Data Management Planning as Infrastructure
Thursday, September 29, 2016 - 14:00
My talk will consider data management plans (DMPs) from two perspectives: looking at the systems which produce them (such as DMPonline) as part of a wider research infrastructure, and also considering the plans themselves as part of a data infrastructure. It is established that good data management leads to the creation of more
reusable data, and can reduce the cost of doing so whilst initial research is being carried out. It’s also recognised that data management planning, before and during a research activity, can help ensure that good data management is practised. As a result, systems have been created to help researchers create and share data management plans and many research funders require their creation. At present, these systems may have an internal view of a plan as a highly complex data object, but they emit plans as documents, in PDF or MS Word format, because that is what funders expect to receive. We are thus failing to make use of the information contained in plans in later stages of the research lifecycle, and in the administration and monitoring of research. We run the risk that the production of data management plans becomes a compliance issue which fails to achieve the intended goal of the data management planning process – the creation of data which is more usable in its original context, more reusable in other contexts, and to do all of this in a manner which is efficient and recognisable to others as good practice.
Is it true, as Eisenhower famously said, that “… plans are useless, but planning is essential”, or can we do more with the plans that our funders and research organisations demand? How can we reduce the burden of creating plans whilst also maximising the value of doing so for the original researchers, their funders, and those who can reuse the data which is produced? What other parts of our e-infrastructure can DMPs draw on and contribute to, and what are the priority actions in a European context? Timely and integrated action should ensure that data management planning is a low-cost activity that makes efficient use of existing infrastructure, helps future planning for infrastructure at every level, and maximises the production of usable and reusable data. Failure could result in the creation of another costly compliance process with little observable benefit.
This talk will represent the collective views of the developers and operators of the world's leading DMP services - DMPonline and DMPTool - and the Research Data Alliance interest group which is looking at the future form of data management plans. We believe we have potential answers to these and other questions regarding DMPs, but these need greater socialisation and acceptance by the many other actors involved. This talk and the discussion which will take place in the session will form part of that socialisation process.
Researchers with responsibility for data management; infrastructure providers with systems that support data management planning or could consume the information in them; research support staff who assist in data management activities; funders and policy makers.
Benefits for Audience:
The audience will learn more about the value of the data management planning process and the plans which are produced. They will gain insight into how these plans and systems can form part of an integrated research infrastructure, from the level of research group to national and international systems. They will understand how DMPs can further the aims of data creators, data reusers, e-infrastructure providers and policy makers. Each will gain clarity on what actions they can take to further the potential gains and will have the opportunity to prioritise actions towards those ends.
Topic 4: Working with data
|Kevin Ashley||Digital Curation Centre|