1. ARL Workshop on New Collaborative Relationships: The Role of Academic Libraries in the Digital Data Universe September 26-27, 2006 ARL www.arl.org Prue Adler Associate Executive Director Association of Research Libraries prue@arl.org

2. Overarching Recommendation NSF should facilitate the establishment of a sustainable institutional framework for long-term stewardship of data. This framework should involve multiple stakeholders by: supporting the research and development required to understand, model, and prototype the technical and organizational capacities needed for data stewardship, including strategies for long term sustainability, and at multiple scales; developing and supporting training and educational programs to develop a new workforce in data science both within NSF and in cooperation with other agencies such as the Institute of Museum and Library Services; and developing, supporting, and promoting educational efforts to effect change in the research enterprise regarding the importance of the stewardship of data produced by all science and engineering disciplines/domains. http://www.arl.org/info/events/ncr.html DRAFT

3. Fund projects that address issues concerning ingest, archiving and reuse of data by multiple communities. Promote collaboration and “intersections” between a variety of communities including research libraries, scholarly societies, commercial partners, science, engineering and research domains, and evolving information technologies. Foster the training and development of a new workforce in data science. This could include, for example, NSF and the Institute of Museum and Library Services supporting a new initiative to train information scientists, library professionals, and scientists, both extant and future, to work knowledgeably on data preservation and stewardship projects as members of research teams. Support the development of usable and useful tools for: automated services and standards which facilitate understanding and manipulating data; data registration; reference tools to accommodate ongoing evolution of commonly used terms and concepts; automated metadata creation; and digital rights management. http://www.arl.org/info/events/ncr.html DRAFT

4. 1. NSF should develop a program to fund projects/case studies for data stewardship and preservation in science and engineering. Funded awards should involve collaborations between libraries, scientific/ research domains, extant technologies bases, and other partners. Multiple projects should be funded to experiment with different models. 2. NSF with other federal agencies such as the Institute of Museum and Library Services should support training initiatives to ensure that information and library professionals, and scientists (extant and future) can work more credibly and knowledgeably on data stewardship -- data curation, management, and preservation -- as members of research teams. http://www.arl.org/info/events/ncr.html DRAFT

5. 3. NSF should support the development of usable and useful tools, automated services (e.g. metadata creation), and standards which make it easier to understand and manipulate data. Incentives should be developed which encourage community use. 4. Economic and social science experts should be involved in developing economic models for sustainable data stewardship – research in these areas should ultimately generate models which could be tested in practice in a diversity of scientific/research domains over a reasonable period of time in multiple projects. http://www.arl.org/info/events/ncr.html DRAFT

6. 5. NSF should place greater emphasis on the inclusion of data management plans in the proposal submission process and in the proposal's review. NSF should be explicit that data management plans should identify the players involved in the custodial care of data for the whole of its life cycle and the means of support. 6. NSF should require data sharing policies be developed for those programs that currently do not require data sharing policies as a part of the proposal submission. In addition, NSF should require that all data sharing policies be more publicly accessible and uniform across NSF. Such policies should be enforced by appropriate mechanisms (e.g. the final report may not be accepted unless the awardee is compliant with stated data sharing plan) and include appropriate training initiatives to ensure that the research community can develop and implement these plans effectively. http://www.arl.org/info/events/ncr.html DRAFT