Educational Technology & Society 1(1) 1998
ISSN 1436-4522

Successful online course design:
Virtual frameworks for discourse construction

Anita Pincas
Lecturer in Education
Institute of Education, London University
20 Bedford Way, London WC1H OAL United Kingdom
FAX: +171-612-6534
Email: teedapi@ioe.ac.uk

ABSTRACT

A mental representation of the virtual context is a necessary basis of successful online conversations. Such a representation is impossible to create without a method of reference back to previous parts of the discussion in order to develop cohesive discourse. The paper surveys the way two different groups of students in recent online courses handled referencing conventions in asynchronous discourse and suggests a way of providing the scaffolding for virtual discourse construction.


* Manuscript received Aug. 27, 1998; revised Oct. 1, 1998

For this paper, an online environment is one of asynchronous textual interaction, as in email. Although it is possible to engage in synchronous text-based communication using chat programs of various kinds and to attach multi-media attachments to email messages, and also to exploit audio and video streams in web browsers for face to face (f2f) real time communication, the world is moving heavily into distance education of a kind that offers learners the opportunity to work at their own most convenient times. It is therefore most relevant to compare asynchronous computer mediated communication (cmc) with f2f communication in 'proximity'.

In order to learn collaboratively in such an environment, participants need to engage in what they can perceive as a normal discussion. For this to occur, contextual factors at least similar to those that surround conversational reality must be present in the online interaction. They will only be present if participants form an adequate mental representation of the virtual space and time within which they interact. Without such a representation, there is no framework inside which the discussion can be managed, and the usual determinants of speech events – addressee/addresser roles, turn-taking, routines, relevance – are confused or absent.

In all human communication, people eventually adapt to circumstances. But for learners, coming new to online discourse, and given a limited time within which to work, it is essential to promote their model of the new virtuality so as to provide the necessary framework. The paper will contrast insights gained from my experience in running and teaching on the following two online courses at the Institute of Education since 1993:

  • Certificate in Online Education and Training, an international online course for educators and trainers.
  • Master of Arts in Education, an international online course for English teachers.
  • The data contrasts two different models of online interaction - (a) a rather free open discussion, which is very commonly used by inexperienced course designers, but is much less successful than (b) a more lock-step approach, which the author prefers (Pincas forthcoming).

    The approach here accepts that collaborative learning has value and that it occurs successfully in online contexts – sometimes more successfully than in traditional classrooms - but that teacher-led instruction is extremely difficult if not impossible online. It follows from this that traditional concepts of teaching are not the most appropriate for online coursework; the focus shifts to the learning process.

    These matters will be the concern of educational research in the coming millennium (Barnett 1997 esp. p.110). The present paper assumes a shift of balance away from the older paradigm of teacher input followed by workshops, seminars, or projects, towards student-student collaboration assisted by teacher facilitation and response.


    Mental representation of the virtual classroom

    This is the key issue for both new tutors and new students unfamiliar with group discussion by internet. The vast unimaginable ether has to be internalised as a virtual classroom of some kind before meaningful and satisfying discourse can occur.

    In our traditional education, every occasion, or series of occasions, has an understood form in space and time. Spatially, it is usually a room, with seats, tables, whiteboards, etc., which itself is within a building, which may be part of a larger complex. Inside the room are the people who have ostensibly gathered for the learning process. Temporally, teachers and learners are aware of blocks of hours, days, weeks, terms, and years linked to the hierarchy of curricula, syllabuses, timetables, lectures, seminars, workshops etc.

    In our usual procedures, we are hardly aware of the forms within which the content of our behaviour is framed. Further, because these forms are so much part of our routine activities, we need to reorient when they change, e.g. if tables are rearranged, or a timetable is altered, but we can do so quite rapidly within the normal boundaries. However, reorientation becomes more and more difficult as the changes become more radical, as is the case for an online classroom.

    To replace the spatial features, an interface with a visual image may suffice. For instance, it is possible to create a physical metaphor like those in Figures 1 and 2*. For email students, the images can simply be given to students on paper. On the Web, each icon in Figure 1 acts as a hyperlink to a conference, and in Figure 2, which represent the learning area of the virtual college, each area leads to a further conference. If students working by the WWW receive the image of a familiar physical environment every time they log on, their representation of the new context will be greatly helped.


