Connecting Formal and Informal Discourses to

0 downloads 0 Views 110KB Size Report
In this paper, we focus on the "Question-Answer" function of the SOUTO system; ... mean a typical type of teacher-learner relationship in school maintained through ... necessarily helps learners locate and see question cards attached to their ..... For example, when students are going to copy their answer cards and paste into ...
Connecting Formal and Informal Discourses to Create Yet Another Zone of Learning Hideyuki Suzuki Ibaraki University [email protected]

Hideo Funaoi Science University of Tokyo [email protected]

ABSTRACT In this paper, we focus on the "Question-Answer" function of the SOUTO system; the hypermedia authoring system, and classroom activities where learners make hypermedia compositions with SOUTO and discuss their compositions with the function mentioned above. Field tests of the system reveal that; 1. The Question-Answer function reconfigures social relationships in the classroom and thus creates a foundation for collaborative learning, 2. The Question-Answer function creates a field of informal talk, 3. The SOUTO system provides a foundation on which both formal and informal talks are constituted and woven together, and thus, transition between them is enabled. Based on these findings, we suggest that educational systems should be designed as mediators which hybridize school-like activities and non-school-like activities. Keywords supporting discussion, reconfiguring classroom activities, informal conversation

CONVENTIONAL CSCL VS. RADICAL CSCL CSCL researches can be roughly divided into two types. One is the conventional CSCL research that is based on existing concepts of what 'learning' and 'school' are, and contributes, as a result, to maintaining the present form of school and classroom activities. The other is the radical CSCL research that are endeavoring to break the cycle of using this traditional concept which is a corner stone in the foundation of such conventional systems (e.g., Scardamalia et al. 1993, 1996). In this vein, we have treated educational systems as tools for school reform. That is, we strive to create yet another type of learning in school through having educational systems take the roles of agents that reconfigure the existing relationship between learners and teachers, learners and learners as well as learners/teachers and tools. In general, so-called "school-like practice" is maintained in classrooms as Mehan (1979) and Edwards (1990) have pointed out. This practice is based on “school-like social relationship”. By “school-like social relationship” we mean a typical type of teacher-learner relationship in school maintained through teacher-learner interaction that continuously casts the teacher as a provider of information or a person who is engrossed in evaluating her/his students' performance with the learners perceived as receivers of the information or persons who are to be evaluated. Educational systems or materials embedded in this relationship appear as tools which enable on-going activity, i.e., school-like practice. In the light of activity theory (e.g., Engeström 1986, Bødker 1991) or actor network theory (e.g., Latour 1996, Law 1992), they are repositioned and given new significance in the relationship, while being exploited as a resource for forming the relationship recursively. School-like practice is constituted on/within the function system comprised of teacher, learners and tools including the educational system. The term “function system” suggests that the origin of school-like practice can not be attributed to particular agents (i.e., the teacher, learners, tools) and that reconfiguring the relationship (network) is the only way to reform the school-like practice. It also implies that introducing a new artifact into the classroom could cause a change in the relationship and, as a consequence, cause a change in the learning activities. This paper focuses on the "Question-Answer" function of SOUTO system (Funaoi 1999). It discusses how the function reconfigures conventional social relationships in classroom and provides learners with a field of collaborative learning.

SOUTO SYSTEM SOUTO (creator/thinker) is a hypermedia authoring system. The system boasts skeleton set of functions to treat multimedia materials and simple GUI that allows learners to access the functions without difficulties. The typical process for making a multimedia composition with the SOUTO system is as follows; (1) Opening Card Editor (fig. 1) to edit cards. The card can include text data, images (jpeg or bitmap), and buttons; (2) Opening Worksheet Editor (fig. 2) to organize cards into a composition. Users can establish linkage between cards; (3) Moving to view mode

to browse the composition; (4) Moving to (1) or (2) for revision. SOUTO is run on a LAN environment and all users’ compositions as well as cards and image data are saved on the server machine within the LAN. This enables every learner to see the other learners' compositions and to import images and cards made by the others. Fig.1 Card Editor of SOUTO system Fig.2 Worksheet Editor of SOUTO system

