This paper proposes an interpretive technique for understanding IS development which uses complementary features of actor-network (ANT) theory and Foucault's theory of discourse. The technique is being applied to a loosely structured project which is working towards the construction of a flexible teaching and learning environment based on the internet. The paper traces the fortunes of two of the non-human actors in this setting. The combination of ANT/discourse theory is found to be an effective descriptive tool for IS development but it is not clear how it could contribute to a methodology. Some possible directions are suggested and balanced against possible organisational and psychological barriers to using such a methodology.
actor-network theory, discourse, IS development, methodology, flexible learning
The search for a theory on which to base the development of information systems continues. The majority of IS development projects still follow, albeit loosely (Fitzgerald, 1997), methodologies derived from the systems approach to problem solving popularised by the RAND corporation in the 1950s (Optner, 1973). This approach takes us through the steps of problem definition, search for solutions, selection of the best alternative, implementation and evaluation. As with many methodologies what starts as a description of how particular projects were done soon becomes a normative or prescriptive theory which promises future success. This may be a reasonable transformation if the original projects were successful, but Schon (1973, pp 203-212) has pointed out the difficulties of trying to use lessons from the past in current projects. After 40 years of IS development several factors encourage IS researchers to look for different theoretical bases for methodologies.
Firstly, a large percentage of computer based information systems are generally acknowledged to provide less than satisfactory service to end-users and to fall short of their original objectives. While some authors have attributed these failures to developers not following accepted theories, there is always the suspicion that the theories themselves may be at fault (Beath and Orlikowski, 1994). Also, researchers concerned with the "impact of IT" have seen the choice of development method as a political choice, with different methods carrying their assumptions about appropriate alignments of organisational and social power (Probert, 1991). And new information technology may require new development methods. Interactive systems are often developed through prototyping, and the growth of the internet will see cooperative systems developed with data and processing widely distributed and with no one clearly in charge of the system. We need a theory of IS development which gives a better account of what is actually going on both politically and technically.
In this paper I consider an adaptation of actor-network theory (ANT) as a candidate for such a theory. In the next section I give a brief description of ANT, its previous application to IT development, and an extension based on Foucault's theory of discourse. In the following section I show how extended ANT was applied to describe the development process in a loosely structured project on the cooperative use of IT. Finally I discuss whether this descriptive method could be used as a basis for an IS development methodology.
Actor-network theory developed from studies in two related but distinct fields: the social practice of science and the introduction of new technologies. An early paper by Latour and Woolgar (1979) looks at struggles over scientific truth in a laboratory, while one of Callon's early studies (1986) considers fishermen and scallops as some of the stakeholders in a changing fishing industry. These examples already exhibit some of the main features of ANT. The actors (sometimes called actants) may be humans, organisations, cultures, ideas, animals, plants or inanimate objects, and these are treated symmetrically irrespective of their ontology. These actors have interests which are represented (in both the semiotic and political senses) by themselves and other actors. The actor-network is a shifting system of alliances and exchanges among the actors.
Latour (1992) claims several advantages for this approach. It is symmetrical with respect to type of actor: it treats humans and machines equally; it is symmetrical with respect to outcome: failures have the same types of explanation as successes; and it is symmetrical with respect to causality: each actor influences and is influenced by other actors and the network as a whole. The equal treatment of people and machines has been criticised but it may be realistic in terms of power relations and it prevents issues from becoming invisible when their representation is transferred (translated) to an actor of a different type. If, for example, some data collection functions are transferred from police informers to computer programs it is still important to be able to talk about the power relations and motives of the collectors and their allies.
The interests of the actors are represented by scripts, usually imperative statements such as "shut the door", "pay your taxes" or "calculate the gross pay". Akrich and Latour (1992) give a comprehensive set of definitions of script-related processes (such as inscription, conscription). These processes describe (amongst other things) the translation of scripts among actors, often involving a change of medium, for example from conditioned response in a human to lines of code in a computer program. Of particular interest is the idea of description (de-scription), the discovery of the words behind the things or actions. This discovery is only possible in contrived, exotic or crisis situations, such as reengineering, consultancy or system failure; in a time such as IS development when nothing is taken for granted.
