Dynamic data and the meaning of publication

Added by Stephane Goldstein on 04 July 2007 13:04

Assiduous readers of this website will have noted the RIN’s interest in exploring the notion of data publication - our study of this area is now underway.  Three current initiatives in the field of chemistry illustrate the extent to which the meaning of ‘publication’ can be stretched, and the way in which ‘publication’ can become meshed into the experimental research process itself. Importantly, they all address the problem of precious information that may be lost because of the obvious limitations that the scholarly journal format imposes on fully and presenting, in a visual way that suits researchers, the increasingly vast amounts of research data outputs.

The JISC-funded Repository for the Laboratory (R4L) project, at the University of Southampton, is a facility that allows data to be captured semi-automatically at the very point of creation in the laboratory, and subsequently managed and enriched.  Thus the repository is not passive, but becomes a dynamic working tool, a virtual space for data to be annotated, discussed (there are blogging and ‘scribbling’ features) and curated.  It can be seen as a halfway house between a conventional repository and formal publication.

Open Notebook Science Using Blogs and Wikis, based on work at Drexel University and as reported and discussed in Nature Precedings, is a not dissimilar concept.  Its premise is that it allows for transparent and open reporting/discourse/commentary on experimental results.  The novelty of presenting some of these in Second Life adds a certain piquancy to the process.

Neither of these initiatives is brand new, so apologies for not being the first to highlight them - but a reminder about such interesting work is timely.  Finally, the third development that I alluded to confronts the data loss issue from another angle.  As reported recently in Research Information, the Royal Society of Chemistry’s Project Prospect, launched early this year, aims to make journal articles machine-readable.  Using text-mining software, this allows for compounds, concepts and data within such articles to be annotated and linked to a wealth of underlying data resources. The consequently increased visibility of this data can also be said to extend the notion of publication.

More about this sort of development, of course, when the RIN study is completed towards the end of this year. 


© Research Information Network 2005–2009