Soft’ systems thinking and information systems: a framework for client-led design

Info Systems J (1994) 4, I 17-1 27 m

‘Soft’ systems thinking and information systems: a framework for client-led design F. A. Stowell and D. West Department of Computing and Information Systems, University of Paisley, Paisley PA1 2BE, UK

Abstract. Since the mid-1980s there has been a growing interest in the application of soft systems methodology (SSM) to the information systems design process. This interest has resulted from attempts to overcome the recognized deficiencies of conventional computer systems analysis methods and techniques. A particular problem which has received attention over the past 5 years is the epistemological and operational differences between the investigative process of the pre-design stage and the technological specification. We suggest that this argument is some- what unproductive and advocate a necessary rethinking about the nature of information systems and the use of technology to support their activities. A re-evaluation of the way that we set about designing computer-based information systems suggests that many of the problems of conventional systems analysis methods may be alleviated by an approach that allows the ‘client’, or ‘user’, to have a greater control over the identification, specification and development of their information system(s). The authors’ belief in this courSe of action has led to the development of client-led design as an underpinning philosophy for user participa- tion in the design of computer-based information systems. Client-led design draws upon and develops concepts and tools from ‘interpretive’, or ‘soft’, systems thinking and, in particular, can be seen as providing a framework for the type of subjective inquiry that Checkland & Scholes (1990) referred to as ‘ideal-type’ mode 2 SSM. This paper is related to the papers published in the Journal of Information Systems (Vol. 3, No. 3), which was a special edition to illustrate the influence of ‘soft’ systems thinking upon information systems design and development.

Keywords: client-led design, information systems, soft systems methodology (‘ideal-type’ mode 2).

INTRODUCTION

The rate at which change takes place within IT means that professionals within the field have

a need to reappraise constantly the design methods and techniques that they adopt. In particular, the responsibilities and tasks of the computer systems analyst (CSA) can be seen to have been changing significantly over the last 10-20 years. It is not only advances in researchers’ thinking about the role of the CSA and their ‘clients’ that have made it important for us to reconsider the role of the CSA but the very nature of advanced technology itself, which can be seen as a catalyst to researchers and practitioners alike in encouraging them to rethink the way in which technology is applied and utilized within business organizations. The now popular notion of an ‘information system’ has embedded the use of technology in the workings of most business organizations, although it is clear from both the literature and from practice that the meaning attached to the phrase ‘information system’ is not universally shared. It is from this observation concerning the use and misuse of terms such as ‘information system’ and ‘computer system’ that our argument for both the theoretical foundation and the practical application of client-led design begins.

A definition of terms: ‘computer systems’ and ‘computer systems analyst’ versus ‘information system’ and ‘information systems analyst’

Before commencing upon a more detailed discussion of client-led design it is worthwhile revisiting some popular concepts and terms in order to explain the authors’ thinking and the origins of the development of client-led design.

Thinking about and practising the provision of computer-based information systems within any organizational setting requires that some fundamental assumptions about computing, data processing, information systems and the role of the CSA are made clear and, if necessary, redefined. For the authors, the popular tendency to use the terms ‘computer system’ and ‘information system’ synonymously represents (i) a lack of thought about the nature of computers and their role in information processing, (ii) the poor and limited use of the word ‘system’ to refer purely to a computer and its related components and (iii) a lack of clear understanding about the nature and characteristics of an information system.

For the authors an ‘information system’ can be taken to be the notional whole that manages the provision, manipulation and use of appropriate data to enable decision-making and resulting action in the realm of purposeful human activity. While this definition does not presuppose the role of computers it does take as given the role of the human being in assigning value and use to data and, hence, the transformation of this data into information.

If we take our interest to be in the use of technology to support such ‘information systems’ then we need to be concerned with the wider implications of ‘information’ and its management and use within the organizational setting as well as the effects that IT may have upon the operation of the information system and, consequently, upon the organization as a whole. In this way the computer systems analyst, or perhaps more appropriately from now on, the information systems analyst (ISA), focuses hidher attention upon the decision-making proc- esses involved in some human activity system (a notional system identified to highlight some purposeful human activity) and the variety of potential technologies that may support these activities. In comparison, the more traditional concept of computer systems analysis has been

concerned with the identification of data-processing activities within the organizational setting and their enhancement (usually in terms of increased efficiency) through the application of computers. A clear distinction can be made between the two approaches in that, whereas the latter practice is based upon the assumption that it is the data-processing activities which are key to organizational performance, the former places emphasis upon understanding the decision-making activities which are then supported by the identification and processing of relevant data.