Scripts are imperative but don't have intentions; actors do. An actor can develop a "program of action" (Akrich and Latour, 1992) perhaps with the intention of maximising the number of actors following a particular script. Some actors may avoid this by following an anti-program. A program of action can include the creation of new actors suitably inscribed. The inscription is most effective if it becomes irreversible, if the actor is, with respect to that script, a "black box" and the script becomes inaccessible to other actors. My e-mail system may, for instance, be carrying a script "always archive a copy for the authorities" which will be more effective because I am unaware of it (or used to it).
Actor-network theory helps us to understand the course of a project or enterprise. We can asks questions such as "How did it come to turn out this way?" (through the changing alliances of actors), "Who is influencing it?" (who has been doing what scripting?) or "Why are some actors acting this way? " (what scripts are they carrying?). These are not questions with deterministic answers but they allow a rich interpretation of the situation.
Some of the more spectacular applications of ANT have been to the genealogy of now well established scientific theories (Latour, 1987), the meaning of simple technical devices (Latour, 1992) or the acceptance of a new product (Bijker, 1992). More recently ANT has been applied to the development of information systems. Monteiro and Hanseth (1996) claim that ANT allows a finer grained analysis of information systems than some other interpretive approaches which can treat all information systems as essentially similar. Walsham (1997) worries that ANT studies are too local, ignoring the wider social environment, and that ANT is amoral and could encourage a devaluation of humans. Grint et al (1996, p.50), although optimistic about the promise of ANT, point out difficulties in identifying the interests of various actants and in particular an ambiguity as to whether scripts are intrinsic to or imposed upon nonhuman actants. While it is clear that the groom says "close the door" (Latour 1992) the message of the data warehouse is not so obvious.
Generalising Grint's concerns, I would say ANT fails to discuss the meanings of scripts, and gives little guidance on how the nature of particular kinds of actors affects their propensity to subscribe to particular scripts. It may be sensible to treat machines and people equally, but they are not identical. ANT may go too far in the direction of flexibility (Latour, 1996b). One way of dealing with this problem is to consider "external" networks. To avoid trying to understand everything at once we normally consider the actor-network for a particular setting or situation, such as the "discovery" of oxygen (Kuhn 1970) or the development of a computer-based information system. We surround the situation with a boundary and imagine that actors outside don't matter, or can be classified as "constraints". But there are networks extending (and folding back) to infinity. We can make the "constraints" a little more realistic by acknowledging their generative networks but restricting these networks to particular issues, "political networks" or "economic networks" for instance. In my work at present I am concentrating on "conceptual networks". I could perhaps call them "semantic networks" although I suspect that term is seriously overloaded; in fact what I mean is "discourse" in the sense developed by Foucault (1972). A script carried by some actors in the current setting finds its meaning in another network which we call a discourse. In an IS development project some scripts may refer to the discourse of accounting, others to the discourse of electronics. The same words may be used but the meanings may be incommensurable. This is a major difficulty in trying to formulate "system requirements".(Gause and Weinberg, 1989) Combining the insights of ANT and Foucault promises a two-way benefit: discourse theory allows us to see where the scripts are "coming from"; and ANT adds substance to the genealogy of the discourse (Foucault, 1980), giving some structure to individual power/knowledge encounters. Some related work on such encounters can be found in Bloomfield and Vurdubakis (1994).
This "discourse biased" version of ANT was used to investigate a project at an Australian university (the Flexible Delivery Pilot Project). Like many university projects, the Pilot Project was supported by a grant (internal in this case), had a limited life (calendar year 1997) but was actually a bureaucratic "time-slice" in an ongoing process, and had an ambiguous management structure - perhaps not unlike many other IS development projects. The idea of this project was to develop and disseminate expertise in the use of networked IT to allow students of the university greater choice in when, where and how they learnt (Document #8). The project was to develop (among other things) "electronically mediated structured teaching and learning environments (hereafter called electronic workspaces)" (Document #6) by drawing together components and experience from previous separate projects. At least that is my idea of the project. From the point of view of ANT the project may or may not be defined by the various actors in various ways and will have to struggle for its own existence (Latour 1996a). In another article (Underwood, 1998) I have discussed whether the project was successful in this sense. In this article I will use the programs of one or two non-human actors to illustrate the descriptive power of the extended ANT, then discuss whether the theory could be used normatively rather than descriptively.
