jwhitney / dataverse Goto Github PK

New Use Case: Had a meeting this morning with the RDA Data Publishing Workflows Working Group and Force11 Data Citation Implementation Group on our project and their resounding feedback and recommendation was that they would like for us to support the ability for one journal article to have more than one dataset attached to it. Right now all the supplementary files are bundled into a single zip file dataset that gets deposited into dataverse as a single study.

This need for supporting multiple datasets per journal article corresponds with the Joint Declaration of Data Citation Principles': Principle #3 Evidence: In scholarly literature, whenever and wherever a claim relies upon data, the corresponding data should be cited.

cc/ @jwhitney @axfelix @pdurbin

Building from discussion in http://irclog.iq.harvard.edu/dataverse/2014-07-14, it's been decided to change the current behaviour where updating same-name files in OJS which have already been mirrored in a linked Dataverse study that has not yet been released does not automatically overwrite the corresponding files on the Datavers side. This was originally done for preservation purposes, but it's been decided that this is too confusing and inconsistent from the perspective of an OJS user, and the versioning mechanism provided by Dataverse after a study has been released is sufficient for preservation purposes, so this should only affect unreleased studies.