Nov 19, 2013

DLF forum notes: Data – sharing – libraries – culture

A recent Digital Library Federation (DLF) forum started with an inspiring keynote by R. D. Lankes and his challenging of the monopolies of content delivery in higher education and the neutrality of the library profession. He argued that librarians should improve society by facilitating knowledge creation, which includes not only providing access to resources, but also teaching literacy, genres, and communication, creating learning environments and motivating people to learn and research. He also emphasized that a librarian is somebody who has either training, professional experience, or spirit, thereby shifting the emphasis from degrees to actual knowledge and passion.

Of lesser success was my leading of a birds-of-a-feather session about Research Data Alliance (RDA). BoFs happened during lunch and it’s probably not a good time to learn about a new organization. Spreading the word about new initiatives is also hard because they are new and there is more anticipation and preparatory work, rather than something that is ready to use. I had several good conversations about RDA, but there could have been more.

All sessions were informative and productive, but the one that got me thinking was a session that I couldn’t attend "Creating the New Normal: Fostering a Culture of Data Sharing with Researchers". It’s a rich topic, so below is some food for thought - my interpretation of the session theme based on the materials prepared for the session by organizers and community notes taken by participants during the session.

The session was based on the Data Information Literacy (DIL) project that is looking to identify skills, capacities and toolkits for data management and leverage information literacy to bridge the disconnect between faculty, graduate students (who are often the data managers in scientific labs) and librarians.

The disconnect stems from different needs in relation to data – data collection vs analysis vs preservation and access. Even though researchers may recognize the importance of data management, they rarely consider it being part of the curriculum or articulated research culture. To simplify, graduate students collect data and both faculty and students analyze it and work on publications. The messiness of data collection and storage is something that bothers many researchers, but they don’t necessarily know what to do about it.

Librarians are increasingly interested in incorporating research data into their library collections and providing access to it as part of their service. They would like datasets to be better prepared for storage and sharing, and they are ready to help. They also have a mission of helping others to address their information needs, but they don’t necessarily have the power to do that. How could these disconnects be bridged?

By working through three scenarios proposed for discussion, the session seemed to approach disconnect bridging via embedding librarians/data specialists into research teams and projects, learning about researchers’ needs and helping them with their data, while teaching them good practices of data management. Librarians’ expertise can be useful in the areas of file organization, metadata, and tools for storage and sharing.

This approach is good, but it’s quite demanding in terms of resources. I agree that data literacy should be embedded in a larger teaching of proper information management (including ethics, security, authoritativeness, etc.) and this could help re-use existing channels of library instruction and minimize resources. At the same time, I wonder whether we should also think about the culture of data sharing in terms of private/public epistemic objects. Data is still a private object, which will constantly undermine the "new normal" of sharing, because we don’t typically share objects we consider private. The new norm then should be that data are "shareable" from the beginning.

In some organizations and domains data are already open and shareable, for example, data from the large observatories supported by the US government. Consequently, data producers in those organizations may have better data information literacy. For other "private" domains, particularly, in the social sciences, there are still a lot of barriers and fears. Would a public culture of data sharing mean peer review of data collection instruments? Survey respondents as co-owners of data? Data curators as independent decision-makers with regard to choosing and preparing data for public use? A lot of possibilities come with the idea of data sharing cultures.