The Genus of Information Infrastructures: Architecture, Governance & Praxis
MetadataVis full innførsel
The complexities in large-scale IT solutions have been acknowledged as a challenge. These complexities arise from interconnections between socio-technical components of what has been referred to in literature using multiple terms such as Information Infrastructures (IIs), Digital Infrastructure, e-Infrastructure, cyberinfrastructure etc. Dividing a complex problem into smaller parts is the prevalent strategy to comprehend complexity in a better way. A similar strategy is to use taxonomy or mental classification to better describe a complex observed phenomenon. In this thesis, I suggest a taxonomy to classify activities within an II as - Architecture, Governance and Praxis (AGP). As taxonomies go, the above classification is comparable to the taxonomic rank of genus (plural: genera), coming from the Latin (genus) and Greek (genos) meaning descent, family, type, race, stock etc. I use the large body of existing literature about IIs and use the above taxonomy to classify the activities in IIs. This taxonomy provides clarity to observe II evolution and provides a systematic view to see what activities helped establish the II. Health Information Systems (HIS) are fragmented and results in duplication of work, ineffective use of resources, incomplete and incorrect information. When data from multiple HIS are brought together, we create an integrated eHealth infrastructure (IeHI) rigged with complexities that are difficult to manage. This thesis covers my participation in the development of two open-source software, DHIS 2 and OpenMRS. These two software systems have communities of software developers, implementers, consultants working at different implementation sites, who come together to establish an II for health. This thesis includes a collection of 6 papers. I start with data collection for HIS using mobile phones (mHealth) and design strategies for scalable mHealth solutions. Then I suggest ways to understand “success” of mHealth implementations. Realizing that design of successful mHealth solutions needs user participation, I suggest OpenScrum, a software development methodology to enable community participation. I then highlight that there is inherent security risk in such contextual software development. Once data comes from multiple sources, it brings forth big data challenges and often data use becomes a problem. We observe that appropriate analytics tools help data use. In the last paper, I highlight that new business models using cloud computing are required to be able to sustain analytics in low-resource IeHI. I observe that the interplay of the earlier mentioned three genera of activities (AGP) results in what is described in literature as infrastructuring i.e. work that is done in conceptualizing, designing, developing, using, and scaling an II. The AGP taxonomy allows observing the interplay of activities within an IeHI at the micro-level and associate them to larger re-combinations at the macro-level. The activities in one genus has effects in other genera. Yet, mapping and quantifying consequences to activities is difficult due to the inherent complexity in II. I call this view of causality, with property of sentience as Karma, a term used in Buddhism to describe the interconnected actions and their effects. I use Karma to explain that as actors associated with an II, the actors perceive effects of actions as good (stabilizing) or bad (destabilizing), whereas these events have perception of causality that is often a time-limited view of the observer or interpretation of the actor.