Ghent, May 28 2013
The purpose of this workshop was to explore where the library sits in the whole data lifecycle, and pinpoint some areas as to what its role might be in the future. The workshop also invited a range of scientists to present their line of research to the audience, with idea about where the library might be able to support them..
Introduction speech by Inge Van Nieuwerburgh (UGent) en Najla Rettberg (UGOE)
What is open science? A researcher’s perspective’ by Erik Mannens (UGent, MMLAB) Presentation | Video
‘Open should be the default, and libraries can be the catalyst to promote this idea' was Erik Mannens' message to the audience.
Key points:
- Know your Data – researchers and librarians have to realize what a commodity their data represents
- Provide the tooling – while so much software is available, the library should be at least aware of what software there is out there
- Push open access in your daily work
Research Data Overview by Sarah Callaghan (British Atmospheric Data Centre) Presentation | Video
Sarah Callaghan guided the audience through the steps of the data life-cycle as well as her personal experiences as a researcher creating data.
Just like with hand-made knitwear, reconstruction is only possible if you know all the details (i.e. the metadata) about how it was originally made.
Key points:
- Make preservation and access to data a reality by promoting use of correct metadata
- Make sure researchers who share their data are properly credited
Meet the scientists, A Researcher’s perspective. Video
This session aimed at highlighting to the audience the daily workflows of researchers and their opinions on what libraries can do for them when it comes to Research Data Management.
Joris Van Zundert (humanities): Joris explained that his data is fairly unstructured, highly complex and heterogeneous. Researchers don’t come from a community of storing and sharing data. Getting credit for sharing and publishing data is key, this will be the biggest push, even more so than funder policies. Libraries could organize storage support as subject data repositories for humanities are limited. Many disciplines in SSH need more machine access to data, and small-scale projects to work this out would be valuable
Aaike de Wever (life sciences): While storage and data centres are fairly abundant for this discipline, encouraging colleagues to share data is still a barrier. The development of accreditation mechanisms still has a long way to go. Libraries should play a role in encouraging deposit, linking the research workflow within the life-sciences. Guiding researchers to repositories is also key, recommending appropriate storage for the right data.
Dirk Van den Poel (marketing): Most data is highly sensitive and he mentioned that even anonymising data isn’t always totally safe. He explained that more open software is being used and there is a push to share the end result data. However this sort of data management takes time and the library could play a role here with support, such as assist researchers to plan the workflow for processing the data, and assist in managing it, so that it can be reinterpreted by other researchers.
Key points:
- Take into account that researchers’ needs differ per discipline!
- Fear of not being properly credited and misuse of rights is widely spread
- Libraries can help sorting out practical issues like rights management, guiding towards repositories, and workflow planning
The data publishing process
'Supporting scientific communities by publishing data by Peggy Schaeffer (Dryad) Presentation | Video
After giving a comprehensive overview of what Dryad does, Peggy outlined several good pointers for libraries and data
'Data Publication: a Publisher's perspective' by Fiona Murphy (Wiley Publishers) Presentation | Video
Fiona from Wiley spoke about what publishing data is all about: why it is important in terms of being cited and credited. The growing pressure funder mandates also plays a role.
Key points from both sessions:
- Surprise researchers by proving how much their data is consulted
- Just like publishers, libraries should re-think the role they can play in the research publication process in all its aspects
- The library is best suited to offer information management expertise, breaking down the barriers and supporting workflows
- Combine management planning with guidance on data repositories , consulting with researchers.
Actively Supporting Data Management : Learning from the Approach of Three Research Institutions
These talks gave best-practice examples of how three institutions have supported research data management.Three very different library approaches to support RDM were presented. It would be interesting to see how they can combine their resources in future training materials!
Developing online data management resources for researchers and librarians' by Robin Rice (Data Librarian, Edinburgh University Data Library) Presentation | Video
Robin led us through the UoE's research data management roadmap and their ground-breaking data policy. She outlined the Mantra course, the reusable materials and the librarian training. This training looked more at ‘facilitation’ of skills, rather than teaching, and getting learners' buy-in from the very start.
'3TU.Datacentrum and Data Intelligence for Librarians: support to staff and researchers on data management' - Ellen Verbakel (Data Librarian, 3TU.Datacentrum, Delft University) Presentation | Video
‘Data intelligence for library staff’ was developed for library support staff in at 3TU (a consortium of three technical universities) specifically to advise researchers on efficient ways to add value to their data. This is a move to position the library more centrally in the support of data intensive science.
RDMRose -an open educational resource on Research Data Management tailored for information professionals by Andrew Cox (University of Sheffield) Presentation | Video
In describing the White Rose ‘learning by doing’ approach, Andrew emphasized the importance of librarians discovering for themselves what research data means. The courses focused more on management issues.
Key points:
- Help participants gain confidence by bringing them in contact with researchers during training
- Be flexible and provide variation in your training materials: no use for a 'one size fits all' approach
- Combine training resources, get in touch with other institutions organizing similar courses
Services and Tools
‘ZENODO - A new innovative service for sharing all research outputs’ by Lars Holm Nielsen (OpenAIRE) Presentation | Video
Zenodo (www.zenodo.org) is the re-branded OpenAIRE orphan repository for both publications and datasets. In terms of data it is suitable for long-term data .The strength of Zenodo is the ‘Community collections’ element which will flag up the relationships between publications and data. The collections can also be exported into local systems, including the links between objects.
CKAN - A Research Data Management Tool. The University of Lincoln's experience in implentation' by Joss Winn (Orbital project, University of Lincoln library) Presentation | Video
For libraries choosing a system to manage research data, Joss provided an overview of his experiences in choosing the right tool for RDM. He published a paper on using CKAN which provides a very comprehensive evaluation of its suitability for libraries.
Key points:
- While RDM might be fairly new, it involves many processes that have existed in libraries for years –visibility, discoverability, curation, preservation
- CKAN is a well designed, flexible and extensible data management system that provides benefits to researchers, data librarians, and their institutions
LIBER/OpenAIRE Panel Session Video
- Wilma van Wezenbeek (TU Delft Library, Delft)
- Lukas Koster (Library, University of Amsterdam)
- Wolfram Horstmann (University of Oxford Library)
- Chair: Kevin Ashley, (Digital Curation Centre)
Discussion and questions focused around which parts of the data lifecycle should the library take part in; how well placed is the library to fulfill data management needs; how can generic library services suit data management.
Some key points from the Panel Session
- Libraries can give practical advice, training and advocacy
- Libraries can help with discoverability, visibility, and offering new contextual information to researchers
- Library can help at the beginning and end of the life-cycle (planning and curation), especially when researchers move onto another institution
- Libraries are not best suited to carry out storage and active data management. This is more the role of data centres and data libraries.
- Starting to carry out a Data Audit would ‘break the ice’ to start data management planning
- In the long term, libraries should mediate with publishers to work on standards for extracting metadata
- LIBER’s ‘10 recommendations for data management’ are a key resource for the community.
- Another valuable resource is the DCC’s guide on ‘Setting up a research data service’
Concluding Points from the workshop
- There is a perceived lack of confidence about dealing with data in research libraries
- Libraries' strengths: Acting in an advisory Role - Guiding researchers to repositories - Making Data Visible - Support for RDM planning
- The library shouldn’t work in isolation – a collaborative approach with other units and other libraries might be the way forward in order to harness richer research information
- The library should play a key role implementing standards. Putting standards in place is crucial, enabling institutions to implement standards.
- Training: Embedding RDM in library courses so that a new breed of data librarians is integrated from the start