Yesterday I attended a round table discussion hosted by the Belmont Forum about the release of their position on data accessibility statements and digital objects management plans. (It’s a bit of a mouthful, but the reason is that they are aiming to be clear and comprehensive around what they are asking to make it easier for researchers, publishers and other stakeholders to get to compliance around this policy.)
You can read their position paper — Draft DAS Statement and Policy for October 2018 Plenary - Google Docs.
This work built on a number of previous workshops, including this one.
There were people representing the following orgs at the meeting yesterday: Belmont, Hindawi, PLOS, Taylor and Francis, the IOP, Digital Science, Figshare, Springer Nature and myself from SAGE.
The Belmont forum is an umbrella group of funders representing 26 funders. NERC form the UK is a member as is the NSF.
The guidelines they have posted are sensible, and well thought out. For me at the heart if their recommendations is the requirement to have a data accessibility statement available outside of the paywall for all articles that they fund.
The DP3 is concentrating chiefly on delivering a set of template Data Accessibility Statements (DASs) for guiding Belmont Forum grantees when publishing their research results. The DAS is included as part of a journal article and articulates which data underlie a paper, where the data are available and under what conditions they can be accessed.”
There was a great discussion at the round table yesterday. Points that came out that I found useful were:
- It is easy to require a DAS (data availability/accessibility statement), but harder to dictate what goes into it. E.g. in life science journals if you need to check for gene deposition it can be harder to implement.
- Though this is hard, there was broad agreement that even requiring a DAS is a step in the right direction
- There was general agreement that doing this increases that amount of data that is being made available.
- The question around where responsibility lies for checking these has no simple answer.
- PLOS have found it useful to have a structured question in the review form. This can help when the manuscript is being assessed.
- At Springer Nature they have about 400 journals that have some kind of requirement for a statement.
- Rebecca Grant and Iain Hrynaszkiewicz wrote a paper showing The impact on authors and editors of introducing Data Availability Statements at Nature journals.
- We talked a bit about the impact on the social sciences and the additional complexity of this kind of data.
- Unsurprisingly institutions were mentioned as a key stakeholder.