    Figure 1. Opening desktop in the FirstClass software


    Figure 2. Image appearing when college link is selected


    However, my own experience shows that without the images, a good mental representation can still develop because, even in the virtual environment, some aspects of the framework remain unaltered, namely the curriculum, syllabus, subject content, and requirements to read, learn and be assessed. What needs to be virtually reconstructed is the essence of the old classroom expectation, namely input from a lecturer and interaction among the participants.

    It is possible to go some way towards reconstructing the expected pattern of teaching. Teacher input may be given in live videoconferencing, or in video or audio tapes. My students have told me that watching videotapes not only helps their subject learning, but gives them a feeling of "being there", while audio tapes allow them to listen while driving, working, etc. But there remains the disjunction between themselves and such films of remote people and classrooms in which they are not themselves present.

    The physical reality for students online is a computer screen. However appropriate filmed teaching may be, if the learning is to develop as a collaborative enterprise between students, interaction of a fruitful kind has to be established. The computer screen therefore needs to be supplemented by an abstract discourse framework within which textual discourse can occur. It is, in fact, not so far removed from familiar abstract frameworks we already use. Everyone is nowadays accustomed to communicating in spite of a disjunction of time and space, by letter, fax, email, telephone, radio and television.

    What is totally new in computer mediated communication (cmc) is ongoing group discussion in a new temporal and spatial pattern. Users need a framework to overcome the difficulty of seeing connections between messages that appear in a linear series unrelated by familiar signals like adjacency, speakers’ gestures and expressions, response expectations in f2f situations, and so forth. An efficient framework for online discourse will have a set of features, starting with the outer boundaries and working inwards through a variety of structures.

    There are two major types of framework: the collaborative and the co-operative. In the former, students do individual work but share it with each other in a helpfully critical discussion. In the latter, they essentially work together on tasks, e.g. projects or essays.

    For a collaborative framework the outer boundary is the syllabus as embodied in a timetable, where, for instance, each week’s work might be delimited to a specified content so that users know what to expect the messages to be about. Each week’s work is further broken down into small segments which can be parcelled out to individual students as indicated in Figure 3. Fuller details of what occurs during the week are shown in Figure 4. The result of such a structure is not only that every student can place him/herself exactly within the framework, but also that, as in all normal discourse, people function best when they know exactly what is required at any point. The smaller the units of content to be dealt with, i.e. the more specifically the content is defined by the timetable, the more successful the discourse is likely to be.


    Figure 3. Weekly task division


    Figure 4. Syllabus timetable


    For a co-operative framework, the outer boundary is also the syllabus, but instead of a fixed timetable, there are activities within it that are more flexible and may depend on students’ own work patterns. In a co-operatively conceived course such as that described in McConnell (1994) for instance, students themselves decide how to group themselves, how to work together, and what their specific goals are. In such a work-pattern, it is necessary to keep co-operative groups small (McConnell suggests 4 or 5 people (op cit. p.97)) because frequent interaction and decision-making is needed. My own view is that this will be a very difficult model to manage if members of the group are dependent on the others to get their own work done; they are likely to become frustrated by even small delays. If however, as in McConnell’s course, they are each pursuing their own goals while helpfully contributing to each others’, then by all accounts it can be very successful.

    With either kind of framework, the interactions that occur within it depend on establishing role relationships, routines of behaviour, and turn taking. The first two are easier to deal with than the last.

    Social contact of some kind can be effected even among distant participants who have never met. By virtue of the fact that they are all students on the same course, many aspects of their relationship are already given. Photographs and biodetails can be sent to all. Based on these fundamentals, it is possible to encourage social interaction preferably before a course starts. In my experience, students readily tell each other about themselves, discuss current events etc.

    Such interaction ought to be kept as separate from the academic work as it is in traditional contexts. A conference called "Social Chat" or "Café" or "Bar" is now the traditional way of encouraging informal exchanges. In my experience it is best for lecturers not to take part, so that students feel they genuinely have their own space. In addition, I normally remind them that they can quite easily set up a group email for themselves alone in order to talk truly privately without the knowledge of the institution, as is normal on campus.