The role of the ‘client’ in information systems design

The identification of an ‘information system’ can be a difficult process not only because of the complexity of the information system involved but also because the information system can be seen as the product of social interaction between organizational members: it is a vehicle for communication that enables and supports the management of the business organization. Consequently, those who are most able to identify and discuss the implications of the information system are those who are, one way or another, involved in its operation. If we accept this point then it follows, quite naturally, that those involved in the information system, the users, or ‘clients’ as we shall refer to them, should by definition play a primary role in the analysis and design of any computer-based support for their information system(s). Although clients may not have a depth of technical knowledge they can be expected to (i) supply the context in which the information system operates and (ii) ‘test’ out proposals based upon their experience of doing the job before the detail of the technology has been considered. It is the recognition of the central role of the ‘client’ in this process which is reflected in the notion of ’client-led design’ as a philosophy for information system design.

The shift in emphasis reported in the previous section concerning the relationship between ‘computer system’ and ‘information system’ has repercussions for the ISA in that different methods, techniques and methodologies will need to be adopted which will facilitate the joint venture in which the ISA and the client(s) identify relevant information systems and then specify the technological devices that may be used to support them.

We can identify numerous participative methods of computer systems analysis which share the objective of getting the client involved in the development, tuning and implementation of a computer system (Mumford & Weir, 1979; Finkelstein, 1989; Avison & Wood-Harper, 1990; McDermid, 1990) but, we argue, these approaches are undermined by the primary desire to fulfil a technological outcome (Stowell, 1991). It is not possible to facilitate true client-led design, we argue, without a fundamental shift in the thinking which underpins the information system design process.

The rest of this paper is directed towards an explanation of the ethos of client-led design and its ‘interpretive’ roots, and a description of the role of the ISA and the part that technology plays in the process of client-led design. Finally, the contribution of ‘systems thinking’ to client-led design is discussed using a particular ‘interpretive’ systems methodology, SSM, to illustrate how client-led design may be put into practice.

CLIENT-LED DESIGN: A FRAMEWORK FOR INFORMATION SYSTEMS DESIGN

Conceptually, client-led design is based upon the interpretive paradigm (Burrell 8, Morgan, 1979), and within that paradigm upon the notions of hermeneutics and phenomenology. Client-led design is a philosophy for collaborative information system design which enables the client(s) to control the development of their technology-based information system. Although the approach is not wedded to any particular methodology, it is important that the underlying paradigm of interpretivism is maintained throughout the design process. The corollary to this is that the methods used as part of the process must themselves satisfy the philosophical underpinning of the interpretive paradigm. An important difference between client-led design and, so called, ‘participative approaches’ is that client-led design is concerned with enabling the client(s) to determine the nature of the IT-based system rather than, through consensus, helping the practitioner to analyse the problem and develop a technical ‘solution’ from expressed views. Instead, in client-led design the practitioner acts as a ‘teacher’ and ‘adviser’ to enable those working in the situation to explore and specify their own requirements and technological needs. In this sense the client(s) have control of the project and may direct the investigation.

On practising client-led design it soon becomes evident that the ISA requires more than purely technical skills to orchestrate the development of a computer-based information system. The whole process of identifying, designing and developing computer-based support for an information system necessitates an intervention into the working practices and customs of an organization‘s culture. The analyst and those associated with the investigation will almost certainly alter the way in which the enterprise operates and the way in which the social networks interact. Their actions are effectively the catalysts of organizational change which go beyond the narrow confines of computer hardware and software.

In differentiating between a ‘computer system’ and an ‘information system’ and in recognizing the role of technology in supporting information system activities we also highlight the need to rethink about the way in which we set about the process of identifying an information system and detailing its possible technological support. It is important that in the early stages of this process the influence of technology should be minimized: we need to avoid a situation in which technology itself is driving the understanding and definition of the problem and the design of a solution to the problem (Stowell, 1991). We need to remind ourselves constantly that techno- logical devices such as computers should only be considered when the information system has been identified and ways of supporting its activities are being considered. In this way emphasis is placed upon the use of technology as a means to an end as opposed to an, end in itself.

