I2S2 Final Report

One of our main aims has been to investigate the research and data management infrastructure
needs of researchers in the Structural Sciences by examining issues relating to scale, complexity and
inter-disciplinary issues over the useful lifetime of research data. It is important to realise that more
and more scientific research is derivative in nature, dependent on data generated, managed and then
made widely accessible to third parties for repurposing and reuse.
We have found that individual researchers, groups, departments, institutions and central facilities
appear to be all working within their own technological frameworks so that proprietary and insular
technical solutions have been adopted (e.g. use of multiple and/or inconsistent identifiers); this makes
it onerous for researchers to mange research data which can be generated, collected and analysed
over a period of time, at multiple locations and across different collaborative groups. Researchers
need to be able to move data across institutional and domain boundaries in a seamless and
integrated manner. Furthermore, there is an acute danger that processed and derived data in
particular is likely to be permanently lost since at present these types of data typically reside on an
individual’s laptop, on DVDs and even on memory sticks.
We have therefore attempted to understand the nature of the interfaces between individuals,
institutions and with central facilities and scoped an integrated data management framework which
allows researchers to work across these boundaries. This has involved aligning infrastructures and
providing data management tools for individuals that fit into this framework. In particular, we have
prototyped a desk-top tool that enables a scientist to store and manage all his/her research data as
they are generated, collected and processed during the course of scientific experimentation. An
integrated approach to providing robust data management infrastructure enables an efficient
exchange and reuse of data across disciplinary boundaries; the aggregation and/or cross-searching
of related datasets; and data mining to identify patterns or trends in research and experiment results.
A related key aim of the project has been to identify the costs and benefits of the integrated approach
proposed by the project. Two parallel benefits case studies have explored the perspectives of “scale
and complexity” and “research discipline” throughout the data lifecycle and resulted in tools that have
the potential to be useful to a much wider community.