    Within a given framework, routines can also be fairly easily established. The timetable and structure shown in Figures 3 and 4 already do this to a large extent by asking for regular student tasks and commentaries on each others’, as well as allowing different spaces for plenary issues and social chat.

    The real difficulty lies in aspects of discourse related to turn-taking which is at the heart of conversation but is absent online. In cmc, everyone can "speak" without interruption at any time. Simultaneous multiple threads of discussion are common, and are indeed one of the advantages of cmc collaboration (Howell-Richardson 1995). These factors make it difficult for interactants to transform a series of written messages on a screen into a mental representation of a discussion.

    In f2f contexts, members of a discussion can, in spite of lapses, hold the unfolding events in their minds well enough to feel that they are keeping abreast. They are helped by being in a room, by having a bounded timeframe, and by seeing the others. Crucially, listeners expect to forget a large amount of what is said. They focus on the here-and-now, but have at their disposal established modes of turn-taking and conventionalised ways of referring back to previous elements in the discourse. These help the interactants to bind different speakers’ contributions into a mental representation of the discussion. Without them, they could not integrate the various messages into a cohesive whole and the interaction would be felt to be random or even meaningless.

    Because of turn-taking and high immediacy, a f2f context rarely shows an original message being tediously repeated. Successive turn are linked to the previous one either by simple contiguity in an adjacency pair, sometimes with brief linking devices, such as "What’s more", "No, but", "Well", or, where the adjacency has been interrupted, with brief signals like "Going back to what Jack said before...", "On the point about language that Alice made,...", etc. They can fairly safely assume that most of the preceding discussion is fresh in everyone’s mind, as long as the interaction has a manageable time span and the topic has a narrow focus, as in most educational seminars.

    In asynchronous group cmc, the usual default order in which messages are presented, namely chronological, is the analogue of a f2f discussion. Why then does it create problems? First, because, as has been pointed out, the factors of space and time are dislocated. All that any individual is aware of is the block of messages available here and now in the computer, having arrived there at different times and each one not necessarily related to the preceding. Continguity is not relevant.

    Second, writing has the effect of changing users’ expectations. Whereas they accept that they will forget a great deal of evanescent speech, they now expect to have total access and therefore some kind of total recall, with the accuracy that graphic records encourage. Ideally, they would want to be able to "flick back" to previous messages. But this is impracticably cumbersome in current online systems.

    These features of cmc taken together can have an effect rather like people in a room all talking at once - it becomes "noise" and can be unnerving in the extreme. However, even f2f talk can be hard to follow if there are more than, say, five speakers. Many threads are started and dropped, the focus often moving onto one clear thread favoured by one or two speakers who may establish dominance by loudness of voice, eye contact, body movement and many other kinesic manoevres. At best, dominance has the effect of simplifiying the discourse by narrowing it, but this is almost impossible to achieve in a series of written messages. Even print in bold, italics, underlined, capitalised, or coloured form does not have the desired effect, as in f2f, where loudly taking the floor holds everyone’s attention. In cmc, there is nothing to prevent all from contributing many messages of any length.

    In comparing how participants handle the cmc situation in the two courses that provide the data for this paper, it is strikingly obvious that the lack of a sufficient framework in the Certificate, compared with the strict framework of the MA (Figures 3 and 4) made a crucial difference to their ability to create the necessary mental representation (Pincas, forthcoming).


    Lack of structures in the certificate course

    The Certificate, while establishing an outer framework of a timetable and modules marked for different topic areas, as well as the division of students into smallish groups for purposes of discussion, did not go as far as the MA in detailing precisely what each student was expected to do and say online. In other words, the framework of Figure 3 applied to the Certificate without the detail inside the box called Week 1. For the Certificate, that box contained a list of readings and some discussion pointers. It was then left to the students to engage with each other.

    The data shows that they found this very difficult indeed. The essence of the problem lay in their feeling that they could not rely on their readers’ memory of past messages or in their readers‘ ability to recover those messages for reference. They strove, therefore, to develop ways of referring back which are quite foreign to f2f discussion. In addition to a reasonable use of the cohesive and referring devices of ordinary language, they resorted to a strikingly high use of actual quotations from previous messages. They were clearly aiming to establish some kind of toe-hold, a framing device for each message of their own.

