Review of Literature

Collaborative writing itself may be broadly defined as the development of a single text by multiple writers. Computer supported collaborative writing, then, may be equally broadly defined as the use of a computer system in the collaborative process, be it as a composition tool, a tool for the communication of ideas, or both. As the term appears here, a text based virtual reality, then, is a system which extends the paradigms of computer communication and composition to include a textual environment which one can experience by issuing textual commands, interact in by using written communication, and effect, using a combination of both. MOO, MUSH, MUD, MUCK, MUSE, and MUQ are all part of a class of interactive text based realities generically referred to in this paper as MU*s. Each represents a different software development effort, with individual characteristics, strengths, and weaknesses. This review of literature provides a technological overview of computer support for collaborative creative writing, and details some of the experiments that have been conducted in it.

In 1989, Ken Kesey and 13 graduate students in a University of Oregon creative fiction class produced a novel called Caverns. In the forward, Kesey describes his motivations and methods.

"After years of refereeing... cruel and futile give and take, [of the traditional workshop method] I hit upon a plan. Have everybody work on the same project. Even better than walking a mile in the other writer's moccasins, mix them up 'till nobody can be sure whose are whose." (Kesey, 1990)

Kesey's first step was to establish ground rules. These included such ideas as: No writing or discussion of the story outside the class period. Kesey's vote equaled 50% of the class. Once this was done, the group began to develop characters and lay out the plot. A section of the plot was assigned to each writer, then written for half an hour. The completed drafts would then be read, in order, into a tape recorder. The tape would be transcribed and made available at the next session of the class for revision.

Kesey's method allowed him access to the actual process of composition - the process where the writer actually invents new text, rather than that of revision, which Kesey felt was the only part of the process traditional workshopping accessed, and it safely defused the sometimes bloodthirsty critical nature of the traditional workshop.

Computer support for such methods of writing instruction takes a variety of forms, from the most basic use as a word processor (in which case Kesey's group may be said to have been computer supported) to advanced tools to allow simultaneous access to the same text, writer to writer discussion, hyper textual commenting, and so on.

Perhaps the simplest method for computer supported collaborative writing discussed in the body of literature reviewed for this article involved the use of a single microcomputer in a first grade classroom, with most of the collaboration taking place outside the text, where it was observed by traditional ethnographic methods, in this case video tape and direct observation. This was collaboration at its most primitive, including such mechanical aspects as spelling, punctuation, and keyboard use. (Butler and Cox, 1990).

This method functioned in a surprisingly similar manner for college engineering students as they strove to create collaborative experimental reports. For them, collaboration became an exercise in revision, allowing easy access to the text already generated, as well as providing a means of creating a single, unified paper. (Arms, 1985)

Basic word processor support of collaborative writing forces most collaboration to take place outside the text. Human to human contact is still fundamental to the collaborative act, as is physical co-location of the collaborators. It also enforces a strictly linear creative process, in which one writer at a time works on the work, and the rest must simply wait. Metacognition, if any, is available for study only if the discussions are taped, which itself may further restrict discussion. Instructor feedback relies for the most part on traditional marking of the finished paper.

A more advanced approach, yet one which is still very common, is the use of electronic mail, for commentary and feedback on the text, and common system of text files for actual text access. (Hall and Hall, 1991)

This method shifts some of the communications burden to the computer, but can be awkward to use, requiring students and instructors to develop considerable expertise with the system being used. This method also, being somewhat more loosely controlled than the preceding, is occasionally subject to fascinating subversions of purpose and control.

