C&C Systems Market Development Division, NEC Corporation

We have developed a synchronous editing system for creating a wall newspaper collaboratively by using local area networks. The educational system is intended to foster writing skills through a collaborative writing process that stimulates intensive interaction among learners who edit the same document.

Learning to write is important for the following two reasons: (1) writing is an essential skill required in daily life; and (2) learning to write can be a method that helps learners develop skills required in our information-oriented society, such as independently searching for information, evaluating it, creating new ideas, and presenting them to someone.

The following sections describe the reasons for focusing on the collaborative creation of a wall newspaper as an activity for learning to write and the design of the editing system we developed. It is also shown how learners interact in experiments using this system in an elementary school.

Taking these characteristics into consideration, activities for learning to write have the following requirements. A learnersí goal of activities should not be to be judged by teachers - but to communicate with readers about what they want to say or accomplish - since one of writingís main features is that it is a medium of communication. Moreover, activities should lead learners to revise their ideas and texts repeatedly since writing is a recursive process.

As the previous section highlighted, the important factors in learning to write through creating a wall newspaper collaboratively are awareness of the reader and intensive interaction among learners. The editing system to support this activity should be designed according to these factors as well as provide various learners with a place to edit the same document simultaneously. Thus, there are two main design requirements:

Requirement 1: Offering learners opportunities to be aware of readers, especially those involved in the writing process.

Requirement 2: Opening each learnerís writing and revising process to all learners who are collaborators in order to increase awareness and foster interaction.

Function 1: Inform learners that someone else is following their editing process on his/her (networked) computer screen. This helps them to be aware of readers involved in the writing process and thus satisfies Requirements 1. Moreover this causes them to interact with one another.

Function 2: Show learners instantly what others are writing and revising. This meets Requirement 2. This helps them share the editing process of every article and thus causes them interaction.

Function 3: Display the whole image of the newspaper always. This satisfies Requirement 2 and helps learners understand the editing process of the whole newspaper as well as individual articles. This fosters learnersí interaction.

Function 4: Show learners which part of the newspaper each learner is editing or reading and allow them to refer and follow his/her editing. This satisfies Requirement 2. This makes learners aware of other learners as collaborators and attracts their interests in other learners editing process and thus causes them to interact with one another.

Our critical concern in the design of the system is to provide awareness to other learners. There have been studies that also pointed out the importance of awareness. Fisheye views (Greenberg, et al., 1996; Gutwin, et al., 1997) are well-known as awareness tools in collaborative editing. They provide both global overview and local detail within a single window by distorting a two-dimensional space. However, they are not appropriate for editing printed matter such as a wall newspaper, because spatial layouts indicating the relationship among objects are difficult to maintain in fisheye space. Therefore we propose an another solution in our editing system.

The editing system is based on the distributed multimedia framework (Tabuchi, et al., 1998) so that it allows learners to edit the same document simultaneously.

Terminals open a shared document and temporarily store it on their own memory. When terminal A changes the document, this information (in Fig. 1, "Move a circle right horizontally.") is sent to every terminal and the document manager via the transmitting manager. According to this information, each terminal changes the document in its memory and the document manager changes the shared document on the server.

Fig. 1. System configuration.

The process of creating a wall newspaper using this system is divided into four phases: Creating a shared document, which is an electronic sheet; laying out articles on it; editing articles; and printing out the document as a sheet and completing the wall newspaper. Each phase is described as follows:

Teachers prepare a shared sheet (with optional sheet size) and register the group members who will edit the sheet.

Fig. 2. A window for laying out articles on a shared sheet.

Each member of the group (editor) edits articles to which he/she is assigned by himself/herself or collaboratively, as shown in the window in Fig. 3. What he/she writes and revises is instantly displayed on both the editing area and the whole newspaper view, which is always displayed on that window [Functions 2 and 3]. And the particular part of the newspaper that each member is editing, called SCOPE, is always displayed on the whole newspaper view [Function 4].

Fig. 3. A window for editing articles.

Group members cannot edit articles of other members but they can read them. They can distinguish such articles from their own articles by their color on the whole newspaper view. If they want to edit other membersí articles, they send a message to its editor asking for permission for collaborative editing. They do this by pushing the button for request. These functions prevent articles from being rewritten without the permission of their editors. Members also have permission to make their own texts and drawings invisible. Therefore they can decide whether they open their editing process to other members depending on their editing stage.

In the editing process, it is important to refer to articles edited by others and to see how others edit them. A group member can read any part of the newspaper by moving their SCOPE. They can also follow someoneís editing by clicking on that memberís SCOPE [Function 4]. When they do this, they see a memberís name on their screen (Fig. 4). Simultaneously a member who is being followed and observed sees the followerís name on their screen [Function 1].

Fig. 4. Observing how another group member edits an article.

The whole electric sheet is divided into A4-size sheets and each sheet is printed out. Group members then put the printed papers together by hand and complete the wall newspaper.

To see if our system supports a collaborative writing process, we observed a class of students using our system. And we focused on: (1) if the system fosters learnersí awareness of readers; (2) if the system fosters interaction among learners; and (3) if this intensive interaction fosters learners to revise their writings.