Client-led design can best be described in terms of five iterative phases of development (Stowell & West, 1994). Phase I is concerned with what we describe as the appreciation, definition and analysis of the problem situation being investigated. We use the term ‘apprecia- tion’ here in the sense that Vickers (1965) used it to refer to the deep understanding of a situation which is brought about by gaining knowledge of the situation through experience and judgement and the way in which the components of this situation are related. In phase 11, the client and ISA define and represent their information system, including the development of the job specifications of staff involved in its operation. Phase 111 deals with the definition of technical

‘Soft‘ systems thinking, information systems and dient-led design

Figure 1. A diagrammatic representation of the process of client-led design.

support for the information system, while phase IV is concerned with the implementation of the proposals. The final phase, phase V, relates to the maintenance of the information system (Stowell & West, 1992). Figure 1 describes the interaction between the various stages of client-led design and clearly illustrates the iterative nature of the approach, emphasizing the continuous cycle of action and learning which is central to ‘interpretive’ modes of inquiry. Above all, the notion of an interpretive process of inquiry is based upon the belief that we should attempt to understand our world at the level of individual and subjective experience. From this starting point, then, client-led design represents an attempt to facilitate the client in identifying and analysing their own information systems and in selecting appropriate technology to support these information systems. This approach represents a fundamental shift in thinking from the more conventional information system design methods whereby the CSA attempts to ‘soak up’ the details of the situation and from this second-hand knowledge provide specifications for the necessary computer-based information system.

Appreciating, defining and analysing the problem situation

Central to client-led design and its successful application, and a major aspect that differentiates client-led design from other participative approaches, is the process of appreciation, definition and analysis that is undertaken by those involved in the study. For example, in order for the IS practitioner to enable the client(s) to articulate their information system requirements, it is important that their awareness of the problem situation as a whole is raised rather than just their understanding of a particular part of it. Moreover, the ISA needs to appreciate the problem in full if the technical advice given is to be of value to the client(s). It is difficult to try and view a situation from another’s perspective and the situation is made more complex by the need to

FA Stoweti & 0 West

recognize the unique contextual setting within which any information system exists. The ISA, therefore, needs to appreciate organizational complexities as much as technical ones, and this process can best be undertaken by adopting an approach to information system design which supports a rich learning process. It should be remembered that clients are unlikely to be fully appreciative of the entire problem situation and so just as the ISA goes through this process of learning so, too, do the clients.

It can be expected that the opportunity provided for the client(s) to express opinions will, at first, produce a myriad of views, many of which may be at odds with one other. This may seem to complicate the information system design process beyond what is required since what we are ultimately seeking is some formalized, clearly defined way of working. But it is this process of gaining an appreciation, by both client and analyst, that is key to the whole information systems design activity. In many information systems design methodologies the amount of time spent first in defining the problem and second in identifying the information system is small compared with the time spent on the technical specification of the computer-based information system. The increasing use of computer-based application generators coupled with the pressure on ISAs to produce quick results encourages short cuts, often at the expense of the early stages of project appreciation and definition.

in client-led design, then, emphasis is placed upon:

1. the early stages of appreciating the situation under investigation; 2. the complete involvement of those working in the situation; and 3. the role of the practitioner as a ‘guide’ in this process of learning and appreciating.

It is important that the ISA uses skills and methods appropriate to the stage of problem appreciation: the ISAs technical expertise should be used only in the technical specification phase of the study and not at the stage where the accent is upon problem appreciation. (It is worthwhile stressing again here that the ISA needs to adopt an overall view of information systems as one in which a variefy of technologies may fit rather than the traditional view of computing systems in which the computer is considered to be the information system.)

The difficulty facing the ISA at this stage may be seen to be a practical one, in that whilst technology requires a functional explanation, information needs understanding and interpreta- tion within a given context. Despite advances in technology, therefore, the human being is still the essential element in an information system since it is the human being who can assign meaning to the data processed by technological devices. One way of describing what the analyst does at this point is as a process of facilitating the ‘unpacking’ of what appears to be the different views of ‘reality’ involved in the situation and then the ‘repacking’ of these views into a form which is meaningful to those in the situation being investigated. Failure to recognize the importance of individual perceptions of reality and the need to appreciate these perceptions is likely to result in the ISA and client gaining only a shallow picture of the problem situation and hence a less than satisfactory application of the supporting technological solution.

