Scientific Data, the latest offering from the Nature Publishing Group, was launched earlier this week: http://www.nature.com/sdata/.
I'm
particularly pleased to see the dedicated Data Citations section; of course, there are multiple ways to present data citations (within References, within Acknowledgements), but this method brings visibility to this practice.
The featured dataset is coordinated by Australian researchers (Uni of
Tasmania, Institute for Marine and Antarctic Studies.
This announcement provoked some email traffic on the JISC list re: citation of dynamic data - one of the most FAQ ANDS receives about data citation. In response to a question about this, the folks from Nature responded as below:
ReplyDelete"We understand that important datasets often grow and evolve. The Data Descriptor - like other publications - is a snap shot of a research project at the time of publication. In line with this perspective, we ask authors to archive a "static" representation of the data alongside each Data Descriptor. Some authors additionally point to a website or resource where users can find the most up-to-date version of the data (see, for example, Hao et al http://www.nature.com/articles/sdata20141 ).
In some cases, as a dataset grows a new Data Descriptor might be warranted, which we would link to the original Data Descriptor. In rare cases where the manner in which the data is archived must be changed, authors could also publish an formal 'addendum' to the Data Descriptor. Retractions would only be relevant if new findings fundamentally undermine the reliability of the data, or if the authors take action to remove the published, static version of the dataset from the public sphere.
Until we have experience of various scenarios I see these types of cases being considered on a case by case basis and us being guided by our Editorial Board. As we build our experience more concrete guidelines or policies with them emerge."