Modelling the Development of Research Data
Ball, A., Darlington, M. and Howard, T., 2010. Modelling the Development of Research Data. In: 6th International Digital Curation Conference, 2010-12-06 - 2010-12-08, Chicago.
|PDF (erim0pos101011ab10.pdf) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader |
There are a whole host of challenges that face researchers as they attempt to manage their data: clarifying intellectual property rights issues, determining the optimal formats and standards to use, forecasting expenditure on preparing data for archiving, and so on. Among these challenges, one that is sometimes overlooked is that of simply keeping track of all the data that need to be managed. In some cases, no measures at all are put in place to track the data, meaning that researchers must rely on their memory to make sense of the files they generate in the course of their research. In others, directory and file naming conventions may be used, according to a researcher's own scheme or one set out by the research group or department. Still other researchers may keep written records of which data arose from which processes. These techniques each present a different trade-off between investment and return in terms of a researcher's effort to document and understand the data, respectively. They also provide third parties - curators, consumers - with different amounts of contextual information with which to understand the data, though this is in turn affected by the quality of the documentation of the technique in use. The ERIM Project is devising an alternative technique for keeping track of data, with the aim of being: intuitive for researchers when recording progress, and for both researchers and third parties when trying to understand the data records; applicable to a wide variety of data types and disciplines (though developed in the context of engineering research); compatible with existing local practices; amenable to automation, at both the writing and the reading stages. This technique, Research Activity Information Development (RAID) modelling, provides a formalism for describing how data records - and conceivably individual data - relate to one another. The RAID models may be represented as diagrams or as XML. Underlying RAID modelling is an ontology for data, their states and the processes that relate them.
|Item Type||Conference or Workshop Items (Poster)|
|Creators||Ball, A., Darlington, M. and Howard, T.|
|Departments||Faculty of Engineering & Design > Mechanical Engineering|
Innovative Design & Manufacturing Research Centre (IdMRC)
Actions (login required)