FACILITATING CLIENT-LED DESIGN THROUGH ‘SYSTEMS’ THINKING

In order to understand the full requirements of an information system analysts should not be constrained in their investigations by the analytical approach used. The approach itself should be ‘agnostic’ to the problem situation: that is, its application should not be limited to a particular problem situation or ‘type’ of domain. In practice, this means that the approach used should not have the rigidity of a step-by-step technique but should be flexible enough to meet different situations and to support an iterative mode of investigation. What we require, then, is a framework to guide the ISAs thinking about the problem situation which will enable them to (i) direct the processes involved in appreciating, defining and analysing an information system and (ii) by the end of its application produce a technical specification to support the defined information system. Although it is important that any approach adopted should not impose a rigid structure on the problem situation (the approach used should mould itself to fit the problem form not vice versa), at the same time the framework provided must be robust enough to support the information systems design processes involved. Such a list of demands may be addressed by some of the ideas and methods that have been advanced under the general heading of ‘interpretive’ or ‘soft’ systems thinking.

Operationaliring client-led design through SSM

We have argued elsewhere (Stowell & West, 1992) that the development of technology-based information systems calls upon a variety of methods and techniques as the client progresses from problem appreciation through technical specification to implementation. However, having outlined the essence of client-led design, our aim is to focus discussion upon what Checkland & Scholes (1990) refer to as soft systems methodology ‘ideal-type’ mode 2 and its usefulness as a means of operationalizing the notion of client-led design of technology-based information systems (Stowell, 1991).

One way in which it may be possible to provide the analyst with a means of guidance that will facilitate learning about both the situation under investigation and the method of approach adopted is to undertake the information system provision process as a piece of action research (Stowell, 1990). In recent times a resurgence of interest in action research both as a method of conducting research and as a process of introducing change has taken place, notably with the development of soft systems methodology (Checkland, 1981). The application of ‘soft’ systems thinking coupled with an appropriate means of producing a technical specification might provide one means of satisfying the requirements of technology-based information system design (Wilson, 1984; Checkland, 1986, 1992; Wood-Harper et a/., 1985; Miles, 1985; Stowell, 1985, 1989; West, 1991; Mansell, 1991; Wood-Harper. 1992). A further argument to support the use of SSM as a method of operationalizing client-led design is that it offers a framework for inquiry which facilitates learning, a fundamental aim of client-led design since it is through the process of inquiry represented by SSM that an appreciation of the problem situation is developed. It is this emphasis upon ‘appreciation’ and learning that represents an epistemo- logical difference between information systems analysis and computer systems analysis in

which attention is placed upon defining the situation in such a way as to enable technological support to be identified.

The idea of adopting an ‘interpretive’, or ‘subjective’, mode of inquiry is not new. However, in past attempts, when the analyst reaches the question of technical development, the subjectivism that underpinned the inquiry up to this stage is engulfed by the functionalist characteristics of the more conventional computer systems analysis methods and the demands of technology. For example, where soft systems methodology has been used as the method of inquiry (Wood-Harper eta/. 1985; Benyon & Skidmore, 1987) the translation of the information system as described by the conceptual model is taken over by the technical specialist and used independently to develop the computer-supported ‘information’ system. This approach often becomes an exercise in generality and the benefits gained from the development of the information system through the use of SSM are lost.

In comparison with this previous trend of ‘front-ending’ a subjective inquiry method onto a more conventional computer systems analysis design method, it is argued here that the design of the information system must continue to be developed by the client and the analyst in the same fashion and within the same philosophy as the problem was initially addressed. One way of carrying this out is to treat the whole project, not just the appreciation, definition and analysis stage, as a piece of action research. In this way the appreciation gained about the situation by the ‘team’ may guide the information system definition, representation and development process. The design of the technology-supported information system can then be considered by both analyst and the client(s) in its widest context rather than as a set of discrete steps in which the information system is determined by the technology.

Attempting to use an ‘interpretive’ approach such as SSM for the development of IT applications is not new, but, carrying through the ‘interpretivist‘ concepts to enable clients to design their own information system is demanding and difficult in practice and, consequently, is rare. In order to provide the bridge between the client’s increased understanding of the form of their information systems and their technical specification, it has been suggested that structured data flow diagrams may be developed from the CM stage of SSM (e.g. Stowell, 1985; Prior, 1990; Doyle & Wood, 1991; Sawyer, 1991). Much debate has taken place concerning the problems, both practical and theoretical, of this relationship and the process of converting one model to the other. However, we suggest that such arguments become irrelevant if one concentrates upon learning and gaining an appreciation of the problem situation when using SSM rather than focusing upon a technical application of SSM’s seven-stage model with the intention of converting what has been discovered into a technical specification. The employment of SSM in this way offers a useful example of the type of application of the methodology that Checkland & Scholes (1990) refer to as ‘ideal-type’ mode 2 SSM, namely the conscious adoption of the underlying conc’epts of a systemic process of inquiry that promotes learning. It is the result of this learning process that then enables one to proceed with the design process in a more knowledgeable manner. For example, if one focuses upon the development of the ideas and understanding that one has of the problem domain (obtained as a result of using SSM as a means of ‘finding out’) into DFDs (rather than upon the translation of CM to DFD), then the epistemological problems may be removed. Therefore, at this point we should

not look to find a strict functional relationship between the two models but instead continue to develop and represent client requirements through an appreciation of client needs (Prior, 1990; Stowell, 1991).