    Their attempts were driven, it seems, by their perceived need to ensure that their reader had the same mental representation of (at least part of) the preceding dialogue as they had. In terms of the now well-known maxims for successful communication (Grice 1975), cmc presents people with a conflict. While striving to be relevant and informative with maximum brevity and clarity, discussants are frequently unsure whether they need to refer back or not. When they decide that they should, they are torn between citing previous messages in full or being as brief as possible. The maxim of brevity would urge them to abridge their citations, but they face a serious problem of how to do this without losing relevance and clarity.

    The difficulty is exacerbated by the email option of responding to a message immediately upon reading it, which gives the respondent the impression of being the second turn in an adjacency pair. Yet experienced cmc users know that this is misleading because the addressee may not receive the two messages in sequence. Perhaps because of this, many people choose other cohesive devices instead or in addition to the Reply facility. They may use the subject header (E.g. "John’s proposal", "Answer to Mary’s question"), or incorporate their reference into the body of their own response (E.g. "I believe, as John also seems to,...."), both of which require more work on the part of the addressee to observe the necessary links.

    Furthermore, it was evident on our course that writers used quotations for cohesion and coherence even when the available software signal (Reply) made it unnecessary. Since Reply leaves the subject header unchanged, a visible "thread" of messages is created if two or more users Reply one to the other in a linear fashion. In the software that these participants used (FirstClass) it was actually possible to sort messages by their subject headers and to read them in sequence, their cohesion being given just by that sequence of headers. Yet people still found the need to refer back and quote.

    This is an important point, because it means that the need was based, not on logic, but on users’ feelings about clarity and relevance. And it was interpreted flexibly and variably by users themselves as the discussion developed. Of course, this is not surprising since the discourse "rules" that have been discovered are, on the whole, as vague as the Gricean maxims, being merely a touchstone of commonly understood, acceptable behaviour.

    In short,when translating oral referencing conventions to written forms, users are faced with two different kinds of choice:
    1. How much of the original message to cite or paraphrase
    2. How to present the reference graphically.

    They experimented in many different ways. First, they tried different modes of presentation. For instance, when they saw a Reply message automatically starting
    "<X> writes:..."
    they corrected it to
    "<X> wrote/said:",
    followed by
    "me here:" or
    "<my-name> writes:"
    and then their own response.
    Or they tried using colour (though once messages are saved, the colours would only be seen by readers who have colour-sensitive word processors or printers).

    Second, they experimented with abbreviating the original message they were referring to. There is clearly a skill in this, and students’ choice of method became more complex as the discussion grew. Their methods varied from citing the whole preceding message (or set of messages if other people had already started commenting on it) through citing only part, to giving some paraphrased version or summary.

    The upshot was that many felt very confused indeed and found the ongoing discussion impossible to follow. The general term for this in cmc descriptions is "overload", though it tends to refer to the quantity of messages more than to the lack of structures of interaction. It is a frequently cited problem but can be solved, as is apparent from the present study, by increased scaffolding within the framework of the discourse.


    Ease of reference in the masters course

    The MA course was arranged with an general framework very similar to that for the Certificate, but it had, in addition, a detailed plan for each week, as seen in Figures 3 and 4. In such a structure, all that is further required for successful reference in the actual online discourse, is that each message be labelled properly. This is done by means of subject headers. Figure 5 shows a sample of these, where the left column is the actual header to be used, and the right column shows what it signifies. Figure 6 shows the email list (in EudoraPro) for one week’s online work.


    Figure 5. Sample list of headers relating to a timetable for one of the modules in the Masters course (Communicative Language Teaching (CLT))

    A

    17:44 16/02/98 -0800

    [Fwd: B FN1- 11 Comment3]

    B

    12:18 07/02/98 GMT

    A - FN 1 - 2

    C

    11:52 12/02/98 +0000

    A - FN 1 - comm

    C

    15:18 16/02/98 GMT

    A - FN 1: Summary (addendum)

