The twin concerns of this chapter may each seem to have a circumscribed quality. Writing is surely something sufficiently clear that we can spot it when it is in progress. For example, if some friend should seek a photograph of "you writing", it would be easy to supply a convincing image. Similarly, ICT is also easy to spot. Typically, it will be identified with the material form of the desktop computer. This technology has a clear thing-like quality. Such circumscribed characteristics of ICT and of writing might encourage a certain slippage in conceptual vocabulary. It becomes natural to conceptulise writing as a kind of behaviour. It becomes easy to conceptualise the computer as a kind of stimulus, a kind of tool-to-hand. Then, perhaps such a perspective cultivates the belief that questions about "the effects of ICT on writing" are straightforward. We argue here that they are not..

In effect, neither item in this ICT/writing relationship has such a singular identity. Consider the text that you are currently reading. Composing the preceding paragraph certainly involved keyboard tapping and screen staring: activities that perhaps could be photographed as "me writing". However, a lot more was involved than those simple behaviours. For example, there was the business of distributing attention. Certainly this applies to the screen, the keyboard, and a set of paper notes. But also – because of a social interruption – it applies to the screen and other forms of event on the periphery. Writing is organised (located, scheduled) to be in a useful harmony with this periphery: sometimes exploiting it, sometimes needing to be insulated from it. So, there is a place where this keyboard tapping is carried out. In the present case, my site-for-writing conveys a sense of design, or intentionality. That is, the place is one arranged to best orchestrate the management of certain material resources, and to filter out events that are thought to be distracting. In addition to this physical context, writing involves a framework of more abstract constraints associated with the task: for example, matters of deadlines, publishers, editors and audience. All of these are built into the organisation of what gets done, including the various technologies that support it. Finally, the present piece of writing is co-authored: a circumstance that clearly disturbs any orderliness in our photographic capture of writing-in-progress.

Thus, our first point is that writing can not be circumscribed in the convenient manner that might support simple research designs. Instead, writing seems to confront us with a rich system of activities. Moreover, these will be realised by individuals in very different ways, depending on their histories of enculturation. Indeed, "cultural practice" may be the best way to frame the achievement of writing. That is certainly the perspective to be adopted here. Our second point concerns how best to conceptualise the writing/technology relationship. Here we propose that they are mutually constituative. This captures the sense in which the second term in our interest – "computers" - can not be any more easily circumscribed than "writing". Computers (and all their infrastructures) are merely inert pieces of material, at least until they are drawn into forms of human activity. If we are sometimes given to identify "properties of the technology", this is because we have noted something orderly about its appropriation into an activity system where we find it operating. Similarly, if we come to refer to "properties of writing", it is because some author is acting in an organised way with technology – ICT perhaps, but a wide range of other technologies could be implicated. Thus, these items are inherently interdependent.

Such a conceptual framework is readily derived from cultural psychology (Cole, 1996). Study of activity in cultural psychological terms entails adopting as an analytic unit the individual-acting-with-mediational-means. That is, psychological phenonmena are everywhere seen to entail activity mediated by some cultural resource or other – an artefact, technology, symbol system, social practice etc. Unlike the im-mediate behaviour of other species, the activity of human beings involves engagement with a material and social world through the intervention of these cultural resources. This invites an analysis of human action that stresses its inherent embededness in culture; observers may thereby strive to describe the relevant dynamics that follow. Here we are addressing one particular mutual engagement with culture – one that concerns the inter-penetration of writing and new technology. In adopting a cultural perspective, we are anxious to resist simple formulations that appeal to "the effects of ICT on writing". Yet we are very aware that new technologies must be deeply implicated in the writing practices of many authors and the nature of that relationship must demand some form of rigorous undersatnding. The challenge, therefore, is to capture honestly the relevant dynamic.

So far, our analytic preference has been presented in rather abstract terms. It may be useful to invoke a more concrete example, by way of analogy. In discussing mediation and its effects, Cole and Griffen (1980) seek a simple parallel for probing the way we speak about the influence of modern technologies that support cognition. They note that it is tempting to speak of "cognitive amplifiers". So, in the present context, some may wish to speak of ICT as amplifying the powers of writers. However, this may be misleading – both in terms of how it conceptualises the underlying activity (writing) and how it conceptualises the impact of the technology (amplifying). To pursue this, Cole and Griffen offer an analysis (somewhat macabre) of the social practice of hunting and killing. They invite us to imagine a traditional society in which prey are captured and killed with simple weapons. Visitors from some modern society furnish these people with guns – a new mediational means to enter their system of hunting. More animals are killed in shorter periods of hunting. Accordingly, we may be tempted to say that the guns served to "amplify" killing. Just as we may be tempted to say that computers amplify the more cognitive enterprises of calculating, writing, or whatever. If all we mean by amplifying an activity is an increase in output – for example, more animals getting killed – then this seems an innocent enough way of talking. However, it is less obvious that the hunter’s capacity to kill has been "amplified" when the new weapon is not to hand. It is not some property of the individual that has been changed by the technology; what has been changed is the manner in which some activity can be carried out - when the technology is available..