The practice of SSM has, over the past 20 years, varied considerably and the methodology has been criticized in the literature in terms of its inability to address power and introduce radical change and it has even been accused of being an instrument for preserving the status quo (Prevost, 1976; Naughton, 1979; Jackson, 1982,1991; Mingers, 1984). However, weemphasize a point made by Checkland in answer to such criticisms that the use of SSM and the subsequent outcome is dependent not upon the methodology but upon the way in which it is used by the practitioner. The potential for radical change (‘emancipation’) comes about through the inno- vative and sophisticated use of SSM, (the possibility of which would seem to be heightened by the debate which is now articulated as ‘ideal-type’ mode 2), to bring about a rich appreciation of the situation in those associated with it. Change resulting from such an appreciation is sponsored by the recognition of its feasibility and desirability by those involved and, if we use client-led design, calls for its implementation come from the client(s) themselves. If we think of this process in terms of technology-based information systems, then ownership of the ‘solution’ by the client(s) can be completed long before we reach the stage of physical implementation.

However, experience indicates that SSM is sometimes used in a functionalist manner in that the seven stages of the methodology are used as if they were the seven stages of a technique rather than as a powerful inquiring system which is undertaken in the mode of action research. Two observations to be made from this which highlight potential problem areas to be avoided when using SSM as a framework for client-led design are:

1. the use of the methodology as if the output of the study will result in a computer-based solution (and, consequently, SSM may be criticized in the same way as technology-driven approaches); and

2. the use of the methodology as if it were a series of discrete steps which takes the practitioner from problem to solution without reflection - as if it were a computer systems analysis technique.

One difficulty which might occur in cases where SSM is used as a front-end to designing technology-based information systems is that there is a lack of appreciation about the nature of SSM itself. SSM is, we argue, the means by which an appreciation of a problem can be brought about. ‘Tools’, such as rich pictures, root definitions and conceptual models, may be used as ways of exploring the problem as well as being a practical attempt to reduce the probability of introducing preconceived ideas about a problem. There may, however, be a variety of tools, in addition to the ones provided by SSM, through which problem appreciation and representation can be attained (West, 1991 ; Stowell & West, 1992).

The current development of SSM (e.g. Checkland & Scholes, 1990) is one of the clearest examples of interpretive systems thinking that offers a practical means of inquiry and, conse- quently, criticisms levelled against the earlierforms of the methodology are now less sustainable. SSM ‘ideal-type’ mode 2 is a useful vehicle with which to structure and direct the process of

am FA Siowe/f & D West

problem appreciation through a never-ending process of learning. The important difference between SSM pre-1991 and now is the clear attempt by Checkland to convey to the would-be practitioner the hermeneutic/phenomenological nature of the methodology not just as a theo- retical concept but as a practical method of inquiry. It is this framework for inquiry which provides both a strong theoretical basis for the notion of client-led design and a means of putting this notion into practice.

END NOTE: SOME PRACTICAL CONSIDERATIONS

A number of important considerations need to be addressed when considering the design and application of technology through client-led design. For example, use of SSM may increase the awareness of those involved in the problem situation but if the problem itself can be relieved by the introduction of technology there needs to be a technical specification. The client is unlikely to possess the technical expertise to be able to define the technology) and hence this role falls to the ISA. Furthermore, if the full potential of SSM is to be realized then there should be no preconceptions about outcomes. This may be problematic for an ISA who has been called in expressly to introduce IT into the company. Finally, if the client expresses the opinion at the beginning of a study that a specific technology is required (e.g. a database), can we justify the use of SSM and hence the exploration of wider issues?

In conclusion, we believe that SSM as represented through what has been referred to as 'ideal-type' mode 2 is an approach which may be of value to the ISA. More importantly, the interpretivist concepts upon which SSM is based, namely the never-ending circle of learning and self-reflection, in principle, may be used, through the notion of client-led design, to address some of the deficiencies of current methods of computer systems analysis (Stowell, 1991).