To apply ANT to this case I studied documents generated before and during the Pilot Project and interviewed most of the key participants. The study was commenced in early 1997 and is still in progress. From documents I extracted (intuitively) a number of key issues and concepts. I used these as a basis for the interviews, but found that each interviewee added their own concerns. All interviewees were, however, comfortable with discussing the case at the "correct" level - they discussed concepts as they related to this case rather than in general, they depersonalised any political issues and they did not introduce low level technical or bureaucratic detail. My position was that of an almost outside observer introduced to the interviewees through a primary informant who was one of the key leaders of the project. "Almost outsider" because I have previously worked at the university in question, knew many of the interviewees personally and was working with my primary informant on a grant application for a related project for 1998. While this casts severe doubts on my objectivity (not a plausible concept in ANT) it gave me the advantage of a wealth of background information.
The actors whose programs I will discuss in this paper are "Lotus Notes" and "team learning". Lotus Notes is data based group support software which is widely used in industry. Although it can be used to implement work flow systems most applications emphasise the database functions, particularly data sharing and synchronisation for remote users. Work flow features are often confined to electronic mail and discussion groups. It is regarded as the leading software of its type and for many years had a monopoly on this market. It is now owned by IBM and its overall program is to maintain its share of the market against threats from Microsoft Exchange and Netscape Communicator. It is not in general use at the university but has been used for several years by information systems students designing a new department for a simulated organisation. Lotus Notes requires a fair amount of technical support.
The organisation simulation (which is essentially a role playing exercise) has evolved from the original Reliable Motor Company by the NCC (1984). The aims were to understand the workings of an organisation and to practice interpersonal skills particularly when working in a team. To keep up with organisational reality IT tools have been introduced into the game, first electonic mail, then Lotus Notes. This allows students to experiment with different uses of IT in organisational communication. The overhead of learning Lotus Notes is justified because organisations in this town are heavy users of Notes, so IS students appreciate having this skill. I have identified the second actor in this paper as "team learning" which represents the behavioural skills that are learned in this exercise, independent of the software used. Team learning also needs to survive, though it is expensive in staff and supported by only some academics. At the start of the Pilot Project team learning was a strong ally of Lotus Notes.
This organisation simulation exercise was one of four strands which were "brought together" in the Pilot Project. The others were an initiative by the IS department to make all their course material available remotely (to their existing students) as well as face-to-face, improvement of a subject on the use of internet resources (for non-IT students, using Netscape Navigator) and the development of a new course which would have students attending the university intensively for 4 weeks then completing project work from their (remote) home locations. The latter two courses were offered by a second department, so the Pilot Project brought together four projects from two departments reporting to a central university education committee. None of these other strands seemed natural allies of Lotus Notes or team learning.
While Notes and team learning had been allies for a long time they have very different "home" discourses. Lotus Notes is linked to the discourse of computing: "Lotus Notes is a client-server platform for developing and deploying groupware applications. Lotus Notes allows people to access, track, share and organize information in ways never before possible, even if they are only occasionally connested to a network."; and "Collaborate - Notes makes it easier for individuals to work together with information. Documents can be passed along from one person to the next, with each adding to or refining information in the document." (Lotus, 1995) Here the work of the team means "working together with information". The first step in building a team is to set up a database.
Team learning, at least at this university, seems to relate more to a discourse of business processes. "We have designed electronic workspaces to support the development and performance of flexible work groups in Information Systems units. We have designed electronic spaces for informal discussions (an electronic café), group product development, work product sharing (an electronic repository), private conversations, communication with staff, electronic submission of assessable work (and return) and electronic access to lecture, reference and other unit materials." (Collings et al, 1997). Here the emphasis is on types of team activity. Although some of these activities are people oriented rather than task oriented they still tend to be related to information and structure - it is as if the "people" or team building activities were seen as just another task. The electronic spaces themselves start as unstructured because part of the "design" task of the students is to structure these spaces. Team building has often been related to another discourse, the psychoanalytical. Freud emphasised the development of relationships between team members and the leader, while Bion traced the difficult paths a team may follow before (hopefully) settling into a cooperative work mode (see deBoard, 1978). This discourse is not seriously discussed in any of the documents relating to the team learning project, or the subsequent Pilot Project. Another discourse which is not fully linked to this particular discussion of team learning is the educational. While many references are made to individuals gaining skills in teams (including team building skills), there are no specific references made to learning from each other.