The example is helpful in sharpening our sensitivity to the three issues at stake here – writing (cf. killing), ICT (cf. guns) and re-mediation (new technologies entering existing cultural practices). First, killing, like writing, is no rigidly defined pattern of behaviour. It involves practices of social coordination – gathering, stalking ambushing – as well as practices of recovery, distribution, honour and so on. In short, the human action invoked in these relationships has a systemic character. Second, the technologies involved are similarly complex. Guns are not artefacts with some singular nature. They derive whatever properties are ascribed to them from how they enter into cultural practices. A gun is different according to it being a starting pistol, a rescue flare, a fairground challenge, or a hunting weapon. Its identity is constituted by the systems of activity with which it is involved. Finally, the parallel pursued here encourages us to notice how the relationship between such activity systems and technologies is itself complex. It is most sensibly considered as a relationship of re-configuration, not enhancement. Killing is not amplified by new technology. Re-mediation involves not so much amplification of some activity as changing the manner in which it is organised or exercised. Guns arrive: the hunting is done differently. ICT arrives: the writing is done differently.

This argument is well developed by Bruce (1997) It is an argument that denies the value of conceptualising writing and technology in terms of "separate realms". Yet examples are needed to reinforce the analysis that this invites for the case of writing. We wish to develop a particular example in the present chapter. We have chosen the case of undergraduate writing. It is authentic, richly-structured and accessible to research. Our aim must be to draw attention to the sense in which it is an activity system: that is, individuals coordinating with a wide variety of cultural supports in the interests of producing text. Our particular interest is in positioning ICT as a developing influence within such systems of activity. Research on ICT and writing tends to dwell upon word processors: how their design re-mediates composition (Erickson, 1992). Yet the increasing use of this one particular technical tool does not suddenly identify writing as something that is now "about" technolgy or something that is now "effected by" technology. Writing has always entailed an activity engaged with technologies – not least pens and paper, but also a whole range of cultural resources that frame up what we do when we say we write. Again, the undergraduate example should illustrate this well. The particular case of ICT serves to capture the richly mediated nature of this activity.

In what follows, first we provide a general introduction to the circumstances of undergraduate writing as a usful model system. Then we organise a discussion around five topics involving the intrusion of ICT into writing. We suggest that each of them furnishes a useful focal point for research. Not that this list is intended to be exhaustive – it does not, for example, touch upon the design of word processor features. The aim is not to partition the domain of interest into five comprehensive areas that identify five independent variables for researchers to study. Certainly each section does have a focus on a particular mediated aspect of writing. However, these are not conceived as independent variables in experiments. Any comparitive analysis that this conceptual organisation provokes is not in the spirit of evaluating "effects". It is more to create a device for drawing attention to the structure of the underlying practice – a structure that might otherwise be hidden from view by virtue of its sheer familiarity. We do this first in relation to the physicality of writing and text – considering the ICT context of writing on a screen. Next we consider how writing is shaped by a technology that is connected to a network of other computers and computer users. Thirdly, we discuss text as traffic within a community infrastructure of ICT. Fourthly, we consider text in relation to audience and the role of ICT in that relationship. Finally, writing is located within a social context of appraisal and evaluation.

The assumption that writing and ICT describe separate realms encourages a particular form of experimentation. Teachers and researchers alike ask what effects technologies have on literacy practices. These literacy practices are viewed as skills that might be isolated from the peculiarities of their situation. Accordingly, experiments are conducted that attempt such isolation. An experimental task might be designed to embody a specific literacy skill (or subset of skills). The task then isolates this skill from others, and from the material conditions of everyday literacy, so that technology’s "effects" upon it can be established. This is most clearly seen in the way in which the experimental research literature separates reading and writing. Experimental reading tasks almost never involve the participant in any act of writing (see Dillon (1992) for a literature review). And although experimental writing tasks unavoidably involve writers in reading their own texts, the separation between reading and writing is achieved as far as possible by excluding the use of source texts from the task (see Ransdell and Levy (1994) for a partial review). Under this approach attention is heavily focused upon outcomes with relatively little concern for processes. If processes are investigated, it is typically with the aim of establishing how technologies change the frequency and sequence of those with an abstract and immutable quality – such as Hayes and Flower's (1980) processes of generating, organising and translating ideas (e.g. Kellogg and Mueller 1993). This approach has made only modest progress and has generated some inconsistent findings (Bruce 1997).

