In this thesis it is argued that the use and design of computer artefacts can be fruitfully studied by focusing on use problems and design contradictions, defined as problems and contradictions in relation to organisational improvements planned by the IT-practitioner. The position taken in this thesis is that the important thing with computer artefacts and systems design is their contributions to human organisations at large. The purpose is to contribute to the design of contributing computer artefacts by studying, first, use problems and design contradictions, and second, the importance of IT-perspectives held by the practitioner. That is, the practitioner's tacit assumptions about what computers really are. Based on Action Science three case studies are made. What was found in these cases were use problems such as context dependency, discriminating information systems, misinterpretations and misuse. Among design contradictions were observed narrow design domains, unimaginative and pointless staff trainings, too fixed positions concerning goals and problems, and an assumption concerning relevant knowledge that effectively was excluding important knowledge related to the organisation at large. In all of these cases a tool perspective was interpreted as the dominating IT-perspective. In order to find out whether or not a tool perspective can be regarded as causing the use problems and design contradictions observed, the thesis examines this IT-perspective more closely. The conclusion made from this examination is that the tool perspective can be causing the problems and contradictions observed. In an attempt to formulate an alternative to this tool perspective theoretical studies into systems thinking and reflective practice were made, ending up in a framework labeled contextual IT-understanding. It is a framework built on principles such as a systems view, purposefulness, product-producer perspective, and constructivism. It is also built on assumptions such as that practitioners have a capacity to make sense out of a situation that initially makes no sense, that their actions are formed through a reflective conversation with the materials of the situation, that design is a matter of reflection, both in and on the actual action, and that in training it is important to make the trainee start to reflect on the tacit assumptions guiding her actions. According to a contextual IT-understanding an information system is a systemic and culturally dynamic artefact and as such, an artefact that will serve certain interests at the expense of others, will be ambiguous, will be misinterpreted, will be misused, etcetera. An information system should also be considered as part of a larger value creating process under the constant influence of a tacit dimension, implying that the relation between an information system and organisational improvements is not causal, that a lot of important design will take place in use, etcetera. More radically though, every information system can be seen as having a tacit dimension acting like an intellectual superstructure, and this tacit dimension can be fruitfully utilized in design. On the overall level a conclusion is drawn that both researchers and practitioners should emphasis the IT-perspectives held by IT-practitioners and the tacit dimensions influencing practice. / digitalisering@umu