There is a definite interest in data management plans (DMP); there was little doubt about that when SND hosted their first network meeting of 2019, on March 12 in Gothenburg. The theme of the day was “Data management plans – how and why?”, and it saw a record number of participants: 117 people from 35 higher education institutions and government authorities had signed up for the day.
On the international scene, data management plans for research data have been mandatory for some time, and in the last few years this has become an issue of concern in Sweden as well. Research funders Formas and the Swedish Research Council (VR) already require data management plans for research projects that receive grants from them, which has raised the question of a national DMP tool. In the beginning of 2019, VR tasked a workgroup with creating a template for data management plans, and the foundation for such a tool. Suddenly, this has become an urgent issue. What should a national DMP tool look like? What possibilities and limitations are there? This is of course a matter of perspective, which became clear during the discussions in SND’s latest network meeting. The topic was discussed from a variety of angles and several actors gave their view.
Important questions and perspectives for a national DMP tool
The opening speaker was Sanja Halling from the Swedish Research Council (VR), who gave a brief presentation of VR’s coordinating mission regarding open access to research data, and introduced the workgroup for a national DMP tool. The group, which consists of representatives from different actors in the Swedish research community, has been asked to investigate the national coordination of data management plans, and to ultimately produce a digital DMP tool. Their work includes mapping the needs of various actors, and what support and tools already exist. Sanja Halling means that the design of a national tool is a matter of finding the smallest common denominator for what a basic DMP tool should contain. It must then be adjusted for each research field and specific local needs of the organisation where the tool is being used.
Several other representatives from this workgroup attended the network meeting. Among them, the SUHF (The Association of Swedish Higher Education Institutions) research data group, represented by Sabina Anderberg. She introduced the recommendations that SUHF’s research data group, together with VR and SND, are working on, and which will serve as a basis for the continued efforts with data management plans. Sabina Anderberg also highlighted other important aspects of the coming development of a national DMP tool; such as the importance of getting researchers involved in developing the tool, so that it will actually benefit research. She furthermore encouraged interested parties to join the common work instead of pursuing these matters on their own.
Sofia Särdquist from the Swedish National Archives (RA) emphasised the importance of coordinating the DMP work with the work with archival descriptions. The information that has to be described according to the National Archive’s general regulations (RA-FS) could also be included in a data management plan.
The technical aspects of a DMP tool were discussed by representatives from SNIC and SND, who are also members of the VR work group. Dejan Vitlacil from SNIC argued that the increasing need for storage of research data creates a demand for more guidelines and policies on how the e-infrastructures should manage data. In order to achieve secure and effective storage of data, you also need information: How will the data be backed up? Do the data contain sensitive information that needs to be protected? Which data shall be saved and who should have access to them? This type of information corresponds largely to what is found in a data management plan. Johan Fihn Marberg and Olof Olsson from SND contributed with a presentation of existing DMP tools and pointed out their strengths and weaknesses. They also posed important questions about possible multilingual integration of university-specific systems, export formats, and how a DMP template could be structured in order to be understood by humans as well as machines.
Norwegian Experiences
From UiO we learned about their current work in creating a DMP template. This DMP template should be designed in order to make it as simple as possible to get research data that meet the FAIR data principles, but still make the template meaningful and user friendly. It should also fulfil the requirements from various research funders, to prevent researchers from having to work with several different documents. One of the templates that are already being used in Norway is the NSD’s DMP template, which was demonstrated by Trond Kvamme. This template is a first version which will be expanded. Kvamme mentioned that the NSD template has been well-used so far, with 1,100 created data management plans since it was launched in November 2017. According to Kvamme, it has contributed to making researchers more aware of how to manage personal data, and how to archive and share their data. But it has been difficult to communicate the benefits that a DMP may have for researchers, and the DMP does not seem to be used as a living document. In the future, NSD see a simpler, machine readable DMP template that builds on adaptable modules.
So what is next for us in Sweden? The investigation into a national DMP tool continues, led by the VR workgroup. There is clearly a lot of things to discuss, and we are eagerly looking forward to see what is ahead.
Here you can see some of the presentations (in Swedish only).