The assumption that literacy practices are comprised of abstract skills that can be studied in isolation from the peculiarities of the situation (including writing tools) and in isolation from each other is problematic. In a review of the experimental literature comparing reading from paper and reading from VDU's Dillon (1990) notes "One is struck in reviewing this literature by the rather limited and often distorted view of reading that ergonomists seem to have. Most seem to concern themselves with the control of so many variables that the resulting experimental task bears little resemblance to the activities most of us routinely perform as 'reading'." (p.1322). The equivalent criticism could equally be applied to the literature on writing. An experimental writing task typically involves composing a short text, in a single session, in the absence of source documents (Snyder 1993) and, as noted by Torrance (1996), often in an unfamiliar genre.

Authentic literacy practices that occur outside of the laboratory take forms that bear only slight resemblance to the isolated, sparsely resourced, tasks enacted inside it. The undergraduate coursework essay writer is embedded in rich contexts including material artefacts such as, journals, books, computers and the internet. The student may also fashion additional resources such as lecture notes, notes from reading, written outlines, and drafts. In a study of genuine coursework essay production (Dymott and Crook 2001a) it was found that mediation by such materials was a pervasive feature of coursework essay production. Great diversity was also found across individuals in terms of the form taken by mediated activity. This mediation was sufficiently strong, and constitutive of sufficiently diverse forms of practice, as to call into question the validity of even apparently basic distinctions between literacy skills, such as between note-taking and drafting. The inability to find a mapping between authentic, richly mediated literacy activity and the literacy skills investigated in the lab should lead us to question the applicability of laboratory findings to authentic activity, and perhaps to question the value of the notion of generic, context independent literacy skills.

In the sections that follow, we draw upon our own research to pursue these reservations and to locate them in relation to new technologies. Generally our research has involved the study of students working under natural conditions of writing and assessment. We have invited them to keep detailed diaries, we have interviewed them, and we have logged the manner in which they use their personal ICT resources. In some cases, we have seen an advantage in a more controlled form of study in which different circumstances for writing are observed and compared. In such cases we have protected the authentic goals and motives of participants and involved them with genre of composition that are familiar. Such an orchestrated research contrast between mediational means can remain useful as a device for exposing the significance of some concealed dimension of the writing situation. In fact, this strategy is illustrated in the next section, where we dwell upon the materiality of writing – as accessed by consideration of how screen-based composition re-mediates what gets done.

A general caution is necessary before proceding. The position we have adopted can create problems with our vocabularly for talking about "writing" here. We are urging that it not be considered in so circumscribed a way that we fail to pursue the rich mediational framework in which it is set. The very term "writing" should be orienting us towards a complex activity system and not constraining us only to consider the canonical image of exercising pen-and-paper or keyboard-and-screen. Yet for analytic progress, there is a need for anchors to be made into human activity and to features of the designed environment. Accordingly, in some of what follows we may sometimes chose to make "writing" refer to the local circumstances of generating text – momentary actions with pen, keys or whatever.

The physical qualities of on-screen text is clearly different from text on paper. These differences have significance for readers’ and writers’ experiences of text. We have investigated students’ activity as they composed short essays; once from sources that were presented on A4 paper, and once from sources presented on a computer screen (Dymott and Crook 2001 [b]). The computer screen constituted a smaller working space than the physical desk top. Whole pages of paper text would be visible at any one time, but the screen typically displayed much less than a single page. Participants often laid out paper so that more than one document was visible at any time. Only one participant attempted this with on-screen texts and accomplished it only with great difficulty. Indeed most manipulations of documents were more readily performed on paper than on screen: paper offered a far more 'direct' (Hutchins, Hollan & Norman 1986) form of manipulation. Scrolling within a document typically involved a pause in reading while visual attention was shifted to an on-screen scroll bar. Yet the tactile properties of pages meant that they could be turned without distracting visual attention from reading. Numerous such differences between paper and screen have been associated with different experiences of paper and computer texts in both laboratory studies (Hansen and Haas 1988) and anecdotal reports (Chandler 1995). One of the more reliable findings from all lines of research is that users of on-screen writing have greater difficulting apprending the documents’ global structure, or developing a ‘sense of text’. Hansen and Haas (1988) offer a framework to explain such findings in terms of the ‘page size’, ‘legibility’, ‘responsiveness’, and ‘tangibility’ of paper and on-screen texts.

