In mid-January, we (and our poster!) traveled to Potsdam for the Symposium Forschungsdateninfrastrukturen 2013. Organized by five projects funded by the German Research Foundation (Radieschen, re3data.org, KomFor, EWIG, BoKeLa), the symposium offered workshops to discuss financial, organizational, technological, and legal aspects of (a) future German research data infrastructure(s). Broadly understood as a network of organizations, policies/legal regulations, services, and software tools, infrastructures were discussed on many different levels. Thus, sometimes the perspective was very granular, looking at the role of the individual researcher within this network, but often the same discussion also took a bird’s-eye view of the landscape, e.g. when discussing visions for and challenges to a “Research Data Infrastructure 2020”.
In his keynote lecture, Torsten Reimer (JISC, @torstenreimer) took both perspectives into account. Using Virtual Research Environments as a concrete example for the state of research data infrastructures, he looked at the current and future role of VREs based on experiences from JISC’s VRE program. An evaluation of the program outcomes showed that sustainability of the VREs was an issue, as a considerable number of them were no longer in operation or had no viable business model. Drawing on the lessons learned from this program and other JISC initiatives, Reimer extrapolated challenges research data infrastructures face, and which threaten their sustainability and cause them not to be used (these included, for example, cost and business models, support offered, usability, interoperability between different platforms, etc.). As so often, Reimer observed, not technology itself is the main issue but to sustain the tools and platforms developed and to make and keep them usable. Thus, the technology needs a supporting infrastructure of (national or organizational) policies and training to support its use – a task undertaken, for example, by the JISC Managing Research Data program.
In the following workshops it seemed that discussions often returned to this “Bermuda triangle” of technology, policy, and the individual researcher as we were struggling to disentangle and clarify their relations. To us, the discussions showed to that the “interface” between infrastructure (itself often a rather fuzzy term) and “researcher” is far from clearly defined and that hence relations between the different protagonists and stakeholders are hard to grasp. Consequently, the buck was passed quite frequently from one to the other – for example with regard to responsibilities in research data management and data sharing. Do we need additional incentives (yes, plenty of use of the “carrots and sticks”-metaphor in the Research Data Management workshop) to convince researchers of the value of data management and data sharing practice, or shouldn’t researchers recognize data management and data sharing as essentials of transparent, reproducible research and a responsible use of public money. If we need more incentives, who is responsible for providing them? The research institutes or universities employing the researchers? The (much too abstract) “system” of scholarly communication?
There seemed to be consent in the RDM workshop on the issue of both incentives and existing data management tools, which were largely regarded as sufficient and well on their way. It became clear, however, that on the “infrastructure” level – understood in this case as the network of local, national, and international actors and policy makers – much more coordination is required to turn the abstract idea of “shared responsibility” into something concrete and workable.
In the workshop on cost, strongly focusing on digital data preservation and dissemination, this question of responsibility was closely tied to that of “who pays”? Is it the government’s responsibility to implement and maintain a (centralized?) research data management and archiving infrastructure? Or will archives, research funders, data producers and/or users pay for data archiving and dissemination?
Many of these questions remained unanswered (which doesn’t mean we didn’t have good and controversial discussion about them!), and to us this made it very clear that there is much to clarify and spell out before we can even envision (let alone begin to realize) a ”Horizon 2020” for the German Research Data Infrastructure. Yet, we are also convinced that this should not keep us from starting off in a small way – while the question who will pay for digital preservation may ultimately have to be decided on an infrastructure level, archives can and have to start using existing cost models to get a first grasp on their cost. By promoting and teaching good research data management to researchers (given they are willing to come) we can help bringing about the change needed in the culture of research and scholarly communication to make data sharing a matter of course. And if it takes carrots and sticks, to hell with it, we will wield those too!
Many of the presentations are available at http://www.forschungsdaten.org/uber-radieschen/projektveranstaltungen/symposium-forschungsdaten-infrastrukturen/. A summarizing report for the workshops (in German) will soon be made available on this page. You may also want to have a look at the Twitter feed for #FDI2013.