Some of the academics involved in the team learning project are quite knowledgeable in psychology and education theory so they are not unaware of these issues. Before electronic communication was introduced to the simulated organisation another information technology was used, but only in the "real" world, not within the simulated organisation. This was video recording. Some student activities (e.g. interviewing) were recorded and played back at debriefings. These debriefings had a considerable psychological slant. It is interesting to speculate whether the alliance with Lotus Notes crowded out these psychological scripts. In any case it is difficult now for staff to follow the group formation activities, since they do not access the students' café discussions.
The issue of scripts having different interpretations in different discourses is apparent with a major script in the project: "develop electronically mediated structured teaching and learning environments". In the discourse of computing this could easily be synonymous with "build electronic workspaces"; in psychoanalysis we would talk about trust, authority and archetypes (is the workspace the home of the sage or the joker?). In educational discourse we might concentrate on how the student (or team?) constructs their own knowledge in this environment.
Towards the end of 1996 there were two events which affected the dynamics of the actor network. The Pilot Project was approved and Lotus Domino became readily available. One of the main aims of the Pilot Project was to promote a campus-wide culture of flexible delivery. The component projects should be somewhat compatible and, as we have seen, only one used Lotus Notes or team learning. The others used mainly Netscape. The program "let them use Lotus" had relied on the educational benefits for IS students and making available (through licensing agreements) free copies of Notes for the students' home computers. Students in the other courses involved in the Pilot Project would not be learning Notes because it was too difficult. The release of Domino may have saved things for Notes. Domino connects Notes data bases, mail and and discussion groups to other widely used software. The "clients" (students) only need a web browser and a mail reader (any brand). A certain amount of database management can also be achieved via a browser. This development probably advantaged Notes (and its ally team learning) in the short term, but other actors became more noticeable. The Domino model turned the Notes problem around. Rather than there being a server looking for cheap and easy client software, there was now a standard and independent client interface able to link to any suitable server. This meant there were now a number of competing products such as Caucus and Top Class.
These competing server products need to form an alliance with another actor, the university computer centre. The Pilot Project is meant to consider campus wide issues, and since the IS department has neither the resources nor desire to provide servers for the whole university, and since many other departments will not wish to manage such a service, the computer centre becomes a key (if currently quiet) player. A computer centre representative said that, while they have not been funded to provide flexible delivery at present, they are investigating server software so the "necessary infrastructure" will be available. It seems likely that the computer centre will take a "middleware role" linking the students to materials on appropriate library or departmental servers.
This means that team learning (separately from Lotus Notes) also needs to make an alliance with the computer centre, or at least inscribe "students can belong to several teams" into future middleware. The program to achieve this so far has seen the inscription of "project" and "cohort" into a diagram of "Flexible Delivery Workspaces" in the Pilot Project final report (Document #10, p31), and the acceptance of this report as providing university wide guidelines. The report also states that "This means that each group must have a work area on a server that they can use in their own way, and must be able to set up their own discussions, bulletin boards, etc., as and when they need to." (p29) This describes what has been done with Notes but there is no reference to specific software here. Team learning is declaring its willingness to change partners, or at least making contingency plans.
Tracing through the story, I think we can see that the alliance between team learning and Notes, although quite fruitful for a time, was always at risk (all alliances are) because there was no common "home discourse" and neither party was able to irreversibly conscript the other.
The example taken from the Pilot Project shows how ANT can trace the development of (some aspects of) an IS project. It also suggests that scripts are not self explanatory and that different actors may find the meaning of what is apparently the same script in different discourses. Can these techniques be used to "help" project development.
For some time (Underwood, 1992) I have been advocating that we should try to preserve ambiguity and contradiction throughout the development process (and in the "completed" system). To engage in (say) data cleansing, which is the latest fashion in reverse engineered consistency, is to remove data which may be important to many stakeholders and to the survival of the system. An ANT/discourse based methodology might be able to preserve a realistic level of ambiguity.
The first possibility is to concentrate on the construction of a local language. Various human and organisational actors might come to use mutually understood language. Marketing, accountants and programmers might become comfortable about what "efficient" means for a particular project. In the Pilot Project case the majority of the university may accept that the Pilot Project report gives a fair enough definition of "flexible learning" for the time being. The local language can probably preserve ambiguity because it is dynamic and underdeveloped and because the actors still maintain an underlying alliegance to their home discourses. The development of the local language can be encouraged or tested by consensus or "unreality" checks: when describing do actors refer to the project network or to their home discourse? Whether such a local language will be sufficiently fluid or whether it degenerates into "groupthink" requires further investigation. In any case this approach may only preserve a subconscious or secret ambiguity under a veneer of consistency.