It would be wrong-headed to expect the computer screen’s physicality to influence literacy practices consistently across individuals. A supposedly 'given' task may be performed quite differently by different individuals who have their own preferred ways of accomplishing it. The participants in our screen/paper comparative writing task had previously taken part in a more naturalistic study documenting how they produced a genuine coursework essay. In the genuine coursework situation they had differed quite substantially from each other in how they worked with source materials. In our comparative situation many of these differences were preserved. Participants constructed different roles for the source texts, in line with their established reading practices. What mattered about the materiality of the sources was therefore different for each participant: it depended on the specific roles sources played in their writing practices. For one participant, the technologies would influence performance by facilitating or interfering with her established practice of skim-reading and moving frequently between documents. For another participant, the technologies would ‘influence’ performance by facilitating or interfering with his established practice of visiting each document only once and reading it thoroughly from beginning to end. Whether or not the the screen 'inhibits' or 'facilitates' a writer’s use of text sources then, depends heavily on the way any individual, with their own history of practice, performs that task.

Having acknowledged that the influence of the computer screen is contingent upon each individual's shaping of activity, we now wish to put in place the notion that the screen, through its material properties, also itself shapes activity. A number of participants in our study used the paper sources quite differently from those presented on screen. They tended to take notes from on-screen documents, and to refer to these notes, not to the texts themselves, when composing. Yet they tended to take no notes from documents available on paper. Rather they would highlight or annotate them, and return to them throughout writing. Participants typically cited aspects of the materiality of each medium - such as those mentioned above - as reasons for adopting different practices with each medium. Material differences between the two media then, helped shape very different forms of practice and, therefore, different experiences of the texts.

Technology then, can best be understood not as a static influence on literacy practice, but as a dynamic contributor to it. What matters about a technology – the affordances and constraints associated with it – are not properties of the technology per se, but emerge only from its relation with the person (and with the rest of the setting) in activity. Furthermore, these affordances and constraints do not simply influence how smoothly or problematically pre-given literacy practices will proceed: they actually shape the practices themselves. Individuals and computers are involved in complex transactions that shape literacy activities.

Writing has an inevitable temporal structure. There is a coarsely-grained rhythm that describes when we initiate and terminate separate episodes of engagement with the task. Then there is also a more fine-grained tempo to writing. This involves a pattern of shifting attention as the writer moves in and out of engagement within a single episode or session. The point of the present section is to highlight how the technolgoical context of the activity serves to choreograph this pattern of involvement. We are particularly concerned here with the more fine-grained level of engagement.

Comments above about manipulating screen-based documents remind us that there is often an issue for writers of coordinating between the writing task itself and a number of supporting resources – such as reference works and other texts. In the previous section we stressed re-mediation of the physicality in this, as it occurs for writing at computers. In the present section, we wish to dwell more on the interactive dimension of the medium, rather than its spatial qualities. The key issue concerns how the involvement of ICT reconfigures the stimuli that manage this mobility between resources. Of particular interest is the networked status of the technology in use.

For a writer, the existence of parallel documents ready-to-hand is often important. This is especially true for the case at issue here; namely, the well-researched writing of undergraduates. Typically, students will position themselves to optimise such engagements with support material. By focussing on the networked nature of ICT, we are raising issues to do with computers creating richer possibilities for movement between documents. This is not a point about the sheer quantity of support documents that might be rendered to hand by easy network accessing. First, it is a point about how ICT reconfigures the whole issue of manageing document access. Second, it is a point about how ICT necessitates changes in the affordances for interacting with the immediate writing environment. This entails changes in the underlying rhythm of writing – understood as a set of attentional commitments. We will comment on each of these network features in turn.

To say that ICT entails a redistibution of document access is to raise the issue of how writing is spread across geographic locations. Our diary records from undergraduates indicate that a writing project can be exercised in a wide range of places. Most obviously, in libraries and university resource rooms; but also in common rooms, friends study bedrooms and, of course, the writer’s own room. To a significant extent, the motive for distributing activity over locations reflects the need to gain access to physical documents that are themselves all over the place. This applies to books and journals but also to material that might need to be borrowed from staff or fellow students. However, on the particular campus we have studied, private study bedrooms now enjoy intranet and internet access. This means that some documents that once required trips to teaching spaces (libraries etc.) may now be obtained online in the student’s own room. Documents borrowed from fellow students might be routed about the campus via email. In short, such networking serves to situate the act of writing more firmly in a single place – the site of one’s networked PC. When we compared the diaries of students with network access in their own rooms to those of students without network access we found a number of differences (Crook and Light, in press). We did not find any difference in the absolute amount of private study. However the students with network access were more likely to conduct that study in their private rooms. We assume this applies to their writing assignments – as these represented a large proportion of what private study involved. Campus networking, then, does not precipitate more private study but it reconfigures how that study is done. In particular, it concentrates it more in a single place. Thus the experience of writing becomes more situated in this sense.

