Funding Sustainability and Domain Repositories This overview presentation will outline several primary issues regarding funding sustainability for domain repositories, including existing funding models, known inefficiencies, and possible solutions. This will be followed by big picture questions that will lead into a larger group discussion. As a 2013 “Call for Change from an Interdisciplinary Working Group of Domain Repositories” noted, “There may not be one solution to the problem — repositories may very well need different funding models across domain and repository type.” https://www.icsu-wds.org/files/Domain_Repositories_Call_to_Action_16_Sept_2013.pdf Jared Lyle (lyle@umich.edu) ICPSR, University of Michigan WDS Members Meeting, Denver, Colorado 11 September 2016

Sources Ember, Carol and Robert Hanisch, Sustaining Domain Repositories for Digital Data: A White Paper, December 11, 2013, http://datacommunity.icpsr.umich.edu/sites/default/files/WhitePaper_ICPSR_SD RDD_121113.pdf RDA-WDS Cost Recovery Interest Group: Ingrid Dillo, Simon Hodson, Anita de Waard, Income Streams for Data Repositories https://docs.google.com/document/d/1PzVXdTH7GhcvHYWd5jGfomkqkKPSWmki yHy5mqEURBE/edit# Cost Recovery for Data Centres Working Group Case Statement: https://rd- alliance.org/sites/default/files/case_statement/RDA_WDS_IG_Publishing_Costs.p df Hodson, Simon, Sustainable Business Models for Data Repositories, RDA Plenary, Tokyo, March 3, 2016, https://rd- alliance.org/sites/default/files/attachment/S.%20Hodson%20Business%20Model s%20Presentation.pdf

OECD Global Science Forum project “OECD Global Science Forum project will run for 15 months or so from April 2016 and will extend the survey of repositories. We will hold two workshops, one to look at innovative income streams and options for cost restraint, the other to test more thoroughly with stakeholders, willingness to pay (for what and how) and the viability of various business models for different levels of curation, data services etc. The output will be positive and concrete OECD policy recommendations for business models that will promote sustainable data infrastructures.” https://rd-alliance.org/group/domain-repositories-interest-group-rdawds-publishing-data-cost-recovery-data-centres-ig/post-3

https://www.flickr.com/photos/athrasher/2823255013

Repository Functions Curation – Adding value to the data Preservation – Insuring the data are available for the long-term Interoperability – Enabling comparison across data collections Citation – Linking data to publications Protecting Confidentiality & IP Community Engagement – Connecting with the domain’s users Innovation – Developing new services

Existing Funding Models Structural/Federal Funding Research and Development Projects Institutional Support Access Fees (Membership) Submission Fees Contract from depositing institution Structural/Federal Funding – “The primary weakness and threat posed by this form of funding is its inflexibility and that it can constitute a single point of failure.” Research and Development Projects – “Research grant competitions subject data repositories to review criteria and time horizons that are inconsistent with their core function as long-lived memory institutions.” Institutional Support – “Contrary to expectations, based on initial discussions in the working group, we found that most data repositories did not receive substantial funding, directly or in kind, from their host institutions. Indeed, a number complained earnestly, even bitterly, that far from subsidising the repositories activities, the host institution sliced off contributions from the central grant / structural funding or research projects.” Membership – “Membership models do not provide public access to data, and they favor researchers at institutions with more resources.” Submission Fees – “On the other hand there is a danger of large administrative overheads and depositors might rush to the cheapest option. The model requires a very clear policy framework to be effective and to avoid such risks.”

Possible New Funding Models Value-Added Services Commercial Services User Fees Overhead Infrastructure Commercial services: Repositories may build services to make money from archived data. This could have limited use for academic users, or only be available to corporate consumers. User fees: Some kinds of data are costly to distribute, and end users may be charged for costs associated with data access. For example, when datasets are extremely large, it is more efficient and economical to provide computing on central resources. Similarly, there are costs to providing access to data in which confidential information from research subjects must be protected. Overhead: Universities allocate a percentage of all research grants (possibly from indirect costs) towards a fund for data archiving. Universities then make decisions about how data are selected for archiving and which repositories are used. Infrastructure: Funding agencies pay for archives directly as a necessary aspect of research infrastructure. The funding model is structured for long-term investment, rather than being tied to three-year grant cycles. While this may appear similar to federal funding for special projects, a percentage of federal research funding would be set aside for digital data archiving and preservation in all disciplines.

Recommendations

Commit to sustaining institutions that assure the long-term preservation and viability of research data. “We estimate that successful domain repositories can be operated at funding levels of less than 5% of the total research budget”

Promote cooperation among funding agencies, universities, domain repositories, journals and other stakeholders. “Funding agencies should re-direct resources to support data curation and archiving. Universities should provide facilities for researchers and assure compliance with archiving requirements. Domain repositories should establish partnerships with universities and institutional repositories to provide facilities and expertise across all disciplines.”

Incentivize PIs to archive data. “PIs should not be faced with a tradeoff between accomplishing their scientific objectives and sharing their data. For example, data curation for archiving should be funded by mechanisms that are outside the funding allocated for the principle aims of a research project.”

Diversify funding

For discussion… What role can the ICSU World Data System play? How can we move beyond policy recommendations? Should we allow natural ebb and flow of business?

Thank you! lyle@umich.edu