One group, which consisted of six students: NS (6^th grade), IN, ST, KN, MI (5^th grade), and AK (4^th grade), was chosen for observation. In this group, two male students, IN and ST, were chosen as the subjects for videotaping; their conversations and actions as well as their PC screen images, were videotaped throughout the class. IN was the group leader. AK, who moved to the school in September, did not join the summer camp and was a complete beginner at computers. The wall newspaper edited by this group had five articles, a newspaper title, and an editorsí list (Fig. 5). Every student of the group, except AK, is assigned the chief editor of an article.

The following section, based on the video data, demonstrates how the group collaboratively edits two articles, A and B (Fig. 5), and how the system supports interaction among the group members during the editing process.

Fig. 5. A wall newspaper collaboratively edited by students.

The editors of article A were ST, IN, and AK. And ST was the chief editor, so he had leadership role in writing the article. After ST wrote one third of the article, IN gave his view on this writing to ST.

Fragment 1: in the 5^th class, Oct. 7

In this fragment, watching the screen of STís computer, IN communicated his opinion on the subject of STís writings and he then proposed how to revise it.

ST adopted INís proposal and informed IN during the 5^th class that he would edit article A accordingly. The following fragment is from the 6^th class.

Fragment 2: in the 6^th class, Oct. 14

In these fragments, IN took the leadership role and, with ST, edited article A by using INís computer. The leadership moved from ST to IN, and the computer being used changed as well. This computer change shows that IN and ST understand that their wall newspaper was shared by the group students and could be edited by using any of their five computers. This shows that the system satisfies Function 2 and this function supports interaction, collaborative editing, and revising.

After these fragments, ST became a leader again and edited article A by using his computer, and IN resumed editing the article of which he was chief editor. But IN followed STís editing of article A, showed in the next fragment.

Fragment 3: at around the end of the 6^th class, Oct. 14

This fragment shows that IN is supposed to increase his awareness of STís activity and the awareness leads another interaction between them. And it is supposed to be done through interaction and collaboration as shown in fragments 1 and 2. This shows that Function 2 is clearly taking place in these fragments and fosters awareness and interaction.

Editors of article B were KN and ST. And the chief editor was KN who had not earnestly engaged in the activity of a wall newspaper creation since the first class. The following fragment is from the 3^rd class which group students started to write articles.

Fragment 4: in the 3^rd class, Sep.16

In this fragment, IN advised KN to edit article B immediately after he saw KNís blank article on his screen. And after that, he frequently referred to KNís editing and advised him to write an article. The frequency became higher as the classes progressed and reached every ten minutes in the 5^th class.

This shows that referring KNís blank image on the screen led IN to try to engage KN in editing article B. Functions 2 and 4 are clearly taking place so we can say that the system supports an increase in awareness of other students and fosters interaction.

In spite of INís repeated advice, KN did not engage in writing and thus made little progress in article B. In the end, IN decided to leave it to ST in a short meeting held in the 5^th class, whose purpose is that each group modified the work plan following the progress in its newspaper. Additionally, he asked AK, who wrote article A with ST, to assist ST.

According to INís decision, ST, AK, and IN edited article B collaboratively in the next class. After they made some progress in it, KN began to engage in their editing willingly. The following fragments show the change of his behavior.

Fragment 5: in the 6^th class, Oct. 14

Fragment 6: at around the end of 6^th class, Oct. 14

In fragment 5, KN consulted IN about his own plan for article B. And in fragment 6, he told IN and ST that he wanted to participate in the collaborative writing between them.

These fragments show that KN is supposed to increase his willingness to write an article. And it is supposed to be done because he saw that his article had been written collaboratively by other students, IN, ST, and AK. This shows the system satisfies Function 2 and this function supports engaging a student in a collaborative writing.

A synchronous collaborative editing system for learning to write was designed and tested in an elementary school. And it was shown how the learners in the experiments interact when using this system. The experiments indicate that (1) students interacted and collaborated well in the writing process; (2) they revised articles repeatedly through interacting with one another; and (3) the system supports the collaborative writing process.

In the next step of our research we plan to analyze (1) the studentsí use of awareness functions in detail; (2) changes in the substances of individual articles and the entire wall newspaper which was edited by students. We also plan to compare the collaborative writing process using this system with those by hand and using other editing tools, such as word processor.

Nystrand, M. (1986): The structure of written communication. New York: Academic Press.

Flower, L. and Hayes, J. R. (1977): Problem-solving strategies and the writing process. College English, 1977, 39, pp. 449-461.

Scardamalia, M., Bereiter, C., McLean, R. S., Swallow J., and Woodruff, E. (1989): Computer-supported intentional learning environments, Journal of Educational Computing Research, 5, pp. 51-68.

Nystrand, M. and Brandt, D. (1989): Response to writing as a context for learning to write. In C. Anson (Ed.), Writing in response: Theory, practice and research (pp.209-230). Urbana, IL: National Council of Teachers of English.

Greenberg, S., Gutwin, C., and Cockburn, A. (1996): Awareness through fisheye views in relaxed-WYSIWIS groupware. Proceedings of Graphics Interfaceí96, pp. 28-38.

Gutwin, C., and Greenberg, S. (1997): Interactive Fisheye Views for Groupware. Short technical note, Department of Computer Science.

Tabuchi, Ayukawa, Ito, and Maeno (1998): A compound Document Middle-ware for group-ware and a Multimedia Group Learning System based on the Middle-ware. Transactions of Information Processing Society of Japan, Vol. 39, No. 10, (in Japanese).