The second possibility is to follow ANT and keep describing. We never allow pieces of our design to become black boxes, we never accept inscriptions as irreversible. Frequent description sessions will reveal which scripts are progressing happily through the system design and which have been de-inscribed. Perhaps we need something like a script audit, where each actor (or program) registers their key scripts at the beginning. These are frequently checked for progress making the ambiguity and politics publicly visible. An actor may acknowledge a script as having lost importance (use Lotus Notes) or may object to a script which is in danger of becoming irreversible (no non-student access to web pages). From my interviews I don't believe the human actors would have any trouble identifying relevant scripts - they are all quite aware of the "politics", it just isn't part of the "methodology". What is really being advocated here is openness (cf Argyris, 1984); but ANT is a French invention, and French writers have not always been sanguine about opennness in organisations (Crozier, 1980). Whether any network can operate with complete openness is an undecided question.
ANT does have a normative face, but only for a single actor or program. Grint & Woolgar (1997, p28) summarise how allies are enrolled.
First the 'problematization' stage identifies key actors who are then persuaded that the solution to their own problems lies with the enrollers. Second, intéressement involves the gradual dissolution of existing networks and their replacement by a new network created by the enrollers. Third, the stage of 'enrolment' proper occurs, in which, through coercion, seduction, or consent, the new network achieves a solid identity. Finally, the alliance is 'mobilised' to represent an even larger network of absent entities.
This is how each program proceeds (with luck). But there are also anti-programs operating at the same time. What is the position of the analyst? They are just another actor with another program. ANT does not countenance that one actor might follow a program for "the common good" or that one program can dominate from the start. As with evolution, we can only tell who was "fittest" after the event. So this normative method (which was probably known long before Machiavelli) only operates for individual goals. This is easily conceptualised if the analyst is trying to sell something or is the clear representative of another powerful actor, but what if they are representing objectivity, ambiguity, job satisfaction or progress? Are they prepared to have their scripts open for description? In the end, as Hedberg (1980) says, the analysts must disappear and each actor must be their own analyst. Perhaps we still have roles as trainers or facilitators although these roles seem to be going out of favour at present.
The study of the Flexible Delivery Pilot Project showed that a combination of actor-network theory and discourse theory provided a powerful interpretive tool. This experience suggests two possible ways to proceed towards a methodology, one based on encouraging a fluid local language, the other on frequently de-scribing scripts from partially built IS artifacts (hardware, software, procedures, people). Unfortunately, for either of these approaches to work, we need potential project managers and project owners to embrace the view that every IS development has a life of its own and that no single actor or alliance of actors can be assured of the outcome. And if we could get them to accept that, a methodology just becomes a useful communication tool rather than a fetish (Wastell, 1996).
Akrich, Madeleine and Latour, Bruno (1992) "A Summary of a Convenient Vocabulary for the Semiotics of Human and Nonhuman Assemblies" in Bijker, Wiebe E, and Law, John (Eds) Shaping Technology / Building Society MIT Press, Cambridge, Mass.
Argyris, Chris (1984) "The Executive Mind and Double-Loop Learning" in Bemelmans, Th.M.A. (ed) Beyond Productivity North-Holland, Amsterdam.
Beath, Cynthia Mathis and Orlikowski, Wanda J. (1994) "The Contradictory Structure of Systems Development Methodologies" Information Systems Research 5,4, pp 350-377.
Bijker, Wiebe E. (1992) "The Social Construction of Flourescent Lighting, or How an Artifact Was Invented in Its Diffusion Stage" in Bijker, Wiebe E.; and Law, John (Eds) Shaping Technology / Building Society MIT Press, Cambridge,Mass.
Bloomfield, Brian P. and Vurdubakis, Theo (1994) "Boundary Disputes: Negotiating the Boundary between the Technical and the Social in the Development of IT Systems" Information Technology & People 7,1, pp 9-24.
Callon, Michel (1986) "Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of Saint Brieuc Bay". In Law, John (Ed) Power, Action and Belief: a new Sociology of Knowledge? Sociological Review Monograph (pp. 196-233). Routledge and Kegan Paul, London.
Collings, Penny; Kleeman, Dale; Richards-Smith, Avon and Walker, David (1997) "Developing New Group Work Practices: An Evaluation of the Design and Use of Groupware-based Systems for a Graduate Course in Information Systems" in PACIS '97 Queensland University of Technology, Brisbane.