QUESTION-ANSWER FUNCTION

The Question-Answer function is designed to facilitate learners’ conversation on their compositions. This function is comprised of the following two sub-functions. Question/answer editor When a learner conceives questions or some comments while s/he is browsing her/his colleagues' compositions, s/he can make a question. Just pushing the "question" button located on the title bar of a card activates the Question Editor with which s/he can write what s/he want to ask without difficulty since the logic of its interface is the same as the Card Editor of the SOUTO system. Giving an answer to a question can be done with the Answer Editor which can be activated by pushing the “Answer” button on the question card. Answer cards are automatically linked to the respective question cards. Original cards (the questioned cards), Question cards and Answer cards can be observed panoramically on the Work Sheet Editor (fig. 3).

fig. 3 the Worksheet Editor shows question/answer cards Questioned/answered notification Users are sent a notification when one of their cards is questioned or answered. The notification provides (a) the name of the learner who made the question/answer, (b) the title and contents of the original card questioned/answered and (c) the contents of the question/answer card. The notification appears when the question/answer card is generated, i. e., either Question Editor or Answer Editor are closed. If the expected receiver is off-line, the notifications are saved on the server machine and delivered afterward when the receiver logs on.

The authors consider that the Question-Answer function contributes to reposition the interwoven agents that constitute the conventional classroom activities, and to create a new relationship on which learner-learner interaction drives their learning. In particular, the function brings the following changes to learners as well as to the SOUTO compositions. Peer learners as audience This function turns peer learners into audience, or significant others. With this function, editing work sheets necessarily helps learners locate and see question cards attached to their compositions because this module provides a juxtapositional, or panoramic, view of the compositions and question/answer cards. This view helps learners to be aware of the attention other learners are paying to their compositions. The questioned/answered notification also helps learners to discern the “eyes” of their colleagues. These functions jointly encourage learners to see the other learners as audience of their compositions. In other words, accountability of their composition toward their peer learners, not toward the teacher, is saliently recognized by the learners. In this situation, the conventional relationship in which learners and the teacher mutually orient toward is resolved and a renewed community of collaborative learners appears. Hypermedia compositions as provisional works Both displaying compositions and question/answer cards altogether and providing questioned/answered notifications ensure learners that the questioning of the compositions is a relevant action. That is, they see the compositions not as untouchable final products but as on-going provisional works, which are always open to discussion. Consequently, cues for revision and re-research are embedded in the process of editing work sheets. Hypermedia compositions as multivoiced products As mentioned above, the Question-Answer function makes revision of the compositions mandatory works in the SOUTO classroom. It is important that the revised compositions do not belong to an individual learner anymore, but that they assume multivoicedness. That is, revision of the composition can be seen as a response to the others’ voice, or questions. This means that revising the composition is a dialogic process and the outcome of the revision comprehends the others’ voice inside. The multivoicedness is an imperative foundation on which the collaborative nature of the learning environment is constituted.

FIELD TESTS We had two field tests which both had university students research computer-related technical/social problems of their choice, e.g., history of CPU development, the mechanism of the hard drive, SPAM mail, cracking and computer viruses, computer aided management, and so on. We required them to make hypermedia compositions based on their own research with the SOUTO system. Fourteen students participated in test 1; sixteen students participated in test 2. Test 1 was comprised of five 90 minutes sessions; test 2 of seven sessions. The first session of both tests included an introduction to SOUTO system. In both tests, every change in the cards and compositions was recorded in sequence in order to trace the process of their work. In test 1, we selected one student as a target for video taping and recorded his actions and utterances during the test. The target, hereafter known as “MK”, was a male student who chose “the mechanism of printers” as his research theme.

