The use of computer networks in education for non-real-time electronic mail messaging and computer conferencing is a relatively recent practice which began in the 1980s (see, for example, Black, Levin, Mehan, & Quinn, 1983; Levin, Riel, Rowe, & Boruta, 1984; Riel, 1983; Sayers, 1989; also, cf. Hiltz, 1988 for an extensive annotated bibliography). The use of computer networks for real-time written interaction, the focus of the papers in this issue, is even more recent, originating, as far as I know, in 1985 with the ENFI (Electronic Networks for Interaction) Project at Gallaudet University, a liberal arts university for the deaf. The goal of the ENFI Project at Gallaudet was to give deaf and hard-of-hearing students the opportunity to use written English in ways that hearing people routinely use spoken English--in authentic interaction and for a variety of purposes, including informal chit chat, more formal discussion of academic texts, generating ideas for compositions, or discussing compositions in progress. As various approaches have been implemented and studied at Gallaudet, we have found that network writing indeed provides a rich and promising new context for communicating in writing (cf. Peyton, 1989, for an annotated bibliography of this work). After hearing about Gallaudet's work, other teachers and institutions have seen advantages to giving hearing students opportunities to interact in writing as well, and are now using and studying various types of networking software and hardware and various approaches to real-time interactive writing on computer networks. Many (but not all) of these teachers and institutions call their work ENFI, as Gallaudet did, so I also use this term throughout the paper to refer to this particular use of computer networks. [1]

The setup of the network varies from school to school, but in most cases (and in all of the classes described in this issue) the teacher and students each sit at a computer terminal and type messages into a private, "composing" window at the bottom of the computer screen. When a key is hit, the message is sent to a public

window, tagged with the sender's name, and made visible to all members of the class. Messages can be sent whenever writers decide to send them. They appear in the public window in the order they are sent and scroll up the screen as new messages are typed and sent. Therefore, discussions can take place just as they would in regular classes, but they are written.

Teachers and researchers working with real-time network writing in classrooms have discovered several important features of the interaction. First, since the teacher sits at a computer terminal writing, it is very difficult and not even desirable to lecture. Second, it is rather difficult to use overhead transparencies and other visuals that many teachers are comfortable with, because using them involves a considerable amount of jumping up and down. Third, because students can type and send messages whenever they want, many teachers, at least initially, experience a sense that they have lost control of what happens in the class. Finally, many students experience a tremendous sense of linguistic freedom and energy when faced with a computer screen, similar to the "flaming" behaviors found in the electronic mail communication of professionals (Kiesler, Seigel, & McGuire, 1984; Rice & Love, 1987). In its lowest state, this behavior takes the form of vulgar or insulting language not seen before in regular classrooms (described in George, this issue; Kremers, 1988; Miller, 1989), but teachers have found that it can also be channelled into language use that is productive and creative.

These experiences with and observations of real-time network interaction in writing classrooms, along with more general conceptions about the potential of computers for transforming educational practice (cf. Bruce & Rubin in preparation, and Sirc, 1989, for overviews of current conceptions), have led to the belief that the use of computer networks in this way might promote "a shift away from the traditional writing classroom" and revolutionize the way writing is taught, by "creat[ing] entirely new pedagogical dynamics" (Batson, 1988b, p. 32). Specifically:

• Because teachers are not standing in front of the class but seated, like students, and their contributions appear as simply another line on the screen, it has been hypothesized that the result might be a shift in teacher role from authority figure and judge to participant-leader and collaborator--"ENFI blurs social distinctions in the class, making the teacher less the center of most ideas and energy" (Batson, 1988b, p. 33). "The teacher is not so much the channel of communication as the moderator, the facilitator. Since students are not physically oriented toward the teacher, they are not as likely to think they have to direct all comments to I that person] " (Batson, 1988a, p. 7).
  
• Because all participants can send messages at will and do not have to compete for the floor, it has been suggested that there might be a shift in classroom interaction patterns--"Activity is spread more evenly . . . contributions are more evenly distributed" (Batson, 1988a, p. 7).
  