Crozier, Michel and Friedberg, Erhard (1980) Actors & Systems University of Chicago Press, Chicago.
deBoard, Robert (1978) The Psychoanalysis of Organizations Tavistock, London.
Fitzgerald, B. (1997) "The Use of Systems Development Methodologies in Practice: a field study" Information Systems Journal 7,3, 201-212.
Foucault, Michel (1972) The Archaeology of Knowledge Tavistock, London
Foucault, Michel (1980) Power/Knowledge Pantheon Books, New York.
Gause, Donald C. and Weinberg, Gerald M. (1989) Exploring requirements : quality before design Dorset House, New York.
Grint, Keith and Woolgar, Steve (1997) The Machine at Work Polity Press, Cambridge.
Grint, Keith; Case, Peter; and Willcocks, Leslie (1996) "Business Process Reengineering Reappraised: The Politics and Technology of Forgetting" in Orlikowski, W.J.; Walsham, G; Jones, M.R. and DeGross, J.I. (Eds) Information Technology and Changes in Organizational Work Chapman & Hall, London.
Hedberg, Bo (1980) "Using Computerized Information Systems to Design Better Organizations and Jobs" in Bjørn-Andersen, Niels (ed) The Human Side of Information Processing North-Holland, Amsterdam.
Kuhn, Thomas S. (1970) The Structure of Scientific Revolutions (2nd edn) University of Chicago Press, Chicago.
Latour, Bruno (1987) Science in Action Harvard University Press, Cambridge, Mass.
Latour, Bruno (1992) "Where Are the Missing Masses? The Sociology of a Few Mundane Artifacts" in Bijker, Wiebe E. & John Law (eds) Shaping Technology / Building Society MIT Press, Cambridge, Mass
Latour, Bruno (1996a) ARAMIS or The Love of Technology Harvard University Press, Cambridge, Mass.
Latour, Bruno (1996b) "Social Theory and the Study of Computerized Work Sites" in Orlikowski, Wanda J., Walsham, Geoff, Jones, Matthew R. and DeGross, Janice I. (Eds) Information Technology and Changes in Organizational Work Chapman & Hall, London.
Latour, Bruno; and Woolgar, Steve (1979) Laboratory Life: the Social Construction of Scientific Facts Sage, Beverly Hills and London.
Lotus (1995) Lotus Notes 3.3 - A Quick Tour of Lotus Notes Lotus Development Corporation, Cambridge, Mass.
Monteiro, Eric and Hanseth, Ole (1996) "Social shaping of Information Infrastructure: On Being Specific about the Technology" in Orlikowski, Wanda J., Walsham, Geoff, Jones, Matthew R. and DeGross, Janice I. (Eds) Information Technology and Changes in Organizational Work Chapman & Hall, London.
NCC (1984) Systems Training - Presenter's Case Study National Computing Centre, Manchester.
Optner,Stanford L. (1973) Systems Analysis Penguin, Harmondsworth.
Probert, Stephen (1991) A Critical Study of the National Computing Centre's Systems Analysis and Design Methodology, and Soft Systems Methodology unpublished MSc thesis Newcastle Business School, Newcastle Upon Tyne.
Schon, Donald A. (1973) Beyond the Stable State Pelican, Harmondsworth.
Underwood, Jim (1992) "Information Systems Work: Contradictions in Practice and Theory" in MacGregor, Rob (Ed) Proceedings of Third Australian Conference on Information Systems Department of Business Systems, Wollongong.
Underwood, Jim (1998) " Using Ambiguity in Information Systems Evolution: Keeping the System Soft?" unpublished.
Walsham, G. (1997) "Actor-Network Theory and IS Research: Current Status and Future Prospects" in Lee, A.S; Liebenau, J.; and DeGross, J.I. Information Systems and Qualitative Research Chapman & Hall, London.
Wastell, David G. (1996) "The Fetish of Technique: Methodology as a Social Defence" Information Systems Journal 6,1, pp 25-40.
#6 - "Teaching Grant Application", October 1996.
(application from IS department for Pilot Project)
#8 - "Towards a Policy on Flexible Learning at the University ... ". July 1997.
(Discussion paper for senior staff)
#10 - "Flexible Delivery: A Guide for Teaching Staff", December 1997.
(this is the final report for the project)