I am curious about this new group Proposal: Research Data Management Since there are also many other groups that touch this topic, I was wondering whether a StackExchange group of its own is needed. What is there difference to the others?
My perspective would be this: as a field, Research Data Management (RDM) is focused more on the structures and best practices that surround the data, documents, documentation, and code that are produced as a part of a research project, rather than the specific data and analytical techniques used.
Within the realm of RDM, our focus is partly on the short term - trying to identify methods for organizing data projects, establishing conventions for file sharing and labeling, etc. But what this field really offers is in the long term view - establishing standards for long term data access, using open data formats, supplying thorough metadata, etc. In a nutshell, I’d say it’s about helping people adhere to the FAIR data principles of findability, accessibility, interoperability, and reusability.
I think many people can recognize the frustration of joining a new organization and overcoming the initial difficulty of navigating a complex and byzantine structure of shared folders and cloud storage. Or trying to reproduce the results of a study from a decade ago only to find that the data referenced in the paper is in a proprietary format that uses software from a company that is no longer in business. Our goal is to identify what the pain points are in handling research data, both short and long term, identify tools and techniques to best overcome those pain points, and figure out how best to communicate to and convince stakeholders that adopting these is to their benefit.
I'd second the statement by Jonathan Hartman but would like to add my view.
The sites warren mentioned are predominantly focused on technical aspects, ie. "where can I can data?", "which statistical analysis should I do here?" etc.
In RDM.SE I can also see technical question for which I would have a hard time placing them into any of the mentioned other sites:
- which metadata standard would be more appropriate in field X?
- how to achieve X with RDM-tool Y?
- which terminology/ontology to use for field X in standard Y given dataset/software/... Z?
I'd imagine a RDM.SE to also cover "softer aspects" like
- how to establish proper RDM-culture within your institution or field? including questions around existing approaches to "introduce" RDM
- how long does "long term archival" have to be?
discussions around which of the above are within scope of a future RDM.SE are to be had, but in my view RDM.SE has a broader scope than any individual of the above. Besides, one could argue that some of the aforementioned sites should be a subcommunity within another site. Most of those discussions are rather opinionated and have no objective answer. the only "objective" criteria is the uptake within the community measured likely by the activity on such a site
I would add that Research Data Management is not only for people computationally working with data (using statistics, data science, AI or other mathematical models) but might also address people taking digital notes or preserving other types of digital knowledge. Those would probably be regarded out-of-scope in Stats.SE or DataScience.SE
Interestingly, there has been a proposal for Library & Information Science many years ago. This kind of goes into the same direction but was closed during beta phase.