• Because participants in written discussions can essentially write together, sharing a screen space and reacting immediately to what others have written, there might be a shift in student writing activity from the solitary production of text to a community of writers who create text collaboratively and comment immediately on the text of others (Batson, 1987).
  
• Because writing on the network involves genuine communication and the expression of ideas for a variety of real purposes (not only for the purpose of being evaluated), there might be a shift in student attitudes about writing--"Students may feel their writing has a more authentic purpose [since] they aren't always writing TO the teacher or FOR the teacher" (Batson, 1988b, p. 33).
  

In short, ENFI seems like a great idea, an innovation that can put into practice some of the best aspects of current theories about reading and writing--an emphasis on reading and writing for authentic purposes, collaborative work, active participation by all students, and the teacher in a less visible, controlling position.

For the past several years I have worked with a colleague, Bertram Bruce, at Bolt Beranek and Newman, Inc., to study how ENFI actually looks in classrooms, to sort out the reality of ENFI from the rhetoric about what might be possible, and to determine what benefits ENFI might hold for students and teachers. We have observed and taken field notes in ENFI classes, interviewed ENFI teachers and students, and collected transcripts of network discussions (a verbatim transcript of everything written on the network is automatically produced at the end of a session) and other writing for the class. In a few cases, we have been able to work closely and intensively with teachers, observing and interviewing frequently and systematically over a period of time and carefully analyzing the transcripts produced. In this paper, l present case studies of two ENFI teachers at Gallaudet with whom I have worked in this way for several years. In these case studies I hope to 1) show how the original goals and visions for ENFI were modified, and sometimes changed considerably, when teachers and students attempted and sometimes struggled to make sense of this innovation in the context of their own institutions, goals, and practices; and 2) raise some issues that need to be considered in attempts to implement ENFI and study its success for promoting students' learning.

Doug
Doug Miller [2] was one of the first ENFI teachers at Gallaudet and also one of the first ENFI "dropouts." After using the network in his first-year composition and literature classes in the fall, 1985, and the spring, 1986, he experienced serious conflicts between the demands of network use and his own teaching goals and habits, and he decided to stop using the network entirely. First, he saw as his primary task in the classroom the dispensing of information to students, and he found the network a cumbersome, inefficient medium for doing that. He believed that he was able to give the same information to his non-ENFI classes much more quickly and effectively.

Second, in his years of teaching he had worked out routines and materials that he felt very comfortable with but found difficult to use in ENFI classes, where he was always seated at a computer:

You're talking with someone who's taught here for 11 years . . . it means a lot of adjusting of things that I've developed and thought about and worked out. I mean there are certain things that I like to do in the regular class. I work with transparencies as a kind of team, pointing to things, making arrows, going from one side to the other. I stand up and talk about the ways that paragraphs are supposed to be arranged and jump around and use my body. I've gotten used to using my body as a way to communicate.... What I've been doing is taking the materials for my regular 103 [freshman composition] and running to my ENFI class in the afternoon. I get them there and I think, "What am I going to do with these things?" I realize I can't even pass them out, because then the students will have to look at something else other than the screen.

Third, in the composition course, his primary goal was to help the students pass the first-year exit exam that consisted of very structured opinion and compare/contrast essays, and he could not see how to make the connection between network writing and the students' need to produce these essays in their prescribed forms:

In English 103 the kinds of writing we're doing and preparing them to do for the writing evaluation is very specialized. We're worried about the form of the essay, the skill of creating something that looks enough like the essay to pass the evaluation. This very direct, glaring goal is right there in front of us, and the students know it.

Finally, he found that the network interaction gave the students a sense of freedom that he could not control:

On the network they didn't want to play, "Let's talk about this book" or "Let's practice writing sentences to please the teacher." They took over. They tried to trick me and sabotage the activity I had planned by logging off and then on again as someone else. They used language they would never think of using in the other class. One day someone's mother was visiting, to see Gallaudet and our new, exciting program, and the screen was filled with this filthy language and there was no way I could stop it without turning the system off. After struggling with it, I finally decided it was too frustrating to try to bring to ENFI the kinds of controls and structures that I normally use in the conventional classroom.