This example serves our general purpose. The empirical comparison inspired by campus networking allows us to notice a significant and general feature of student writing as a form of cultural practice. Namely, through an inevitable distribution across sites of social and study activity, writing involves other people and, thereby, a potentially wide variety of opportunities for exploration and interaction. ICT (in the form of netwoked study bedrooms) reconfigures the dynamic of writing practices in that sense. It does so by virtue of its disturbance to the socially-distributed nature of the activity.

The second point we wish to make about networked ICT and writing arises from the strong interactive properties of the networked computer. In this case, we appeal to system logs that have been gathered from a broad sample of students with PCs in their own rooms (Crook and Barrowcliff, in press). The most striking message of these records is that students use this technology a lot. From around mid-day until well into the late evening there is a 50% chance that a study bedroom computer will be active. This hints that there are a great many things that can be done with this technology. Indeed, that notion of versatility gets nearer to the main point we wish to make about the finer detail of usage patterns. One distinctive feature of a networked computer is that it makes a large number of resources available at one site for action. Sitting at this technology, the user is able to send electronic mail, have synchronous text conversations, read a news ticker, listen to an MP3 file, watch the television, surf the web – as well as interact with a word processor. All these things may be done in parallel. Perhaps the common image of the PC as a recreational technology is the image of games playing. Yet these students spent rather little time engaged with conventional computer games. Rather, what they did was more about multi-tasking: moving in and out of a wide range of separate applications in a style of working that is best described as "animated".

One analysis we conducted concerned all sessions where a word processor was opened for at least an hour and where the document title implied a course-related writing project. System logs allowed the pattern of computer activity across that hour to be followed. On average, during such a word processing session a student would attend to another task (shift input focus) once every four minutes. How this more animated style of writing should be judged is not central to the present discussion. Certainly, some of these movements between computer applications involved movements between resources that were central to the composition task – text files, websites and so forth. However, it was also clear that many such movements serviced more recreational interests – changing background music, responding to instant messages and so on. None of which is to suggest that private spaces have not always been rich in such affordances for fragmenting some core study activity – such as using a word processor for composition. However, ICT creates an additional layer of such alternative possibilities. Writing tasks executed at this networked technology are clearly reconfigured: this is a technology that strikingly concentrates at one site a wide range of highly interactive affordances.

Yet a final point regarding this theme is important to make, as it echoes something we have already noted about the inherent variation in activities implicating such rich technologies. In interviews it became apparent that students managed the potential of ICT for multi-tasking in different ways. Half of them reported genunine concern that they spent too much time in playful use of their computers. Perhaps as a result of such worries, many reported strategies for filtering out certain sorts of competing computer activities that might be accessible during a period of planned writing. Others, however, seem relaxed about this feature of the technology and were more vigorous in responding to interactive options when they arose. This is the point we have stressed before: individual writers shape the inclusion of new technolgies in distinctive ways. Accordingly, simple generalisations about singular "effects" are again seen to be inappropriate.

The present section concerns how writers’ involvement with ICT can impart a more fluid quality to text. Usually this point would preface a discussion of word processors and the manner in which they allow writers to cut, shape, paste or otherwise manipulate text. Here, however, we consider how text becoming fluid involves manipulations that have a more social focus. At an early point in the development of computers for education, researchers noticed how this technology could potentially "socialise the writing process" (Daiute, 1983). One way in which this may be achieved is by recruiting text into practices of interpersonal communication. Evidently, such practices have been cultivated by the popularity of email, instant messaging, chat rooms, and asynchrnous discussion forums. Text composed in these ICT contexts come to acquire distinctive registers (Ferrara, Brunner and Whittemore, 1991). However, there are other ways in which computers mediate through entering situations in which learners (or writers) are interacting. That is, computers are a technology for collaboration. Such joint activity may occur "at" them, "around" them and "through" them (Crook, 1994). Accordingly, student writing may become one such species of activity that can be incorporated into all of those collaborative arrangements

It has been observed that undergraduates are reluctant collaborators in relation to the familiar task of producing essays (Hounsell, 1987). Our interviews and diaries suggest that it remains relatively unusual for students to work together "at" computers in the interests of shared writing. Yet we might expect this to happen more as a result of universities promoting ICT use so urgently. What our interviews do reveal is that student collaboration over study does occur, but it has a somewhat improvised or serendipitous quality. This includes unannounced visits to the rooms of friends, some of which develop into work-related exchanges. On such occasions, the desktop computer offers a particularly visible surface for supporting joint composition. Similarly it might at the very least precipitate a critical discussion of someone’s writing-in-progress. In this sense, the potential for ICT to enter such casual exchanges among students may implicate the technology in supporting more social forms of writing among students.