    C

    12:19 14/02/98 GMT

    A - FN1 - 1

    E

    20:35 11/02/98 PST

    A - FN1 - 10

    F

    18:57 12/02/98 -0600

    A - FN1 - 10 Comm3

    G

    12:06 13/02/98 +0100

    A - FN1 - 10 Comm 4

    E

    17:10 12/02/98 PST

    A - FN1 - 11

    F

    21:58 08/02/98 -0600

    A - FN1 - 12

    F

    22:40 08/02/98 -0600

    A - FN1 - 2 Comm1

    G

    19:01 09/02/98 +0000

    A - FN1 - 3

    F

    22:54 10/02/98 -0600

    A - FN1 - 3 Comm1

    E

    10:55 11/02/98 +0100

    A - FN1 - 4

    B

    17:46 12/02/98 GMT

    A - FN1 - 6 Comm 2

    E

    13:00 13/02/98 +0000

    A - FN1 - 6 Comm3

    H

    16:46 12/02/98 -0800

    A - FN1 - 8

    E

    19:12 13/02/98 +0100

    A - FN1 - Comm

    B

    17:46 12/02/98 GMT

    A - FN1 - Comm

    C

    23:55 15/02/98 GMT

    A - FN1 - Summary

    A

    13:36 08/02/98 -0800

    A - FN1 10

    A

    08:34 10/02/98 -0800

    A - FN1 10 Comm 1

    I

    23:51 12/02/98 -0500

    A - FN1 5 Comm1

    D

    13:22 13/02/98 +0000

    A - FN1 5 Comm2

    Andrea

    16:22 13/02/98 -0800

    A - FN1 5 Comm3

    J

    23:51 12/02/98 -0500

    A - FN1 9

    K

    17:49 12/02/98 +0100

    A - FN1-10 Comm 2

    K

    17:59 08/02/98 -0800

    A - FN1-5

    K

    20:57 09/02/98 -0800

    A - FN1-6

    D

    16:57 12/02/98 +0100

    A - FN1-6 Comm 2

    L

    13:35 08/02/98 +1100

    B - FN1 - 1 ( Complete )

    Anita Pincas

    12:56 09/02/98 +0000

    B - FN1 - 11 ( comm 1 )

    M

    01:07 13/02/98 +0900

    B - FN1 - 11 Comm 4

    N

    01:23 11/02/98 +0000

    B - FN1 - 11 Comm2

    L

    13:49 11/02/98 +1100

    B - FN1 - 12 ( comm 1 )