In short, Doug felt that in his non-ENFI classes he could accomplish his goals much more easily and successfully, dispense the information he needed to, and prepare the students for the test. And he felt less pressure because he was using teaching approaches and materials he was more comfortable with.

At the same time, when he dropped ENFI he felt a sense of loss. For one thing, he had gotten the impression that his ENFI students were smarter, better students. He wondered whether they actually were better students or whether they had simply had the chance, in the discussions on the network, to display their intelligence, while his other students were always listening to him lecture. He also wondered if there wasn't a place for the wild abandon that the network seemed to provoke, if he could only find a way to channel it:

The advantage of ENFI is that it dislikes formulas. It breaks down the tendency to codify the whole process of writing. It makes communication in writing more organic, more natural, more spontaneous, more realistic, less artificial, less contrived. It seems there should be room for that someplace in the system; the question is where.

For the next two years Doug taught the same composition and literature classes, He couldn't figure out how to fit interaction on the network into the course goals or the time available, and so he taught without it. But, in the summer of 1988, he experienced an "ENFI revival," a discovery that he now calls "one of the high points of my teaching career." He was in the last week of a three-week, intensive creative writing course for bright, motivated, deaf high-school students in Gallaudet's "Young Scholars' Program." Because classes met for six hours a day, he had a lot of time and, in fact, needed some variety. So he tried what he calls "script writing"--giving students a script and a list of characters and having them create their own dramas on the network. His role was one of manager or director:

I stopped trying to maintain very tight control, and I gave them the control. That's when all the excitement happened. I could now assume a role that more normally lends itself to the system--not having all the answers; not trying to manage and give feedback to every conversation; not trying to make sure students were paying attention to every word I was saying. I would pop in and out and offer advice, and they would consider it and either accept or reject it.

In those few days of dramatic production on the network, Doug was amazed at the "incredibly rich ways in which language was used, the levels of meaning and subtleties in language that were taking place."

. . . It's not the kind of communication that typically happens in a classroom. Normal classroom communication is not real. There's a sense in the students that what we're doing is not important; it' s only a rehearsal; it's only preparing them to do something else that is important. There's the sense that "I'm writing these sentences or this practice essay so l will later be able to write the real essay." There's the sense that they're writing to the teacher. They don't know the teacher. They don't care about the teacher.

But writing on the network has a sense of absolute urgency. Their ego is at stake; the personality they are showing in their writing is part of them; they're defensive about that; they're very careful about that; they want to assert themselves and function within this social context; it's real.

Doug decided to take what he now knew about the network back into his first-year composition class, to use one of three class days on the network dramatizing scripts that he created. The course goals hadn't changed--the students still needed to pass the same exit test--and Doug's desire to help them pass the test hadn't diminished. He now believed, though, that he had discovered the key to success with the network in "script writing," because as students created scripts, they would be discovering and inventing ideas together. The transcript of these ideas could then be used to help them write the required compositions.

I observed the class many times that semester, during ENFI and non-ENFI sessions, and found that on the network the original hopes for ENFI, described at the beginning of this paper, were indeed being realized. In the non-ENFI class, there was an atmosphere of intense seriousness. Doug lectured (using lots of transparencies) about ways to write various types of essays, and the students took notes; or Doug led a review of writing exercises the students had completed outside of class, and the students watched . There was very little student talk, and that came almost entirely from one student who made occasional comments or asked questions. Some students never spoke at all. In the ENFI class there was a much stronger sense of student involvement. Students were divided into groups on separate channels, and everyone had a role in the script. Doug was simply another actor, and because participants were identified on the network only by their character in the script, students often didn't know whether or not Doug was involved at all, or which character he was. There was also a stronger sense of excitement and engagement, and a lot of laughing as students worked together to create their scripts, played and competed with language, and displayed their linguistic showmanship. They didn't realize that they were "writing," and if they did, they seemed to consider it their own and not simply a product for a grade. At first, there was the familiar use of insults and obscenities seen initially in many ENFI classes, but soon most students dropped that kind of language and entered more appropriately into the drama they were creating. In fact, language seemed to become a vehicle for competition, for display, for showing off one's intelligence. In questionnaires and interviews about the network, most students said they enjoyed the network-created scripts and thought they helped them to generate ideas, get insights into the topic, and learn from the ideas of other students. They liked the non-ENFI part of the class too and felt they had learned from the lectures, writing exercises, and practice essays, but they liked the balance and thought that all lecture and structured writing exercises would be boring. All but one student said they would choose to take a class like this again.