ANALYSIS OF CONVERSATION -MK’S COMMITMENT TO ZB In this section, we discuss how the Question-Answer function shift classroom activities based on video data from test 1. The following fragment shows how responding to a question was treated by MK; and what kind of relationship he participated in. On the fifth session of test 1. M was notified by the system that student ZB put a question card to his composition. The question was about difference between additive and subtractive process of color. On seeing the question card, MK started to browse web pages to search for information to answer the question. In the following fragment, ( ) refers to an unclear utterance; refers to a caption, IS refers to instructor. Fragment

to

I,

one

of

the

MK[01]: Mr. IS, this is about( ). Is it OK to answer this based on what I’ve found? I searched and searched. But I have not located information seems really good. IS[02]: Why don’t you answer: “This answer is based on information I’ve found so far”? MK[03]: Oh, yes. IS[04]: “This is what I’ve located”, sound OK? MK[05]: I can do that, right? IS[06]: Yes. IS[07]: This problem suddenly happened. MK[08]: Is this my fault? IS[09]: No, no, not exactly. MK[10]: I could answer the question by using information I have right now, but IS[11]: All right, why don’t you do that? MK[12]: Well, I am thinking about searching for a better one, you know. MK[13]: Hey, ZB, have you already got answers from anyone? ZB[14]: Yeah. MK[15]: Will you look and check my response? ZB[16]: Why? MK[17]: Well, I haven’t found a perfect answer, but I have information that might satisfy you. ZB[18]: Oh, yeah

At the beginning of this fragment, MK tries to sound out IS’s receptiveness to his compromise, i. e., answering ZB’s question based on information he has already had [01, 05]; and IS endorses MK’s compromise [02, 04, 06]. In the middle part of the fragment, MK expresses his intention that he would like to search more to find a better answer [12]. This statement [12] is inconsistent with his former statement [01, 05]. MK does not intend to flatter the instructor by showing his willingness to do more research. This is apparent if you see the successive exchange [13-18] where MK visits ZB, who sent the question, and explains his attempt to answer her question [17]. This statement suggests that, at least at this moment, MK orients toward ZB, the person who sent the question, not the instructor; and that MK recognizes that he may be responsible for her understanding of the matter in answering her question. The fact that MK did not stop searching in spite of the instructor’s endorsement [02, 04, 06] of MK’s first proposal to compromise [01] indicates MK’s commitment to find a good answer for ZB. The authors presume that the Question-Answer module contributes to the emergence of that kind of commitment in the SOUTO classroom,. This module the sends questions directly to the person who asked without going through the instructors. The person who is asked receives the question cards inscribed with the interpellants’ name. Answer cards are also sent to the interpellants directly. This information pathway formed by the Question-Answer module helps learners recognize that their compositions as well as question/answer cards are located in the field of learnerlearner interaction.

ANALYSIS OF CONTENTS COMPOSITIONS

OF

QUESTION-ANSWER

CARDS

AND

A detailed examination of the contents of the question/answer cards reveals that the Question-Answer function creates a field of small talk where learners engage in informal and person-to-person conversations. It is a world of vivid colloquialism filled with vulgar expressions, laughter and fury. It is an alternative to the so-called formal classroom conversation. (Strictly speaking, small talk and formal classroom conversation are reciprocally constituted as figure and background). In this section, we have discussed (1) the different features between discourse which appears in Question-Answer cards and compositions, (2) how the hypermedia compositions, products of the classroom activity, and small talk generated around the composition are related.

Compositions: Peremptory talk about “FACT” Compositions appear as a fixed “front”. In the process of making compositions, the authors of the compositions might sometimes be uncertain about their ideas or knowledge and suffer from a lack of information. However, this area of uncertainty disappears when the composition is completed. Thus, the compositions appear as assertive and coherent. The following cases clearly exemplify the nature of the compositions. [ST(ID for the author): “PC recycle” (Title of the composition)] #01 The bulk of PC users is constituted by cooperate users. Many PCs are purchased by a lease contract. Thus many PCs are discarded at the expiration of the contract. [MT: “Risk of Internet shopping”] #02 If the business is legal, the company is not liable to refund a customer (skip) It should be noted that things are treated differently in the cases of deception and fallacy.

