Open access

The ability to view, reproduce, and build upon other works is a core tenet of the scientific process. Redivis provides tools for data storage, access, analysis, documentation, and publication that can help researchers and institutions implement public access and open science practices.

Not all research can be made fully public. Redivis supports a responsible approach to data sharing by allowing data owners to make datasets and their metadata as accessible as possible while applying appropriate protections to sensitive or restricted information.

Background

In August 2022, the United States Office of Science and Technology Policy (OSTP) released a memorandum on public access to federally funded research (known as “Nelson Memo”) which outlined new requirements affecting both faculty and students who conduct research using federal funding:

1. Make publications and their supporting data resulting from federally funded research publicly accessible without an embargo on their free and public release.
2. Enact transparent procedures that ensure scientific and research integrity is maintained in public access policies.
3. Ensure equitable delivery of federally funded research results and data.

1. Public accessibility

Redivis helps data owners make research data as broadly accessible as possible while supporting appropriate protections for sensitive or restricted information. Public datasets and workflows can be explored without an account, while researchers can create a free account when they need to request restricted access or work with data.

Open science does not require all research data to be released without restriction. Instead, data should be made as open as possible and as restricted as necessary. Redivis supports this approach by allowing data owners to:

• Make data and documentation public by providing public access to dataset descriptions, metadata, documentation, sample data, and complete data when appropriate.
• Keep restricted data discoverable by making a dataset's overview and metadata publicly available while restricting access to the underlying data.
• Provide responsible access by allowing data owners to limit access to approved researchers through setting application requirements.

Redivis provides the technical tools for configuring and administering these access models. Institutions, dataset owners, and researchers remain responsible for determining appropriate conditions based on applicable laws, consent, licenses, data-use agreements, ethical obligations, and institutional policies.

2. Transparent procedures

Redivis systems are built to automatically capture and document any work done on datasets and analyses in standardized formats, while giving researchers and administrators the ability to supplement or override these when necessary.

• Datasets are automatically versioned. All changes to variables and rows are recorded and made available to administrators and data viewers.
• Redivis offers a no-code interface to build transform queries that compiles to SQL code. This code is available to the analyst and any viewers.
• Workflows are self-documenting. Every step taken in an analysis workflow is recorded sequentially in a visual format that is easy to follow.
• Workflows are version controlled. Every step is versioned and time-stamped, allowing users to revert to a previous iteration, or a viewer to understand how queries might have changed over time.
• Workflows can be forked, allowing researchers to build on, adapt, and extend prior research analyses while maintaining a connection to the original work.

3. Equitable delivery

Redivis is free and accessible to any researcher, reviewer, or casual data browser. By taking multiple complex data storage and analysis systems and centralizing them with clear UI, Redivis makes full workflows easier to understand without specific technical knowledge.

• Redivis uses transparent, open-source formats, allowing data, analyses, and workflows to be exported and used within other systems.
• APIs make data and workflows available to researchers and administrators using their own systems.
• No-code interfaces allow researchers at any skill level to build and understand research steps.
• Common, open-source coding languages (SQL, Python, R) form the foundation of tools supported in analysis workflows.
• All researchers receive a free data storage allotment and can explore and query data at no cost.