But more than a month before the class ended, many weeks before he had intended to, Doug stopped using the network in this class. He explained

I think what is happening on the network is valuable, but there's not enough time for us to pursue that. The problems that I anticipated with finding ways to incorporate the ENFI activity into the objectives of this class are becoming more apparent. Because I'm so conscientious about getting them ready for the writing evaluation, I feel guilty about wasting or using class time giving them a script and assigning them roles and turning them loose when that isn't immediately helping them to write essays that conform to the standards of the test. 103 is so tightly structured and the work is so directly related to the tests, it's hard to break out of that and create other kinds of assignments and ways of evaluating student performance. So I feel very limited. There's so little time left, and I feel I need to give them more coaching. The writing evaluation is the thing that's shaping us, the big monster that's lurking in the shadows, that terrifies everyone, including me. (Emphasis added.)

The following semester Doug didn't use the network at all with his English composition class, and it looked as if that was the end of ENFI for Doug.

Harry
Harry Markowicz was also one of the first ENFI teachers at Gallaudet, like Doug using the network for the first time in 1985. His classes are primarily for college preparatory students, who have not yet passed the required reading and writing tests to be accepted into first-year English. For the past few years Harry has been attempting to develop his students' reading comprehension, and their abilities to write focused, extended text about academic topics, by conducting discussions on the network about books and articles the students have read on the theme "Growing up deaf." The students then write compositions about the same theme.

This approach with the network has seemed promising, so during the fall 1988 semester I visited some of Harry's classes and collected the transcripts of all of the network discussions. To my surprise and dismay upon reading the transcripts, I found most of the interactions to be very different from the characterization of ENFI at the beginning of this paper and from Harry's hopes for them. Instead of equal student participation with Harry acting as coach and facilitator, the discussions looked more like traditional classroom discussions. They usually revolved around a sheet of questions that Harry had prepared about the reading. Harry was the manager and leader, the principal channel of communication. He had the majority of the turns (around 37%), initiated almost all of the topics, and the students addressed almost all of their comments to him (a pattern that Thompson, 1988, has also found in her networked classes). Although the students initiated topics not central to the discussion of the book, Harry initiated nearly all of the book-related topics (around 90°~0), often with a specific display (known-answer) question--"What about the tricycle event? Do you remember what happened?" or a more open-ended request for information--"I want you to tell me what are the main events which are described in this chapter."

In fact, these discussions looked similar to a pattern found in much of oral classroom discourse, with the teacher asking a fact question, the students answering, and the teacher evaluating the responses and asking another question to probe for elaboration or to introduce another topic (Cazden, 1988):

It is true that students had more freedom to participate--they could all respond to a question or statement at once with no obvious or implicit constraints on the timing or length of the response--but the distribution of participation was by no means equal. In fact, some students participated a great deal and some, very little, as shown in Table 1, which illustrates the number and percent of turns taken by each participant in two class sessions near the end of the semester. On November 7, all that one student said was, "Good morning!" and "No, I never heard anything about it." Another said only, "no way!" and "ufo."

The students may have felt that the writing on the network had a more authentic purpose than their other writing for the class, but their exit test, an opinion essay evaluated mostly for grammatical correctness that they had to pass before being accepted into for-credit English courses as freshmen, was a powerful reality to them as well. Their comments in network discussions indicated that many would have preferred to spend their time doing structured grammar and vocabulary exercises and writing practice essays, work they saw as more directly related to the skills they would be

Will we do something different beside using the computer all the time??? I mean I would like to praticing our writing and to improve our vocabulary like some other classes do in Eng 50.