REFERENCES

Avison, D.E. & Wood-Harper, A.T. (1990) Multiview, an Exploration into lnformation system Development. Black- well Scientific Publications, Oxford.

Benyon, D. & Skidmore, S. (1987) Towards a toolkit for the systems analyst. Computer Journal 30,2-7.

Burrell, G. & Morgan, G. (1979) Sociological Paradigms for Organisational Analysis. Heineman, London.

Checkland, P.B. (1986) Soft systems as process: a reply to M.C. Jackson. Journal of Applied Systems Analysis, 9, 37-39.

Checkland, P.B. (1992) SSM and information system de- sign: history and some current issues. Systemist, 14(3), 9&92.

Checkland, P. & Scholes, J. (1990) Soft Systems Method- ology in Action. Wiley, Chichester.

Doyle, K. & Wood, R. (1991) Systems thinking, systems practice: dangerous liaisons. Systemist,, 13(1), 2830.

Finkelstein, C. (1989) An lntroduction to lnformation Engi- neering. Addison-Wesley, Wokingham.

Jackson, M.C. (1982) The nature of 'soft' systems thinking: the work of Churchman, Ackoff and Checkland. Journal of Applied Systems Analysis, 9, 17-29.

Jackson, M.C. (1991) Systems Methodology for the Man- agement Sciences. Plenum Publishers, New York.

McDenid, D.C. (1 990) Software Engineering for Informa- tion Systems. Blackwell Scientific Publications, Oxford.

'Soft' systems thinking, information systems and client-fed design

Mansell, G. (1991) Action research in information systems development. Journal of lnformation Systems, 1, 29-40,

Miles, R.K. (1985) Computer systems analysis: the con- straint of the hard systems paradigm. Journal ofApplied Systems Analysis, 12, 55-65.

Mingers, J. (1984) Subjectivism and soft systems method- ology--a critique. Journal of Applied Systems analysis, 11,85103.

Mumford E. & Weir, M. (1979) Computer Systems in Work Design-The ETHlCS Method. Associated Business Press, London.

Naughton, J. (1 979) Functionalism and systems research: a comment. Journal of Applied Systems Analysis, 6,

Prevost, P. (1976) 'Soft' systems methodology, functional- ism and the social sciences. Journal ofApplied Systems Analysis, 5, 65-73.

Prior, R. (1990) Deriving data flow diagrams from 'soft systems' conceptual model. Systemist, 12(2), 6575.

Sawyer, K. (1991) Linking SSM to DFDs: the two epistemo- logical differences. Systemist, 13(2), 7680.

Stowell, F.A. (1 985) Experiences with SSM and data analy- sis. lnformation Technology Training, 48-50.

Stowell, F.A. (1 989) Organisational power and the metaphor commodity. In: Systems Prospects: The Next Ten Years of Systems Research, Flood, R.L., Jackson, M.C. and Keys, P. (eds), pp. 147-154. Plenum Publishers, New 'fork.

Stowell, F.A. (1991) Towards client-led development of information systems. Journal of lnformation Systems, 1, 173-1 89.

Stowell, F.A. & West, D. (1992) SSM as a vehicle for

69-73.

client-led design of information systems: utilizing 'ideal- type' inode 2. Systemist. 14(3), 99-106.

Stowell, F.A. & West, D. (1994) Client-led Design: A Sys- temic Approach to lnformation System Development. McGraw-Hill, Maidenhead (in press).

Vickers, G. (1 965) The An'of Judgement. A Study in Policy Making. Harper & Row, London.

West, D. (1991) Knowledge elicitation as an inquiring sys- tem: towards a'subjective' knowledge elicitation method- ology. Journal of lnformation Systems, 2, 3144.

Wilson, B. (1 984) Systems: Concepts, Methodologies and Applications. Wiley, Chichester.

Wood-Harper, A.T. (1992) Viewpoint: action research. Jour- nal of lnformatron Systems, 2, 235-236.

Biographies

Frank Stowell is Professor of Information Systems and Associate Head of the Department of Computing and lnfor mation Systems at the University of Paislejl His primary research interests are in the development of methods of information system definition which accord with the inter pretivist systems paradigm and, particularly with the em- powerment of the client over the design process

Daune West is a Senior Lecturer in the Department of Computing and lnformation Systems at the University of Paisley Her PhD from Portsmouth University was in sys terns and information systems development Her research interests include the use of systems concepts and tools in the development of information systems, particularly during the process of information systems requirements analysis