As it’s nearly Halloween, let’s have a little horror story: You’re working in the lab late one night, when your eyes behold an eerie site…error message, a 404 error on a public data directory you had previously bookmarked that contained essential data for a project you just began, comparing Covid-19 outcomes under different regulatory schemes. How could the NIH website have a 404 error on such essential information? You’ve never seen anything like it. It may take you a bit longer to truly realise that all that data you were relying on for your project? data collected by hundreds of scientists and with millions of dollars of public funding? It’s gone. *Lighting flash, horror movie sting!*
Starting in January of this year, for some researchers this wasn’t just a scary data story. They went to formerly excellent databases or to essential retrieval tools, and found information deleted, moved, de-indexed, or otherwise rendered unfindable or unusable. At first many assumed it was a glitch, but as researchers and the librarians they consulted began talking to each other, it became clear that a massive purge of US public data was underway. The most affected data concerned topics like: climate change, public health (especially HIV/AIDS, Covid-19, social determinants of health, and vaccination), gender identity, and information on the contributions of those from minoritised groups.
In addition to the deletion of public data, public-facing educational materials, such as those in educational webpages, public plaques, and displays in national museums, were also being deleted, minimised, or de-indexed. Research papers and commissioned reports by public agencies and their employees were also being deleted or de-indexed. Essentially, knowledge generated by a vast network of scholars over decades, was being erased or supressed. It was a book burning for the digital age.
As scholars and librarians noticed this was being done, they also got to work on a rescue plan for this public knowledge, at first on an ad-hoc basis, and later organised into the Data Rescue Project. Organisations and individuals with the capacity to do so began backing up public data, targeting the most at-risk datasets first. They launched subprojects cataloguing government websites and even physical signage at national parks and monuments, and connected with other organisations (such as the Internet Archive) already involved in these efforts. If you would like to support their efforts, there are a number of things you can do:
- Get involved in a rescue project: check the list of current projects to find something you can help with, either by archiving data or feeding back public data you had previously downloaded
- Share information about the project among your networks: Academic research is both a massive global effort and an interlocking series of small communities. Within your networks are people with a lot of capabilities to help
- Share your data story: If you’ve used US public data in your work, share your story here. This helps campaigners advocate for robust, secure, data infrastructure that is insulated from political machinations by demonstrating it’s value, importance, and global reach. In possible, donate to the project
You may be thinking ‘what’s happening in the US is terrible of course, but I don’t actually use US public data and I don’t live or work there. Why should I care?’ Well, there’s really two answers to that. First, while the work of the Data Rescue Project and their allies is inspirational, there is no doubt that some knowledge has nonetheless been lost. Maybe not any information you would directly use, but when the scholarly record starts getting holes in it, knowledge is lost and you end up having to re-tread work already done. My colleague, Franchesca Soldati wrote more about the consequences of this in her recent blog on disappearing journals .
However, the second reason to care is more basic. It isn’t you today. It isn’t here today. Did American scholars think it would be them ten years ago? This ties back to this year’s International Open Access Week theme: who owns our knowledge? If data you created can be deleted forever by a third party you don’t own it. It’s very normal for scholars to rely on government or institutional repositories to hold the knowledge they generate. These are considered safe houses for such work, but should we so uncritically rely on these? Scholars generating or using very large datasets, such as climate scientists or astronomers, probably do not have access to the personal computing power to store and retrieve the data they use, and so it may be difficult for some fields to imagine an alternative anyway. Distributed community-run data storage models, such as the ones employed by the Data Rescue Project itself, provide one possible practical solution, but these require significant commitment and community buy-in. We all need to work together to preserve the knowledge we create for future generations.
For now, scholars are left with hard questions: how can I be sure I’ll be able to access the data I need in the long-term? Is my data held somewhere secure, or is it vulnerable to political machinations? What should I do to protect the knowledge I use and the knowledge I value the most? All of these questions come back to the central issue of ownership and control. When you’re beginning a project, especially a funded or work-for-hire project, consider the issue of ownership and rights. Email us at openresearch@abdn.ac.uk if you’d like to talk about how these issues might impact your own work, and don’t forget the Digital Research Team if you need help setting up secure storage for your data, or drafting a data management plan.
References and Further Reading
- ‘Press and Presentations’, Data Rescue Project: a full list of media about as well as presentations by the Data Rescue Project, available here
- Baraniuk, Chris. 2025, April 23. ‘Inside the desperate rush to save decades of US scientific data from deletion’. BBC Earth. https://www.bbc.co.uk/future/article/20250422-usa-scientists-race-to-save-climate-data-before-its-deleted-by-the-trump-administration
- Santarsiero, Rachel. 2025, September 30. ‘White House Ramps Up Effort to Remove and Alter Information on Climate Change Orgs Working to Rescue Climate Data Before It Disappears’. National Security Archive, George Washington University. https://nsarchive.gwu.edu/briefing-book/climate-change-transparency-project/2025-09-30/disappearing-data-part-ii-distorted
- Lucas, Julian. 2025, March 14. ‘The Data Hoarders Resisting Trump’s Purge’. The New Yorker. https://www.newyorker.com/news/the-lede/the-data-hoarders-resisting-trumps-purge