We talk to each other through computer which doesn't have helped us alot. This class seemed like one of class being offered as Group discussion where we share our ideas not talking about our weakness in english grammar structure.

How can the computer helps me with use proper english which i want to pass writing test. I wanna to pass it so badly.

I want to write a paragraph often to improve my writing

could you give us to write more not in computer. i feel i learned nothing in this computer. if i write more i would learn more because it helps me to remember etc. in english.

Unlike Doug's case, ENFI was very much alive and in use in this class, but it looked to me, and apparently to the students, very different from the original vision.

During this period of my study of Doug's and Harry's implementations of ENFI, the short movie, "Bambi meets Godzilla," popular in the 60s, kept coming to mind, unbidden. In the movie, Bambi is peacefully eating grass and flowers in a meadow with sweet music playing in the background. Suddenly, the huge foot of a monster appears and flattens him, and all we see are his little legs sticking out from under the foot. That's the end of the movie except for a long list of credits. At Gallaudet, it seemed as if a huge, indomitable force, in the form of teacher goals and teaching styles, institutional gateposts, and student attitudes, was in conflict with ENFI as it was originally envisioned, and I found myself appreciating the title of one of Donald Graves' conference talks, "When bad things happen to good ideas." It seemed as if, at least with these two teachers, ENFI had come to a rather bad end.

Fortunately, the situation and my views of it did not remain in that state. First, in the fall, 1988, Doug decided to try ENFI again, this time for script writing with a group of very bright, literate students in his Introduction to Drama class, which doesn't have an exit test. The students recreated on the network plays they had read and discussed in class (Oedipus Rex, Hamlet, A Doll's House, The Cherry Orchard, Six Characters in Search Of an Author, The Good Woman of Setzuan, and Death Of a Salesman) as a way to get involved with the plays, understand the conflict, and think and react the way the characters would. As before, Doug played a role along with the students. There was still vulgarity and sexual innuendo in some of the ENFI scripts (Doug says they're always there), but at the same time the students had the chance to represent a wide range of characters in a wide range of cultures, contexts, and historical periods and experiment with linguistic ways to show action, portray characters, and express thoughts and emotion without having to dress up or be "good actors." They could invent worlds together, completely out of their own creative thoughts and language, and they generated some stunningly imaginative and inventive scripts, as shown in these excerpts from network recreations of A Doll's House and Oedipus Rex. [4]

After using network writing in this context and in this way, Doug now believes that this type of interaction "taps a latent urgency to communicate freely, openly, and without the inhibitions and controls found in natural, face-to-face communication" (Miller, 1989), and that when this energy is channeled, it creates a context for the creative and imaginative use of language that will not occur in conventional classes and traditional writing assignments. In fact, he believes that for him at least, ENFI is not appropriate for the highly structured setting of more traditional, conventional courses and that new courses need to be created in which the true potential of the network can be explored. He plans to design such a course, with a title something like, "Creative Interaction on an Electronic Network ." In order for ENFI to be successful from Doug's perspective, he had to change the context in which it was used. Instead of ENFI revolutionalizing his traditional classroom, he took ENFI out of that type of classroom and created a new context for it.

My rethinking about the characteristics and success of ENFI came about in a different way in Harry's case. Here it was my own perspective rather than his practice that changed. Upon closer examination of the transcripts from his classes, l realized that my initial dismay over the very traditional character of his ENFI classes was unfounded. Although neither the interaction patterns nor student attitudes toward network writing conformed to the original vision for ENFI described at the beginning of this paper, the discussions in Harry's class were far from a useless waste of time, to be discounted and ignored as not representative of the real "ENFI." In fact, in the context of a rather traditional teacher-led discussion, these students were engaged in extremely valuable uses of written English: thinking about and discussing at length a text that had personal significance for them; working through the meaning of that text together; learning to express their ideas in interaction with others who expressed different ideas, in different ways, and with a teacher who not only pushed for clarification and elaboration, but also made available his own ideas and ways of expressing himself. In short, English was being used in complex and genuine ways for thought and self-expression, occasionally in the form of brief extended texts. The following example shows one of many such opportunities.

