This thesis is based on my experiences participating in a global Open Source development project of a health information system (HIS). The software is developed in multiple countries, including India, Norway and Vietnam. It is also used to gather health information in the same countries.
The author has participated in the development and training in an action research approach. I have developed an integration between their health system and another Indian HIS, which is my main empirical findings. My participation has also involved training sessions with users and training of Indian developers in India. Based on the experiences from this endeavour I have explored the use and effect of Intermediary Objects.
Based on my research I argue that concept of Intermediary Objects can be used to explore and analyse a development project, but the concept is too broad and undefined as it is now. There is not a wide range of research on the concept. I also argue that the concept needs a clearer definition and I also propose some new properties that would help define the object, so that it can be used in a better way to describe and analyse a design process.