Emancipate your data | Chemistry World

abernard102@gmail.com 2013-09-25

Summary:

"The topic of sharing data, big and small is, well, rather big at the moment.1 Data is the currency of science; sharing it enables research, and doing so transparently and openly supports reproducibility. However, approximate estimates are that more than 80% of the data collected in chemistry is not available in any open or searchable manner, which represents an enormous wasted resource and creates the potential for a lack of transparency. With funding agencies in the UK and abroad also pushing to increase transparency for publicly funded research, it is time to rethink how data is shared. Share and share alike Digital repositories, initially associated with open access publishing but now focusing on data, are a possible solution. My own journey here began with an experiment in 2001, when Peter Murray-Rust and I published an article showing how data could be semantically integrated2 into the narrative of an article, a result of the Chemical Markup Language project.3 Peter coined the expression datument4 in 2004 to describe these enhanced articles, and I set out to create more examples (about 40 during the period 2005-2013)5 as an encouragement to others, most of which have links to our own digital repository.6 One of these coincided with the launch of Nature Chemistry in 2009. Anticipating that they might be willing to experiment, I submitted (and they accepted)7 an interactive exploratorium, hosted by the journal itself, and followed this up with a variation looking at bonding to helium.8 Both items contained interactive figures, the latter also incorporating re-usable data, and the full data sets in each case were deposited into this new breed of digital repository. However, when David Scheschkewitz and I recently submitted another such article to the same journal, we were informed they could no longer handle such interactive tables. We had to find a way to host and present our data-rich object ourselves ...  Our solution is to propose a two-component model, in which the article and the data are separately published. In our case, Nature Chemistry has published the article, and Figshare the data.  In this scenario, the article is the ‘narrative’ into which the relevant, yet still distinct, data is woven. A key point here is that the data itself is citable (a principle of the recent Amsterdam Manifesto on sharing data). With both components being individually citable, they become equal symbiotes. The narrative9 cites the data and the data10 can refer to the narrative. The data inherits trust from the narrative’s peer review, and the narrative inherits a date stamp and integrity from the data.  Each can have a different publisher and, importantly, the presentation of each can be optimised for its own needs.  Ideally, the data should have two layers: the raw data and an interface to present it. The reader can then access either layer, depending on their needs. Software (for example, data mining, semantic annotation and search11 or error detection) would focus on the raw data ..."

Link:

http://www.rsc.org/chemistryworld/2013/09/open-repository-data-sharing-rzepa-figshare

From feeds:

Open Access Tracking Project (OATP) » abernard102@gmail.com

Tags:

oa.new oa.data oa.publishers oa.policies oa.comment oa.open_science oa.figshare oa.repositories.data oa.repositories

Date tagged:

09/25/2013, 17:43

Date published:

09/25/2013, 13:43