Harry's goals for network interactions are being realized. His task now is to make connections for his students between their writing on the network and other types of writing, including the kinds of writing they believe they need to do in order to succeed in the university.

Although these case studies involve only two teachers, they teach us a great deal about ENFI in particular, and more generally about implementing innovations involving computers in writing classes. First, we find, as many others studying educational innovations have found (Spaulding, 1989; Swanson-Owens, 1986, for example), that when any new technology or approach is implemented, it enters a "meaning system" (Swanson-Owens, 1986) that is already in operation. This system includes an institution, with its well-established procedures for assessment; individual teachers, with their own goals, theories of learning, and preferred teaching styles; and students, with their own expectations about the nature of teaching and learning, most of which have been shaped by their past educational experiences. When ENFI entered this institution and the classrooms of these teachers and students, it changed. Our original conception of a revolutionary classroom dynamic was never realized in the way we had imagined. Social distinctions were not necessarily blurred, shifting the teacher out of the role of authority figure and manager of classroom activity and discussion. Students may have felt more freedom and incentive to send funny or even vulgar or insulting messages when they wanted to, but at least in Harry's class, when the real work started, the teacher called the shots. Shifting the teacher out of the role of authority figure may not even be realistic or desirable in some instances, where the goal is to focus students' attention and energies rather than to promote free creativity.

Neither did writing on the network necessarily result in a shift in interaction patterns. Some students who were otherwise silent may have opened up on the network, and others who dominated classroom talk may have suddenly become silent; but the overall pattern appears to be more or less the same. Some people participated a lot, others very little. It may be unrealistic to expect anything different in a discussion of more than a couple of people.

The network did provide the opportunity for students to write in ways other than formal, formula-driven, solitarily produced compositions for the teacher to read and evaluate. Writing did take on a more authentic, immediate, even self-creating quality. However, for many of the students in this study, the "real" writing they did, the writing that counted, that either allowed them to advance through the institution's gateposts or held them back, was solitarily-produced text. So instead of changing their views about writing, some students wondered why they were writing on the network, which was so different from the work for which they would be evaluated.

The second thing we can learn from these case studies is that even though ENFI may look different from what was initially envisioned, the ENFI picture is not at all grim. Real-time written interaction seems to offer some extremely rich and valuable writing experiences for students. In both Doug's and Harry's classes, students showed intelligent, creative thought and self-expression on the network that the constraints of more formal compositions, and especially prescribed exit tests, may not have encouraged them to do. It may be that our traditional curricula, views of how learning takes place, and ways of measuring learning need to be changed radically so that writing and thinking in interaction can hold a central place in our planning and assessment, and its full potential can be explored and developed.

Finally, this study has taught me something about program evaluation. In evaluating a new technology or approach, there is often the tendency to believe that once we understand what the innovation is, we then need only to know "whether it works." The desire for proofs of success through externally-imposed measures of outcomes often takes too much of our attention away from the classroom, where real teachers and students determine both what the innovation is and whether it is successful. The very different approaches and results that Doug and Harry have had with ENFI have convinced me that we need to understand clearly "what our innovations mean to the teachers [and students] who have to use them" (Swanson-Owens, 1986, p. 95) before we can begin to ask meaningfully whether or not the innovation has succeeded.

1. It is by no means necessary to label real-time written interaction on computer networks "ENFI," but I use that term here for two reasons: because it should be helpful to separate this particular approach from all of the other ways that networks can be used; and because the teachers I will be quoting use it.
   

