the four sites studied, there were two patterns of wrap-up activities. One type involved reflective activities in which students returned to ideas previously encountered during the project, discussed concepts, and presented their work. The other type was a teacher wrap-up.

The second grade and fifth grade classes spent the greatest amount of time engaged in wrap-up activities. As the second grade students were finishing their final Zero-g Challenge in mid May, the class composed a good-bye letter to the project participants. In late May, after the students’ project work was completed, the second grade class viewed the free-fall video again, and attempted to explain motion in zero-g using the knowledge they had gained.

The fifth grade class participated in wrap-up activities, but unlike the second grade class, their work was confined to their school setting, and did not end up being communicated on the network. Instead, the students presented their zero-g designs to other fifth graders in the school. As reported above for the second grade class, this activity also involved reflective thinking and discussion of the students' work and the concepts they encountered during the course of the project.

Unlike the elementary school classes, the middle school and high school students did not participate in reflective wrap-up activities. Wrap-up activity occurred primarily among the teachers. For the middle school teacher, this involved making sure all of the students had completed their work and had sent it over the network. The high school teacher collected her students' ideas and composed a recreation design herself, and sent it to the network.

For many project participants, the Wrap-up stage signals the conclusion to their work and obligations. However, network projects hold the potential for value to non-participating individuals and institutions. Distribution of knowledge gained through the project activity to a wider audience is the function of the Publication stage. The Publication stage of the Zero-g Project included electronic postings and conference presentation. The thesis of one of the authors of this paper was posted on the WWW soon after completion of the project. The thesis included a detailed history of the project, e-mail messages, and analysis of interviews, observations and completed work. In addition, accounts of the project have been presented to educators and scholars at research conferences and workshops.

As we can see from the analysis of four schools participating in the year-long Zero-g World Design Project, network-based educational activities undergo a lifecycle, starting out with some pre-activities (proposal, refinement, organization), continuing through the activity’s mature state (pursuit), and then through its post-activity (wrap-up, publication). Furthermore this lifecycle is recapitulated within the overall lifecycle of a project, at both micro and macro levels.

There are several reasons why it is useful to know about the lifecycle of network-based learning activities. First of all, an understanding of lifecycles helps the participants to understand their multiple roles in a network activity, which change as the activity proceeds through its lifecycle. The role of the project’s leader in the Zero-g Project, for example, was quite different in the preliminary stages than in the mature functioning stage and yet different in the closing stages.

The Zero-g World Design case study highlights the importance of active, effective moderators to initiate and sustain the interaction in an networked learning community. The construct of mediation in learning has been a central construct in socio-historical theories of learning (Vygotsky, 1978). This construct takes on a new appearance in network-based learning environment, and thus helps us better understand its importance even in more familiar face-to-face learning environments. Interaction in networks tends to stretch out over time, which also makes the importance of mediation and mediators easier to see. Most failures of attempts to build successful network learning environments are due to the lack of appropriate mediation at the appropriate times in the unfolding process of a network learning interaction.

From this analysis of network activity lifecycles, it is possible to identify essential elements of mediation. Levin (1999) has embedded these into a web-based interactive guide for people interested in creating and implementing network-based educational activities.

It is helpful to know about the lifecycle of network-based learning communities so that systemic support for the projects can be embedded in the institutional structures within which these activities occur. For example, Riel (1993) used this kind of lifecycle knowledge to build "learning circles," systemic organizational frameworks that proved sustainable and scaleable across many years.

Knowledge about the lifecycle of network-based learning communities can help teachers, administrators, and learners to integrate their involvement in these communities into their involvement in other concurrent educational activities, leading to a more powerful overall environment for learning.

This material is based upon work supported by the National Science Foundation under Grant No. RED-9253423. The Government has certain rights in this material. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. We would like to thank all the teachers, students, and others who participated in the Zero-g World Design Project.