However, there is a further sense in which writing is socialised by ICT. This is captured in the idea that computers provide a technology for student peers to interact "through". Text can become more fluid by the ease with which it can be passed among computer users populating a common network. Our system logs of networked computers in student study bedrooms revealed that local file transfer via email and instant messaging was very common. It must be said that much of the transfered material was not at all related to the curriculum. However, again the computer has emerged as an intriguing technology through the way in which it resources both playful and academic concerns at a single site. Our expectation is that practices of electronic communication that evolve to serve playful interests will be gradually appropriated into the demands of study. This leads us to expect that undergraduate text will move more freely among peers thanks to the transporting infrastructure supplied by ICT networking. It leads us to expect that literacy will become more social in this sense.

What students told us about their shared use of lecture notes reinforced the idea that trends of this sort could be active. Most students reported exchanging notes from lectures. Most students with computers in their private rooms reported doing this via electronic mail. It must be admitted that coursework writing seemed more protected in this sense. Coursework, unlike lectures, was more likely to be a topic of conversation – something the student sought benchmarking reassurances about in relation to personal progress. However, again we may predict that this more familiar form of writing will also become loosened from its tie to private studying and become a more common currency in the electronic exchange of documents.

The previous section converged on observations that were somewhat speculative: based only upon emerging trends in ICT use.. The present section needs to be yet more tentative, by virtue of the early state of development of the relevant ICT structures. Again, the theme is an increasingly social dimension to the activities of writing. In the previous section we dwelt upon the potential of ICT for making text more mobile among students. Arising from this was the expectation that writing would thereby become more socially distributed. Certainly, it allowed illustration of the sense in which literacy practice are "social" and the manner in which new technology is involved with this. The concern of the present section is with the role of audience in a broader sense. How can ICT be implicated in student literacy through shaping it to become a more public form of intellectual practice?

We wish to invoke the emerging phenomenon of the academic course website in order to explore these notions of audience. University teaching managers increasingly expect academics to make use of web technology to resource students taking their courses. In short, they expect to see course websites. Analysis of what this precipitates in one representative institution (Crook, in press) suggests that academics are not yet very vigorous or imaginative in how they make use of this new infrastructure. Indeed most of them simply do not use it at all. Those that do present fairly pedestrian material dominated by collections of lecture notes. As it happens, this corresponds very closely with what students report they need. Yet, how might it be otherwise?

Elsewhere, one of us has discussed this more fully in relation to primary education (Crook, 1998). There it was suggested that ICT offered a powerful technology for making visible learner activity within the community of the classroom. The notion of a course website (or a classroom website) acknowledges the idea that this technology concerns something potentially local in character. Indeed what is found at such sites may well be "local" in the sense of relating to needs and interests within that learning community. But typically, the origin of such material is the organiser of the course, or the teacher of the class. Here we are noting that the local website also offers a versatile tool for making visible material that comes from the learners themselves. Not only is this a powerful resource for the classes traditional challenge to create "common knowledge" (Edwards and Mercer, 1987), it can create an authentic sense of audience among novice writers, and it can resource future generations of learners in a class by virtue of "leaving tracks".

It is the unfortunate fate of much student writing that the products are rarely shared with other people. Often such writing may only be seen by the person who set the task. The sense of genuinely writing for others is important to cultivate; the practices of writing that arise where that motive is in place are likely to be distinctive. Moreover, benefit acrues to the "others" also - through the possibility of vicariously learning from access to the work of peers (Mayes, J.T. (1995). Outside of educational contexts, the web has become a lively forum in which opporutnities for creativity in the written word is richly celebrated. Such potential has hardly been explored within universities for the case of student writing. Yet it is clear that ICT in the form of web-based publishing offers considerable prospects for this further sense of socialised literacy.

The example discussed above should convey again the broad scope of factors we wish to embrace as describing literacy practice for the undergraduate writer. It reminds us of the wide variety of ways in which the context for producing a text relates to a technical setting. Resources found on a course website are relevant to shaping the form of an undergraduate composition – most controversially, but not inevitably, in terms of offering the student cut-and-paste solutions. However, the course website also has a potential to shape writing activity through its capacity to create audience. This was more the emphasis of the last section, although it identifies a form of influence yet to be well developed. Again, developed or not, the example allows us to see the way in which literacy practices extend deeply into the socio-cultural context in which the composition is set.

