Part of the Deep Dive: Data Governance Webinar Series
The Data Rescue Project (DRP) emerged as a collaborative effort to protect and preserve public datasets that are at risk of being lost, altered, or rendered inaccessible. Our work is based on the belief that public data belongs to the people and that preserving it is a civic responsibility. In this presentation, we reflect on how current threats to public data access expose weaknesses in the systems we rely on to ensure long-term access and stewardship.
Our approach is not only technical; it is deeply community-driven. The Data Rescue Project brings together librarians, archivists, technologists, data stewards, and researchers from across disciplines. We recognize that no single organization or agency can solve the systemic problems of public data infrastructure alone. In moments of urgency, grassroots efforts can illuminate what larger institutions often miss: that data preservation is not only about files and formats, but about relationships, trust, and collective responsibility.
This talk will highlight coordination among members of the DRP and our partners, including efforts to amplify active preservation work being done by groups like the Environmental Data and Governance Initiative (EDGI) and data repositories such as ICPSR’s DataLumos. The DRP uses data vulnerability and access as a lens to ask bigger questions: Who is responsible for public data? What does true accessibility look like? And how can we build systems that are not only open but resilient in practice?
Video transcript
Presenters: Lynda Kellam & Mikala Narlock, Co-organizers of the Data Rescue Project
Background and Context
The US federal statistical system is decentralized and complex, consisting of 13 principal statistical agencies including the Census Bureau, Bureau of Economic Analysis, and National Center for Education Statistics. These agencies collect data and produce statistics about the economy, society, and natural environment. American government information is in the public domain except under limited circumstances, making it widely trusted and authoritative globally.
However, over the past eight months, there have been unprecedented attacks on US federal government information and data sources—some taken down, altered, or made inaccessible due to lack of data stewardship within agencies.
Origins of the Data Rescue Project
The Data Rescue Project began in February 2025 as a partnership between three data professional and librarian organizations: IASSIST and the Data Curation Network. Building on previous efforts like the End of Term Web Archive (established 2008) and EDGI (Environmental Data and Governance Initiative, created in 2017), the project initially aimed to help patrons find data and amplify safeguarding efforts.
The project differentiated itself by focusing on social data at high risk, including data on marginalized communities and sexual orientation/gender identity—areas not addressed by other groups at the time.
Core Goals:
- Coordinate to avoid duplication
- Communicate actions being taken
- Rescue data through thoughtful, sustainable workflows
Key Accomplishments (February-September 2025)
- Rescued thousands of data sets across 80+ at-risk government offices
- Mobilized 500+ volunteers capturing hundreds of terabytes of mirrored data
- Created public-facing website with resources, tools, checklists, and library guides
- Developed shared Google spreadsheet for coordinated rescue efforts with nomination system
- Uploaded nearly 800 data sets to Data Lumos (ICPSR repository at University of Michigan)
- Generated almost 8,000 downloads from Data Lumos (February-September 16)
- Built Data Rescue Portal cataloging nearly 1,250 rescued data sets with location information
- Created Data Rescue Event Toolkit with free templates for institutions
- Save Our Signs initiative: Captured almost 8,000 photos from 300+ national park sites (as of September 15) in response to executive order targeting signage about slavery and civil rights
Key Partners
- EDGI & Public Environmental Data Partners (PEDP): Coalition focused on environmental data and recreation of removed tools
- Safeguarding Research and Culture: German organization scraping websites and storing as BitTorrent
- America’s Essential Data: Collecting data use stories
- Population Reference Bureau: Created Federal Data Forum online community
- Tracking Government Information: University of Minnesota effort tracking removed websites
- SUCHO (Saving Ukrainian Cultural Heritage Online): Provided infrastructure guidance based on 2022 war response
Critical Challenges Ahead
- Restricted access data: What happens to data requiring application processes when those processes become unavailable?
- Data integrity and collection: Impact of dismantled programs or inadequate staffing on ongoing data collection
- Long-term infrastructure: Supporting current efforts and preparing for future crises
- Community sustainability: Maintaining diverse, cross-professional, international network
How to Help
- Share resources with professional and personal networks
- Join rescue efforts and help identify at-risk data
- Share stories about why public data matters to everyone
- Collaborate on data rescue or storytelling events
Core Questions for the Future
- Who is responsible for public data? Everyone must educate themselves and others about its importance
- What does true accessibility look like? Must avoid commercialization and strengthen nonprofit communities
- How can we build resilient, open systems? Governance must be cross-community; silofication and territoriality harm public data efforts
Central Principle: Public data are the people’s data and must remain a public good.