The Guardian: “…In the early days of the Data Rescue Project, there was a mad dash to save data from any agency they thought Trump or Doge would target next. Volunteers would download every dataset from the Centers for Disease Control and Prevention (CDC) or National Oceanic and Atmospheric Administration (Noaa) webpages as backup copies in case they ever went dark. But as the pace of deletions seemed to slow down last fall, the group had more time to consider ways to build longer-term data resiliency. The ultimate goal is not just to save data, but to make it accessible and discoverable to the broader public in the same way that libraries curate research and books, said Kellam. All archived information is now uploaded on to a searchable public repository for at-risk government data hosted by the University of Michigan, DataLumos. Volunteers also write metadata – short descriptions to help users understand what the data measures – for every item they preserve. “It’s one thing for one person just to download and copy some information, but that’s not really preserving it,” said Frank Donnelly, the head of geographic information systems and data services at a university library who began volunteering last winter. “You need to have a larger ecosystem where you’re saving data and the metadata that goes with it.” They are also trying to be more intentional about what they save. It’s hard to track how many datasets the US government produces, but there are at least 500,000 federal datasets listed on Data.gov, a site that makes federal data publicly available.

As of late April, Data Rescue Project volunteers have archived more than 3,000 items across hundreds of government departments, according to their public tracker. They have been downloaded from their public repository more than 18,900 times. Some items are more than just single datasets, but entire websites and databases, like a full archive of all Nasa webpages and the entire US Fish & Wildlife Service’s Feather Atlas, which contains high-resolution images of the feathers of North American birds. Volunteers have also managed to save several datasets before they were deleted. HIFLD Open was a collection of more than 400 maps showing critical infrastructure, like hospitals and highways, that emergency responders use during climate disasters. The Department of Homeland Security took it down last summer, but another two groups, the Public Environmental Data Partners and Fulton Ring used HIFLD data that volunteers archived to rebuild a public version of the deleted federal tool, which they are calling HIFLD Next….”

See also Tom’s Hardware: “Internet archival sites struggling to preserve the internet because of skyrocketing hard drive prices due to the AI boom — Wayback Machine and Wikimedia punished by stratospheric storage pricing and stricter anti-scraping measures blocking the wrong bots..”