As these examples clearly show, information in compositions is talked about/displayed as peremptory “FACT”. In other words, “completed information and talk” is required in the compositions. It should be noted that the subject of the talk does not appear in the surface, that is the talk within the compositions is de-personalized. Moreover, making compositions is the process of concealing deficit and vacillation and having the compositions express fixed and authorized information. Question-Answer: The world of person-to-person conversation Talking privately In the Question-Answer, learners talk privately. In the case followed, student ZB expresses her question through examining information in the composition in light of her own experiences. [ZB: Question on the “History of CPU” (by FY)] #03 I often notice Intel’s Pentium and Celeron (is this spelling OK?) in appliance shops, but I’ve never seen the other companies’ CPUs. They don’t seem so popular. Can you actually name personal computers that have non-intel CPUs inside?

In the following exchange, ST expresses her personal opinion when answering MT’s question. [MT: Question to “PC recycle” (by ST)] #04 This appears to be an unavoidable problem, so (skip) I really hope the recycling of PCs will work out well. [ST: Answer to MT’s question (#04)] #05 I am sure we are all seeing the same goal. Establishing a recycling system as a resource reservation is not only your hope, but mine as well. Thank you for your comment.

In the following exchange, FY expresses her visible anger in responding to UK’s question. [UK: Question to “Internet job hunting”] #06 How many students will attend the meeting (a job fair)? Who is going to organize it? Do I really have to attend this? It seems dull. [FY: Answer to UK’s question (#06)] #07 I think you can skip it if you are afraid it will be dull. No one is forcing you to be there.(skip) Anyhow, I recommend that you do your job hunt as you like. I couldn’t care less about your job hunting

It would be fair to say that questions like #06 can not appear in a public and formal classroom conversation; FY’s answer to the question is possible in a person to person dialogue. The examples above show that through the reciprocation of the Question-Answer, information in the compositions is re-recounted based on their personal opinion, experiences and emotion of the students. This is no less restitution of vivid self that was once hidden when the compositions were completed. One of important educational consequences of making learners participate in this kind of conversation is that learners are given the opportunity to understand things in light of their personal and embodied experience. Apology and irresponsibility The Question-Answer also creates a field in which the backdoors which were closed when the compositions were completed, are more easily left open. In the following answer, ST states that she is not sure about the meaning of the term “lease” used in her compositoin. [ST: Answer to the question on meaning of “lease”] #08 You are right. The expression “purchasing by lease” seems odd. “Leasing” is borrowing, not buying. In fact, this expression was copied from a book, so I am not sure.

Here, ST admits the expression “purchasing by lease contract” is not adequate. This expression is actually correct though the point is that the adequacy of the expression is denied through examining from her own sense of language. It is also interesting to note that ST revealed her uncertainty by stating “this expression was copied from a book, so I am not sure”. In the example below, MT confesses the limitation of her research. [ST: Question to “Risk of internet shopping] #09 Can I get a refund in case of deception and fallacy? [MT: Answer to ST’s question (#09)] #10 I am very sorry but I was not able to find out what happens in the case of deception and fallacy.

In the Question-Answer, talks which never show up in the compositions, or talks which are eliminated to make compositions “complete compositions” appears. This talks includes expressing one’s lack of knowledge, limitations, apologies and so on. It is, of course, possible to make this kind of talk in more formal conversation in classroom. However, the talks risk the compositions, that is, the talk may abort the compositions and mark failure of the task. In contrast, the effect of talk made within the Question-Answer function is basically enclosed within person-to-person relationship. Consequently, learners can accept that the weak points in their compositions be pointed out. Sometimes, they may take on a more defiant attitude. Such irresponsibility can be noted in the following answer. [OH: Answer to a question about the definition of Knowledge Management] #11 I guess it might be a project to reformulate information and knowledge by forming databases.

As previously mentioned, the compositions appear as fixed and as an authoritative text. However, the juxtaposition of personalized talks shown above with the fixed text relativize the fixedness since this juxtaposition reveals that the static and authoritative nature of the compositions is a product of a certain practice to render the composition as fixed text. Through this process of relativization, the compositions are reconstituted as artifacts for collaboration. Joking, capering, decrying The world of Question-Answer is filled with joking, capering and decrying. In the following example, MC is playing with words by contrasting two similar pronunciations.