Not all integrations of purposeful writing with MAILBAG into the classroom grew out of symbiosis between QUILL and a teacher's purposes. In one case, a teacher completely rejected MAILBAG because it conflicted with her views of the appropriate way to teach writing. This teacher started out using MAILBAG in the usual way, and students began sending messages according to their own purposes, such as love letters to one another. When the teacher discovered this, she immediately made MAILBAG unavailable[...] (Bruce and Rubin, 1990)

The shifting of part of the communications burden to the computer in the above method of computer support for collaborative writing has another interesting effect. Since all communication which takes place on-line must be in a text form, students find themselves writing, albeit informally to real audiences - their peers. This may explain the occasional subversions of purpose, and certainly offers a chance to practice the skills being taught in a fashion which is anything but drudgery. (Kinkead, 1988)

The e-mail/common file method of support suffers a lack of integration, essentially using standard large computer/network functions to accomplish a very specific set of tasks. This set of tasks may be integrated into a single, user friendly system, offering essentially the same services, but in a controlled, electronic bulletin board style of system. Such a system may be designed for the student and instructor of limited computer experience. If properly constructed, such a system allows the tasks required for the e-mail/common file system to be performed quickly and easily. The author of this study has previously been involved with such research, resulting in the DISCOURSE bulletin board system.

As support for collaborative writing grows more and more sophisticated, many intriguing new features are added. One such system is called the PREP editor. PREP borrows heavily from the hypertext model to facilitate the creation of the collaborative text itself. Text is segmented into "chunks" then assembled in vertical columns. A paper, the outline it springs from, and comments by other students or the instructor may all appear on screen at the same time, and are bound together by hypertext links. (Charney, 1991)

As research into text based virtual realities has expanded, it has taken a number of tacks, most often attempting to duplicate traditional classroom study. While this can be argued to be a useful place to start when bringing a student to text based virtual realities, MU*s offer possibilities for collaboration which simply cannot exist outside them.

All too frequently I log onto an educational MUD to find myself in a virtual representation of a university campus. [...] within these buildings are elaborately programmed classrooms. Teachers can lock students in and others out; they have tools for delivering lectures, for silencing one or all members of a class, and controlling who speaks when. (Fanderclai, 1995)

One example of a collaboration beyond the scope of the classroom, but well suited to MU*s is the use of the MOO environment to create enactments of a given piece of literature. This is done by first selecting a very "environment oriented" work, such as Dante's Inferno, and then creating the environment within the MOO, configuring "conversational robots" to read the lines - essentially producing a mini - play of the piece, complete with all the allegorical meanings visualized and expressed. (Harris, 1998)

Consideration has been given to the nature of collaborative learning in a text based virtual reality, both specifically to writing and to the very nature of discourse of any type within MU*s. Some argue that MU*s encourage a different type of discourse altogether, more similar to the oral discourse defended by Socrates in Plato's Phaedrus, due to the immediacy of feedback and the more conversational nature of text-based interaction possible in MU*s, as well as the sense of place and the control over it.

MOO provides a possible example of how [Computer Mediated Communication] can bring us closer to this ideal advanced in Phaedrus. [...] MOO characters not only interact with each other, but they can manipulate the objects of their environment as well. [...] just as the unmediated co-presence of orality and the social isolation of print are not inherently incompatible, neither are electronic media and the concept of cohesive polis. (Langham, 1994)

Consideration has also been given to the sociological perspective, where MU* presents some interesting new challenges for gathering and analyzing data from interactive electronic sources for purposes of ethnographic study, namely in the processing of logfiles generated during the normal operation of the environment.

Although doing a good job of systematic observation is without doubt much more difficult and problematic than simply walking around and describing what one sees, it is still true that in the study of virtual communities we can exploit some peculiar advantage. In many cases observation can be carried out even without informing the people being studied. (Paccagnella,1997)

Current research has also explored the cognitive and metacognitave environment of MU* based writing instruction, finding that the text based nature strongly encourages writing (and thinking) about writing during the writing process.

My students [...] online writing activities allow them to engage in group discussions about reading and writing, consult with writing tutors, and conference on their writing with me--cognitive activities that are the mainstays of my classroom. But further, the synchronous activities allow for online metacognition that face to face activities do not, and perhaps more importantly, the saved , printed logs of the activities allow for further reflection and metacognition. (English, 1998)

As the review of literature suggests, more current research has begun to diverge into the technical - how computer supported collaborative writing might be accomplished - and the pedagogical - how text based virtual realities might be used to teach writing. The study covered by this article was carried out to explore the gap between. Kesey suggested that collaborative writing gives the writing instructor access to the writing process. This project was an attempt to explore actual dynamic of collaborative writing during the writing process in a MU* environment.