    O

    16:03 13/02/98 +0900

    B - FN1 - 12 Comm2

    P

    20:45 08/02/98 +0000

    B - FN1 - 3

    Q

    11:35 08/02/98 +0700

    B - FN1 - 3

    R

    11:07 09/02/98 +0900

    B - FN1 - 3 Comm 1

    A

    08:30 18/02/98 +0000

    B - FN1 - 3 Comm 1

    M

    02:51 12/02/98 +0900

    B - FN1 - 6

    A

    14:58 15/02/96 +0000

    B - FN1 - 6

    M

    08:46 13/02/98 +0900

    B - FN1 - 6 Reply to Comm

    R

    23:10 12/02/98 +0900

    B - FN1 - 6 Comm 1

    A

    14:58 15/02/96 +0000

    B - FN1 - 6 Comm 1

    A

    08:30 18/02/98 +0000

    B - FN1 - 6 Comm 1

    R

    00:08 09/02/98 +0900

    B - FN1 - 7

    A

    14:59 15/02/96 +0000

    B - FN1 - 7

    A

    08:30 18/02/98 +0000

    B - FN1 - 7

    R

    10:31 09/02/98 +0900

    B - FN1 - 7 Comm 1

    A

    14:59 15/02/96 +0000

    B - FN1 - 7 Comm 1

    A

    08:30 18/02/98 +0000

    B - FN1 - 7 Comm 1

    P

    08:16 13/02/98 +0900

    B - FN1 - 7 Comm3

    N

    21:15 13/02/98

    B - FN1 - 7 Comm4

    T

    19:31 10/02/98 +1000

    B - FN1 - 8

    G

    15:24 14/02/98 +1100

    B - FN1 - SUMMARY

    U

    04:02 13/02/98 -0800

    B - FN1 Task 1 Comm 1

    U

    02:05 13/02/98 -0800

    Fwd: B - FN1 Task 10

    P

    00:54 12/02/98 +0900

    B FN1 - 5

    P

    01:17 13/02/98 +0900

    B FN1 - 6 Comm

    A

    14:59 15/02/96 +0000

    B FN1 - 6 Comm

    X

    16:17 18/02/98 -0400

    B FN1 11

    X

    21:12 12/02/98 -0400

    B FN1 7 Comm 2

    X

    21:11 12/02/98 -0400

    B FN1 9 Comm2

    W

    11:26 11/02/98 +0200

    B FN1- Comment

    S

    04:09 12/02/98 +0900

    B-FN1-9

    R

    23:19 12/02/98 +0900

    B-FN1-9 Comm 1

    A

    15:00 15/02/96 +0000

    B-FN1-9 Comm 1

    A

    08:30 18/02/98 +0000

    B-FN1-9 Comm 1

    Anita Pincas

    11:18 14/02/96 +0000

    FN1 - Anita's comment


    Figure 6. Email headers for one week of the masters course


    It will be seen that every header signals the week’s topic and the task number. This is an eminently simple and learnable system. Because the week’s work has been broken down into a small number of questions, people can easily understand its overall structure and form a mental representation of the discussion within these boundaries.

    Moreover, it overcomes the problem of how to refer back to previous messages. If Reply is selected then it has the same subject header as the original, and the respondent can add something like "Comm" to signal a comment on it. If headers are part of an already understood framework, there is a clear pattern of messages and responses.

    In my courses – 6 separate cohorts since 1994 – the system of headers has been the pivot about which their discourse turns, and references are very easily handled. There have been no indications at all from the students that have followed this pattern that they find any difficulty at all with it.

    Nonetheless, there are variations in their procedures, which depend upon factors other than the simple need to refer back. Some do plunge in, relying on the subject header; some quote the previous message or part of it, others paraphrase it. Other students address each other as at the start of a conventional letter, using "Dear <……>", "Dear<….> and everyone", "Dear Everyone", " Hi everybody", or even merely a self announcement like "Ta Da!" or " Sorry to all group A members". See Figure 7 for partially classified list of the different modes of reference.


    FULL CITATIONS- 5 +/- quotation from earlier message
    Secondly James said in CLT3 - 5 comm 1 ( and I completely agree with him "How do…
    I liked David’s comment on Littlewood’s terminology in B - CLT3-2
    In A - CLT3-11, James says "some focus on form is needed, the
    As Sheri said in her CLT3-3 comm, "Most native
    I would like to reiterate a point I made in CLT3, Task 3.
    Quotation of workshop question
    "To what extent should language teaching be a formal study of language or simply involve the learner in natural communication?" Sherri certainly ended…

    INCOMPLETE CITATIONS - 4 (assumes the conference is known) +/- quotation from earlier message
    I think that James makes good points in comment 1.
    …(Hugh's Task 3) …
    Kah Leng pointed out in her comment on my task #9, that …
    I would like to make some brief comments on the questions David (S) raises in the 3rd section of his task answer

    REFERENCE + paraphrase
    Topic
    - Oblique 3rd person reference
    I would like to add something to the issue of...
    The question of learner errors and how best to respond to them occupies a….
    Dear Sherri, In relation to your $64,000 question,
    Second person
    Your distinction reminds me of …
    Third person (+ paraphrase)
    plain
    Secondly, James said...
    As Barbara says...
    As James points out
    Sherri points out the issue of …
    …kind of feedback James mentions ("on-line type within the context of the…
    It appears from the lively discussion you have sparked off, that …
    Dear Deborah, I wanted to reply to your comment but didn´t have time.
    Alejandro reminded me of my promise to send to you some …
    grateful
    First I’d like to say how interested I was to read Deborah’s...
    Dear David, your questions were helpful for me, and I hope this mail will be a
    Thank you Fiona and Hugh for...Thanks Sheri for...
    Dear James, Thanks for finding the time to provide some theoretical background …
    agreement
    I would like to agree with Joanne when she says...
    As James, I think...
    Dear David When I was reading about Community Language Learning I thought it is a very interesting approach. However, I would agree with you that …
    I guess this might be the case with Fiona’s..
    I was very interested in Deborah’s attempt to...Deborah’s system might be...
    I think that James makes good points... I would compliment what James says
    This would complement what James said about the student's focus…
    I agree with what Adriana says
    I couldn’t agree with Hugh more....
    Kah Leng pointed out...
    I very much agree with David Turner's sentiments…
    As Barbara says Littlewood divides feedback into two categories.
    I would like to agree with Joanne when she says that it doesn't really matter
    First of all, I'd like to say that I side with Keh Leng when she said,
    …that it is possible to correct them as James suggests by recasting, but this is…
    I fully agree with you when you urge the theorists …
    With this in mind, Hugh draws some very considerable conclusions.
    Like Hugh, I really do find …
    evaluative
    Another exquisite served up by Chef Jon. It wasn't so much an Hors d' Oeuvre,...
    In enjoyed Sheri’s task on...
    I would like to make some brief comments on the questions David (S) raises in the 3rd section of his task answer, as I found them very intriguing
    I liked your definition of your role as 'a pointer out of things' and I am