[MC: Answer to the question about the origin of the abbreviation “An-gura”] #12 I do not know right now when the abbreviation “An-gura”(a Japanese word refers to underground) emerged. I am sorry, but I am sure “An-gura” has no relation with “angler”, although both give us an image of “dark place”. (note: In general, Japanese does not distinguish sound “R” and “L”, so “An-gura” sounds like “angler” for Japanese people).

The following exchange is a good example of mutual joking. [OH: Question to “Knowledge Management” (by SK)] #13 I understand that Ikujiro Nonaka is one of the Knowledge management researchers, but, I would like to know what his parents were thinking when they named their new born son Ikujiro. [SK: Answer to the question of OH(#13)] #14 This is a very difficult question. Do you realize that? I would like to date you to discuss this problem.

The following question is about the definition of knowledge management. The point here is that IZ factitiously misreads the composition and proposes a vindictive interpretation of knowledge management. [IZ: Question to “Knowledge Management” (by SK)] #15 Knowledge Management is a system which enables employees to cavil one another. Right?

Both question [13] and question [15] are based on an unexpected reading of the compositions. Nonaka’s given name is a trivial matter when we read about knowledge management. The author of the composition did not expect the reader to conceive of that question. Likewise, the author does not expect the interpretation given in question [15] when she explained the definition of knowledge management. These questions tease no-nonsense reading of the composition and question authority of the concept of knowledge management and its founder, Ikujiro Nonaka. Joking, capering and decrying are dangerous noises that spoil normative reading of compositions. However, they also form a productive syncrisis and anacrisis that drive creative learning through dialogue (see Bakhtin 1984). Informal talk brought into compositions In the previous sections, we discussed the consequences of the apposition of informal dialogue supported by the Question-Answer function with the compositions. In short, the compositions were embedded in the informal conversations and personalized, de-fixed, de-authorized and de-crowned. In this section, we focus on how exchanges in the Question-Answer are penetrated and influence the compositions. Basically, the process follows the reverse course. For example, when students are going to copy their answer cards and paste into their compositions, statements like “thank you for your comments”, “well, this is what I can say to you”, “I guess” and so on are usually excerpted. It is fair to say that this is a way to give a fixed and authoritative nature to the compositions through a de-personalization of the person-to-person conversation and being a concealment of the speaking subject. This does not necessarily mean a complete detachment of the compositions from the voice of individuals. For example, when they edit cards in the compositions as a response to a question, they often add redirect statements like “I edited the composition. Please see that” or “I added some examples to the composition” to the answer cards. When accompanied by such statements, the compositions appear as a hybrid of public and personal nature. Here, we show how compositions and exchanges on the Question-Answer function mutually penetrate. For that purpose, we will focus on FY who is working on “Internet job hunting”. When she made a card (#16) titled “Job fair”, she got a question card (#17) from UK. [FY: a card “Job fair” in “Internet job hunting”] 22nd Jan. #16 The date and frequency strongly recommended.

of

the

job

fair

depends

on

each

school.

(skip)

Participation

is

This card is followed by UK’s question (#17), which can be divided into three parts. [UK: question card] 22nd Jan. #17 {q-a: How many students attend the meeting (a job fair)?}{q-b: Who is going to organize it?} {q-c: Do I really have to attend this? It seems dull.}

FY’s answer to the question is as follows [FY: answer card]

22nd Jan. #18

{a-a, a-b: I will inquire into them}{a-c: I think you can skip it if you are afraid it will be dull. You can decide what you want to do by yourself.}

In this answer, FY replies to q-a and q-b by showing her will to inquire, and she responses anger at q-c. A week later, FY edits the card “Job fair” as follows. [FY: a card “Job fair” in “Internet job hunting”] 8th Feb. #19 {A-b: The job fair is organized by colleges or universities}(skip) {A-c: You are not required to attend}but, it may help you to get a good start (skip)

