A Framework for Avoiding the Open Research Data Dump (AFFORD)

AFFORD is a 2-year project (2023-2024) funded by swissuniversities and hosted by the University of Zurich through the Interface Group  and the Center for Reproducible Science. Its principal objective is to establish a support framework that lowers the barriers to publishing data and other forms of research output according to the FAIR principles.

FAIR Open Research Data

Requiring open access to research data (ORD) is a necessary step towards credible, reliable, and reproducible research, but it is not sufficient. For instance, if experimental data are made available in a format that only a small number of labs can read, or if files are published without proper documentation to provide context not sufficiently captured by metadata, then open access to these data is practically useless. However, publishing FAIR ORD is highly resource-intensive, especially if large amounts of data are produced in a project. Unless adequate support is provided, the requirement for ORD is bound to produce data dumps where data are freely accessible, but of limited utility.

Developing a Sustainable Support Framework

AFFORD aims to establish a sustainable support framework that lowers the barriers to publishing FAIR ORD by bundling know-how, workflows, and tools under the umbrella of one organizational entity. It uses the Sinergia project Fluid Dynamics of the central nervous system as a reference project and accompanies it in the full cycle of ORD generation from experiment planning to publishing. This project-based, data-driven approach will help the framework reach a sufficient level of maturity before it is made available to all university researchers.

Project management: Vartan Kurtcuoglu (The Interface Group), Leonhard Held (Center for Reproducible Science)

To produce FAIR and fully reusable ORD that will improve research in general, substantial effort is required. Funding bodies such as the SNSF could provide human resources dedicated to ORD for each supported project, but this would be highly inefficient. Alternatively, universities could concentrate expertise and resources in dedicated entities to support scientists in producing usable ORD of high quality and utility.

Challenges and Goals

The implementation of such a support model is what we propose to investigate with AFFORD. Doing so, we will address the following three gaps:

1. If scientists are to produce FAIR and fully reusable ORD, they must invest resources, but they usually do not have sufficient data on how much is required.

   Goal: AFFORD aims at providing field-tested and refined guidelines for the production of FAIR and fully reusable ORD in research projects, including resource estimates and processes for resource monitoring.

2. Ideally, ORD management starts at the planning stage of a research project but, in practice, it is often initiated posteriori. Especially with the ambitious goal of FAIR and fully reusable ORD, the process can be much more costly and difficult when initiated after output has already been produced.

   Goal: AFFORD aims to estimate resources in both scenarios - sequential and concurrent ORD management - and will hence allow a comparison, something that is currently not available.

3. To faithfully satisfy ORD mandates, the efficiency of ORD management must be increased.

   Goal: We propose that this be realized through centralized ORD support. The resource data collected will allow us to propose efficient designs that institutional decision makers can use for the implementation of sustainable ORD support frameworks.

AFFORD will provide field-tested guidelines for the production of FAIR ORD, resource estimates for concurrent and retrospective ORD management and processes for resource monitoring. These will need to be adapted to a certain degree when they are applied to research projects in different fields. However, the field diversity within the reference project will ensure that guidelines and resource estimates are not excessively insular and that they can serve as more than mere starting points.

• Up to now, the developed guidelines, tutorials and templates are available on the AFFORD website.
• Recommendations for the processes, resources and structures needed to provide sustainable ORD support will be published as a white paper.

We further propose to build a central ORD index from where all of a research project’s output can be accessed (single point of entry to increase findability and accessibility according to FAIR principles). This ORD index should allow for a rapid overview of a project’s ORD, with access or links to the data and all relevant information and documentation. It is not intended to serve as a data repository and should thus be simpler to host and maintain.

• Up to now, we have implemented an internal ORD index for the reference project. A template of such ORD index  is publicly available.
• The finalized structure including DOI for systematic referencing will be released in Summer 2024.

We will consolidate the experience gained in AFFORD to make a concrete proposition for sustainable FAIR ORD support, including proposition of the organizational entity or entities that may be suited for this purpose. Its impact will largely depend on whether the framework will be adopted by institutions. We will work towards this end through the Swiss Reproducibility Network and locally advocate for its implementation at the University of Zurich.

ORD mandates will, by themselves, not produce more credible, reliable, and efficient research. When faced with such mandates in the absence of a favorable cost-benefit structure, most scientists will only make minimum effort to fulfill them, if they do it at all. This is the core challenge addressed by AFFORD, aiming to lower the barriers to publishing data and other forms of research output in an accessible and reusable form.

AFFORD aims to address all Swiss researchers who aspire to produce FAIR and fully reusable ORD in their scientific projects. The main outcomes of the project will be:

1. guidelines for ORD management including resource estimates,
2. a concept for a sustainable ORD research support framework that can be adapted and realized at any (Swiss) university,
3. the proposition of a specific implementation of the framework tailored to the University of Zurich.

The guidelines will be made openly available and will remain of value for the foreseeable future. The ORD research support framework concept will also be made openly available. The project-based, data-driven approach adopted in AFFORD will improve the adoption of the proposed support framework by providing reliable resource requirement estimates to decision makers at the university level. It will further accelerate acceptance by the research community by ensuring that the support framework has reached a sufficient level of maturity before it is made available to all university researchers.