    Figure 7. Samples of references during one week’s work on the MA


    Figure 8 gives an indication of the pattern and quantity of interaction between the 21 online students during one week of a ten week module. Research into the quality of this interaction is ongoing, but initial inspection shows that it is very rich indeed when compared with typical classroom workshop talk.

    Column 1 : 21 students did one task each, ie 21 tasks.
    Column 2 : between them, they produced, in addition, 28 comments on others’ tasks.
    Column 3 :number of comments made on each of the 12 tasks. It can be seen that this is not a mechanical activity, since different tasks aroused different degrees of interest.

    Word count and time spent
    The tasks and comments constitute 23,846 words, which, at the rate of speech of 150 words per minute, would amount to a seminar lasting 2.6 hours
    21 students produced 49 messages, ie an average of 2.3 messages each.
    The average word count output per student is therefore: 1036 wds per week, or 10,367 wds per term.

    Figure 8. One week’s tasks and comments


    Conclusion

    In f2f situations, one type of solution to the difficulties of discussion among large numbers of people is seen in the highly structured committee context, where convention gives a chairperson the right to enforce an agenda and hold discussants to prescribed turns and roles. In cmc, it is only possible for a chairperson to manipulate discussion after messages have been posted, e.g. by removing some or rearranging their order. But this is merely to repair the interactions, not guide them let alone control them (except to the extent that the discussants might learn a mode of procedure).

    But the great benefit of cmc systems is that they allow us to step outside the single linear conversational model we are bound to in normal talk, and exploit the ability to run several discussions at the same time, the possibility of everyone making contributions without interruption, and so on. One way of assisting this new kind of discourse is to structure the framework, as proposed in this paper. It is only when management of discourse ceases to be a problem that participants can fully develop their collaboration and share ideas in a fruitful learning environment.

    Unfortunately, experience of listservers and numerous other cmc discussion areas reinforces the impression that open-ended discussions without a clear scaffolding will remain the norm in many circumstances, with all the concomitant discourse difficulties described above. It is hope that the present paper will contribute to a perception of the problem, if not to a complete solution.


    Acknowledgement

    * Figure 1 is the opening desktop in the FirstClass software used by Abacus Learning Systems for the training courses, and Figure 2 is the image that appears if the College link is taken. They are produced with the kind permission of Abacus Learning Systems <learning@abacus-uk.com >.


    References

    • Barnett, R. (1997). Higher Education: A Critical Business, UK: SRHE & Open University Press.
    • Grice, H. P. (1975). Logic and Conversation. In: P. Cole & J. Morgan (Eds.) Syntax and Semantics 3: Speech Acts, NY: Academic Press.
    • Howell-Richardson, C. (1995). Interaction across computer-conferencing. In: R. Howard & I. McGrath (Eds.), Distance education for language teachers, Clevedon, Avon: Multilingual Matters Ltd.
    • McConnell, D. (1994). Implementing Computer Supported Cooperative Learning, London: Kogan Page.
    • Norrish, P. (1990). The Graphic Translation of Text, PhD Thesis, University of Reading, United Kingdom, p.16.
    • Pincas, A. (forthcoming). Mental representation of cmc discourse: the importance of cross-referencing.

    decoration