2. Both teachers have consented to the use of their actual names in this paper.
   

3. The network writing is shown here without corrections to spelling, punctuation, or grammatical form. It has, however, been edited somewhat to provide a sense of continuity. For that reason, turns are numbers so it is clear how far into the discussion or play the excerpt was taken and which turns were left out. Because the students are deaf, their writing at times has features characteristic of non-native English speakers.
   

4. As before, these two play excerpts have not been corrected, but have been edited to provide a coherent thread to the discourse. For a complete version of these excerpts, cf. Peyton and Miller, 1989.
   
   

Batson, T. ( 1987, January). The ENFI Project: Computer networks and the writing classroom. Project proposal to Annenberg/Corporation for Public Broadcasting. Washington, DC: Gallaudet University.

Batson, T. (1988a, April). Ex cathedra: The "pulpit model" of teaching and computer networks. Paper presented at the University of Chicago's Twelfth National Institute on Issues in Teaching and Learning. Chicago, Illinois.

Batson, T. (1988b). The ENFI Project: A networked classroom approach to writing instruction. Academic Computing, 2(5), 32-33, 55-56.

Black, S. D., Levin, J. A., Mehan, H., & Quinn, C. N. (1983). Real and non-real time interaction: Unraveling multiple threads of discourse. Discourse Processes, 6, 59-75.

Bruce, B., & Rubin, A. (in preparation). Introduction. Electronic quills: A situated evaluation of writing with computers in classrooms. Hillsdale, NJ: Erlbaum.

Cazden, C. B. (1988) . Classroom discourse: The language of teaching and learning. Portsmouth, NJ: Heinemann.

Hilz, S. R. (1988). Annotated bibliography: Publications on computer-mediated communication and education. In Teaching in a Virtual Classroom: Vol. 2 of final evaluation report, A Virtual Classroom on EIES. New Jersey: New Jersey Institute of Technology.

Kiesler, S., Seigel, J., & McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. Amencan Psychologist, 39(10), 1123-1134.

Kremers, M. (1988). Adams Sherman Hill meets ENFI: An inquiry and a retrospective. Computers and Composition, 5(3), 69-77.

Levin, J. A., Riel, M. M., Rowe, R. D., & Boruta, M. J. (1984). Muktuk meets jacuzzi: Computer networks and elementary school writers. In S. W. Freedman (Ed .), The acquisition of written language: Revision and response (pp. 160-171). Norwood, NJ: Ablex.

Miller, J . D. (1989, March). Script writing on the ENFI computer system. Paper presented at the 23rd Annual TESOL Convention. San Antonio, Texas.

Peyton, J. K. (1989). Computer networks for real-time written interaction in the writing classroom: An annotated bibliography. Computers and Composition, 6(3), 105-123.

Peyton, J. K. & Miller, D. M. (1989). Dramatic interaction on a computer network: Creating worlds with words and ideas. In D. Beil (Ed.), Teacher's guide to using computer networks for written interaction. Washington, DC: Realtime Learning Systems.

Rice, R. E., & Love, G. (1987). Electronic emotion: Socio-emotional content in a computer-mediated communication network. Communication Research, 14(1), 85-105.

Riel, M. (1983). Education and ecstacy: Computer chronicles of students writing together. The Quarterly Newsletter of the Laboratory of Comparative Human Cognition, 5(3), 59b7.

Sayers, D. (1989). Bilingual sister classes in computer writing networks. In D. Roen & D. M. Johnson (Eds.), Richness in writing: Empowering minority students (pp. 120-133). New York: Longman.

Sirc, G. (1989). Response in the electronic medium. In C. Anson (Ed.), Writing and response. Urbana, IL: National Council of Teachers of English.

Spaulding, C. L. (1989). The effects of ownership opportunities and instructional support on high school students' writing task engagement. Research in the Teaching of English, 23(2), 139-162.

Swanson-Owens, D. (1986). Identifying natural sources of resistance: A case study of implementing writing across the curriculum. Research in the Teaching of English, 20(1), 69- 96.

Thompson, D. P. (1988). Conversational networking: Why the teacher gets most of the lines. Collegiate Microcomputers, 6(3), 193-201.