In this card, q-b: ”Who will organize it?” is answered by A-b. q-c: ”Do I have to attend?” is also answered here by A-c. It is fair to consider A-c as variant of a-c in #18. In A-c, the aggressive nature of a-c is resolved and the information is expressed more coolly as a general fact. In this card, previous exchange [#17-#18] appears as transfigured version. This means that the voice of UK as well as FY penetrates into the composition. As a consequence, the composition is anchored in the private dialogue while maintaining its public and formal nature. This is a significant phenomenon in terms of collaborative learning. Imbuing compositions with the participants’ vices modifies the compositions into shared facilities for collaborative learning. The multivoiced compositions then assume incommensurable features inside them, i. e., they are products of formal classroom activity that express individual ability; they are also shared products of collaboration. This hybridism can appear on the premise that their multivoicedness is visualized for learners. The SOUTO’s Question-Answer function supports visualization of multivoicedness by providing a juxtapositional view of compositions and question/answer cards. Interestingly, FY edits the answer card (#18) after she finished editing the card in the composition (#19). [FY: answer card] 8th Feb. #20 {a-a: I think half of the students skip the job fair} {a-c1: I think you can skip it if you are afraid it will be dull. No one is forcing you to be there }(skip){a-c2: Anyhow, I recommend that you do your job hunt as you like. I couldn’t care less about your job hunting}

In this card, q-a:” How many students attend the meeting” is answered by a-a. a-c2 is added as an answer to q-c in #17. It is notable that a-c2 is a stronger version of a-c1: that is, it is more offensive than a-c1. What is the significance of this revision at this point of time? The answer to q-a appears only in this card. This implies that FY intends to keep the question-answer sequence “q-a=>a-a” within the field of informal and private conversation. Furthermore, while FY answers to UK’s question by showing the organizer and merit of the job fair coolly in the composition, she emotionally and obstinately attacks UK in the answer card. It is note worthy that the talk of FY extends into two regions, that is, the region of compositions (public talk) and that of Question-Answer (private talk), and assumes a multi-layered nature. In other words, FY displays two different identities simultaneously: One is an identity of learners making formal talk about the compositions. The other is that of students engaged in emotional and private reciprocation of words with their peers. The Question-Answer function of SOUTO mediates between two different worlds and sets learners at the intersection of the two worlds. This position would be the zone of possibility where a new region of discourse anchored to both formal and informal worlds and at the same time separated from both.

DISCUSSION Observation in the SOUTO classroom clearly showed the Question-Answer function contributed to change in the existing social relation in the classroom. However, what we saw in the classroom was not a monolithic “alternative activity” brought about by the Question-Answer function, but rather a continuous constitution of different regions (i. e., the region of formal classroom discourse and that of informal small talk) where students stand on the intersection of these regions. The region of small talk closely resembles “a carnivalized world” (Bakhtin 1968). According to Bakhtin, a carnival is “a special type of communication impossible in everyday life," with "special forms of marketplace speech and gesture, frank and free, permitting no distance between those who came in contact with each other and liberating from norms of etiquette and decency imposed at other times". In this statement, “everyday life” refers to medieval life ruled by the authority of the church and official norms. Bakhtin also discusses the incommensurableness of a carnival with everyday life. That is, “during carnival time life is subject only to its laws, that is, the laws of its own freedom" and “while carnival lasts, there is no other life outside it”. The region of small talk we observed was full of the carnivalistic characteristics such as abusive language, anger, parody, mutual mockery, and degradation of authorities. In addition, it was also observed that the region was ruled by laws incommensurable with that of school-like activities. It is fair to consider that the region of small talk contains elements of carnivalistic nature in relation to the school-like activities. It is notable that SOUTO links the two regions. In other words, SOUTO appears as a boundary object (Star 1987) upon which different activities are constituted and woven together, and thus, transitions among them are enabled. SOUTO can be a tool for making hypermedia compositions into a product for classroom activity, it is also a tool for person-to-person small talk about the composition, that is, carnival. This hybridism allows the students to construct their learning based on both of the regions, i. e., that of school-like activities and that of carnival. Hybrid tool ensures hybrid learning. What is the significance of this linkage? From the viewpoint of designing of collaborative learning environments, the linkage supports learning through discussion. We believe that learning through discussion occurs only when the discussion assumes an emergent nature. That is, in the discussion, knowledge is not taught from one to another but comes about through the interaction/negotiation among interlocutors. Bakhtin (1984) finds this emergent nature in Platonic dialogues. Bakhtin claims that this emergent nature of Platonic dialogues appears only when the dialogues keep the carnivalistic nature within them, and that Platonic dialogues lose creativity, that is, turn into catechism, once the carnival is removed. The SOUTO system linking the region of school-like activity with carnival (region of small talk) can be an adequate tool for collaborative learning through discussion. From the viewpoint of school reform, this linkage is expected to constitute yet another zone of learning. The region of carnival is an integral part of students’ lives and their reality as students. However, it has been excluded when student activities are organized as formal learning activities. Making learners stand between a school-like world and the carnival world provides the possibility to realize an unprecedented learning activity with the missing part, or the region of carnival, as one of the constituents. The newly formed zone of learning is a field where school-like activities are questioned and relativized. It is also a field where learning is anchored to both sides of school life, i. e., classroom activities and small talk, and where the students appear as hybrid agents. This study suggests that educational systems should be designed as boundary objects that constructively produce conflict between existing school activities and carnivalistic activities, and establish traffic between them. The point of this design strategy is not to replace school-like activities with “non-school-like activities”, but to make these activities encounter and hybridize into yet another zone of learning. It is important that, in this work, the schoollike activities are not concealed from the students, but are visualized for them and thus repositioned by the juxtaposition with the other activities.