In this section, we use a further example that stresses "audience" in order to highlight another dimension of literacy practice liable to re-mediation through new technologies. Most undergraduate writing will have an instrumental quality. It will be constructed to do something:to effect some influence on the reader. Most often that reader might be narrowly defined to be the person who set the task or, more generally, the person who is going to grade the work. In short, much student writing gets positioned in a loop of feedback. It forms the basis of a possible dialogue between student and tutor. This feedback contract overarhces the production of any particular piece of writing. It characterises part of the general cultural setting in which the undergraduate writes. As such it must be a pervasive influence on the writing enterprise.

We believe that the dialogue cultivated by a piece of written work is an interesting – but neglected – dimension of a student’s developing literacy practice. We also suggest here that the form of that dialogue is significantly coloured by the nature of specialised technical and social practices that are used to manage it. Traditionally, such tutorial feedback on writing was a strongly interpersonal event. At its most elaborate, it might involve a student in some detailed one-to-one discussion with a tutor, ffocussed on some composition that the student had submitted. More recently, such feedback may have been distilled into something a little more terse: corrections noted on the text, comments made in margins, evaluative remarks appended to the finished product. While more economical in form, these practices arguably still maintain a strong flavour of an interpersonal exchange prompted by the student composition.

Such scribbles, notes and evaluative remarks can be viewed as the "social practice" underpinning this particular form of tutorial dialogue. Accordingly, we can imagine instruments – alternative forms of mediation – that could be inserted into this process, and which might alter it in significant ways. The example presented in Figure 1 is a case in point. It is a form of technology for managing the feedback obligation: namely a coursework coversheet through which tutor comments can be organised and attached to the work in question. Interestingly, its design critically depends upon the availability of ICT. For printing the assignment-specific details to cover more than 100 courses requires the computational help of a teaching programme database. What is made possible by that technology is a single sheet of paper that summarises the teaching context for each assignment; including in its design space for a tutor to mark and comment on the work itself. The example in Figure 1 is based on a real case in active use: altough details identifying the department have been edited.

--- Insert Figure 1 about here –-

Not that this protection of anonymity should imply that there is something improper about what is illustrated. As it happens, the department concerned achieved the maximum score in a thorough teaching quality audit. Moreover, the instrument was originally conceived to be only in the best interests of good teaching and learning practice. It is not for us to judge whether it has been successful in that sense: not least, because we have no empirical evidence regarding its use. However, the example remains a good basis for general consideration of this aspect of students’ literacy practices. What the instrument achieves is a method for proceduralising the act of feedback. Probably the stimulus for doing so was a well-intentioned one based upon the ambition to ensure that a minimal level of tutor commentary did occur. Yet there is a danger that such proceduralisation renders the feedback less personal, perhaps serving to create a sense that a social exchange has been reified (Wenger, 1998).

The issue worthy of further understanding is how the impact of feedback is altered by practices that seem to mechanise the communication involved. Even the more modest tradition of marginal scribbles and summary comments at the end of a composition can still convey a sense of managing an authentic dialogue with the reader. The example reminds us that that sense of student writing being a literacy practice includes an embedding of text in an institutional conversation. ICT can act to re-mediate the form of that exchange

Perhaps understanding the significance of ICT in psychological domains can be impeded by the bounded nature of the technology itself. The technology of interest seems conveniently self-contained as a material tool. Accordingly, design features of that tool may be explored as independent variables in experiments, for it is is easy to tinker with the functional properties of the underlying software. Similarly, it may be easy to tinker with the way in which this object is inserted into the teaching contexts that realise some curriculum. Yet it only works to approach ICT as a cluster of independent variables if we are also comfortable that there is some complementary dependent variable: something that can be straightfowardly measured as "writing" – the output in an experiment. It is tempting to regard writing in a smilarly singular manner. The core activity of manipulating a pen or a keyboard seems to give it this focus. Perhaps no writing researcher frames their interest in such a narrow behavioural way. Yet the sense in which researchers approach writing as a cultural practice" often takes us no further than a nother set of behaviors: those associated with the construction of plans, drafts and annotations. In the present chapter, we have chosen examples that help make visible a much wider sense in which writing must be studied as cultural practice.

Hopefully, it is evident from these examples that ICT is increasinlgy deeply implicated in the undergraduate’s experience of writing. So rapid is the spread of this technology within educational contests that it beomces urgent to understand the significance of its role for student writers. However, that understanding can not be modelled on claims about how isolated technical variables have singular effects on isolated dependent variables. The influence in this relationship is very much one of reconfiguring: the availability of ICT is ensuring that writing gets done differently. Our ability to notice just how it is re-mediaged requires us to use methods that respect the systemic nature of this ICT/writing conjunction.

Bruce, B.C. (1997). Literacy technologies: What stance should we take? Journal of Literacy Research. 29, 289-309

Cole, M. (1996) Cultural Psychology. Cambridge: Cambridge University Press.

Cole, M. & Griffen, P. (1980) Cultural amplifiers reconsidered. In Olson, D.R. (Ed.), The social foundations of language and thought. New York: Norton.

Crook, C.K. (1994) Computers and the collaborative experience of learning. London: Routledge

Crook , C.K. (1998) Computers in the community of classrooms. In K. Littleton and P. Light (Eds.) Learning with computers: Analysing productive interaction (102-117). London: Routledge.

Crook, C.K. (in press) The campus experience of networked learning. In C. Steeples and C. Jones (Eds.) Networked learning. Berlin: Springer-Verlag.

Crook, C.K. and Barrowcliff, D. (in press) Ubiquitous computing on campus: patterns of engagement by university students. International Journal of Human Computer Interaction.

Crook, C.K. and Light, P.H. (in press) Virtual society and the cultural practice of learning. In S. Woolgar (Ed.) Virtual Society? Oxford: Oxforrd University Press.

Daiute, C. (1985) Issues in using computers to socialize the writing process. Educational computing and technology Journal, 33, 41-50.

Dillon, A. (1992) Reading from paper versus screens - a critical-review of the empirical literature Ergonomics 35, 1297-1326

Dymott, R., and Crook C.K. (2001 [a]) Technology mediated practices: ethnographic analysis of student essay production. Paper presented at 8^th Annual conference of Writing Development in Higher Education: Changing contexts for Teaching and Learning. Leicester, April 24-25

Dymott, R., and Crook C.K. (2001 [b]) Student essay production: more than a problem solving task. Paper presented at International Conference on Communication, Problem Solving and Learning, Glasgow, Scotland. June 25-29

Edwards, D. and Mercer, N. (1987) Common knowledge. London: Methuen.

Ferrara, K., Brunner, H. and Whittemore, G. (1991) Interactive written discourse as an emergent register. Written Communication, 8, 8-34.

Hansen W.J., & Haas C. (1988) Reading and Writing with Computers: a Framework for Explaining Differences in Performance. Communications of the ACM 31, 1080-1089

Hayes, J. & Flower, L. (1980). Identifying the organization of writing processes. In In Gregg and R. Steinberg (Eds.) Cognitive Processes in Writing. Hillsdale, NJ: LEA

Hounsell, D. (1987) Essay writing and the quality of feedback. In J. Richardson, M. Eysenck and D. Piper (Eds.), Student learning: Research in education and cognitive psychology. Milton Keynes: SRHE & Open University Press

Hutchins, E. L., Hollan, J. D., & Norman, D. A. (1986). Direct Manipulation Interfaces. In D. Norman and S. Draper (Eds.) User-Centered System Design. Hillsdale, N.J.: Lawrence Erlbaum. 97-124.

Kellogg R.T., & Mueller S. (1993) Performance amplification and process restructuring in computer based writing. International Journal of Man-Machine Studies 39, 33-49

Mayes, J.T. (1995) Learning Technology and Groundhog Day. In W.Strang, V.B.Simpson & D.Slater (Eds.) Hypermedia at Work: Practice and Theory in Higher Education University of Kent Press: Canterbury

O'Hara, K. & Sellen, A. (1997) A Comparison of Reading Paper and On-Line Documents. Proceedings of CHI `97, March 22-27.

Erickson, B. J. (1992). A synthesis of studies on computer-supported composition, revision, and quality. Journal of Research on Computing in Education, 25, 172-186.

Ransdell S.E., & Levy C.M. (1994) Writing as Process and Product - The Impact of Tool, Genre, Audience Knowledge, and Writer Expertise Computers in Human Behavior 10 511-527

Snyder, I. (1993). Writing with word processors: A research overview. Review of Educational Research, 35, 49-65.

Torrance, M. (1996). Is writing expertise like other kinds of expertise? In G. Rijlaarsdam, H. van den Bergh, & M. Couzin (Eds.), Theories, models and methodology in writing research . Amsterdam: Amsterdam University Press. 3-9

Wenger, E. (1998) Communities of practice. Cambridge: Cambridge University Press.

Figure 1: An anonymised coversheet generated by a computer database to service the commentary made by tutors on student writing.