CONCLUSION In this paper, we focused on the "Question-Answer" function of the SOUTO system; the hypermedia authoring system, and classroom activities where learners make hypermedia compositions with SOUTO and discuss their compositions with the function mentioned above. Field tests of the system revealed that; 1. The Question-Answer function reconfigures social relationships in the classroom and thus creates a foundation for collaborative learning, 2. The Question-Answer function creates a field of informal talk, 3. The SOUTO system provides a foundation on which both formal and informal talks are constituted and woven together, and thus, transition between them is enabled. Based on these findings, we suggested that educational systems should be designed as mediators which hybridize school-like activities and non-school-like activities.

ACKNOWLEDGEMENTS This study is supported in part by the Grant-in-Aid for Scientific Research (B-11558013, 1999-2001) & (B13558023, 2001) from the Japanese Ministry of Education, Culture, Sports, Science and Technology.

REFERENCES Bakhtin, M. (1968). Rabelais and his world. Iswolsky, H. (Trans.) MIT Press Bakhtin, M (1984). Problems of Dostoevsky's poetics. Emerson, C. (Trans.) Manchester University Press Bödker, B. (1991). Through the interface: a human activity approach to user interface design, Lawrence Erlbaum Associates: Hillsdale. Edwards, D. (1990). Classroom discourse and classroom knowledge. In C. Rogers, and P. Kutnick (Eds.) Readings in the social psychology of the primary school. Croom Helm Engeström, Y. (1986). The zone of proximal development as the basic category of educational psychology, The Quarterly Newsletter of the Laboratory of Comparative Human Cognition, Vol. 8, No. 1, pp. 23-42. Mehan, M. (1979). Learning lessons: The social organization of classroom instruction. Harvard University Press Funaoi, H., Akahori, K., Itami, M. and Itoh, K. (1999). Development and Evaluation of a Hypermedia System for Collaborative Presentation, Technical Report of IEICE, ET99-13, pp.25-32 [in Japanese] Latour, B.(1996). Aramis or the Love of Technology, Harvard University Press Law, J. (1992). Notes on the theory of the Actor-network: Ordering, Strategy, and Heterogeneity, System Practice, Vol. 5, No.4, 379-393 Scardamalia, M. and Bereiter, C. (1993). Technologies for knowledge-building discourse, Communications of the ACM Vol. 36, No. 5, pp.37-41 Scardamalia, M. and Bereiter, C. (1996). Student Communities for the Advancement of Knowledge, Communications of the ACM Vol. 39, No. 4, pp.36-37 Star, S. L. (1987). The structure of Ill-structured solutions: Boundary objects and heterogeneous distributed problem solving. In Gasser, L. & Huhns, M. N. Eds.) Distributed artificial intelligence, Vol. II, Pitman. pp.3754