The U.S. government spent taxpayer money to build a climate data website, ran it for a full 15 years, and then shut it down with its own hands.
But what the shutterers didn’t expect — because open data is legally owned by the entire public — a group of laid-off workers and 2,500 ordinary people willing to pay rebuilt it within a year.
It sounds like an inspirational story about data defeating power. But the fiercest debate in the community points precisely at what the inspirational narrative ignores: raw data sitting there is, to ordinary people, essentially nonexistent. What’s truly valuable is that layer of experts who got fired.
How a 15-Year-Old Public Website Was Shut Down Overnight
In June 2025, the Trump administration shut down Climate.gov as part of sweeping cuts to the U.S. National Oceanic and Atmospheric Administration (NOAA).
The site launched in 2010 as the federal government’s most important climate-education platform for the public. It translated complex satellite remote-sensing data, atmospheric-chemistry observations, and ocean-temperature records into charts, articles, and teaching tools ordinary people could understand. Farmers used it to judge planting windows; teachers used it for lesson prep; journalists used it to check climate-change facts; coastal-city planners relied on its sea-level-rise data for flood-defense budgets.
Before it was killed, Climate.gov drew nearly 1 million visitors a month.
The shutdown went far beyond “temporarily offline for maintenance.” The entire 10-person team was laid off, and the site was redirected to a stripped-down page with only scattered fragments of content. NOAA itself lost more than a fifth of its staff in the reorganization — some weather-forecast offices were so short-handed they couldn’t even spare anyone to launch weather balloons, the daily starting point for weather forecasts.
Next, the Fifth National Climate Assessment — the most comprehensive analysis of climate change the U.S. government has ever produced — vanished from the official website. The report had taken four years and hundreds of scientists to write.
If the data hadn’t been openly licensed, the story would have ended there: government deletes, data disappears.
Why It Couldn’t Be Killed — Open Data Is a Legal Firewall
The U.S. has a rule: data produced with taxpayer money belongs to the public domain and carries no copyright restriction. Anyone may legally copy, distribute, and use that data.
What does that mean? The government can shut down the website, but it can’t shut down copies of the data.
Rebecca Lindsey was the former project lead for Climate.gov. After being laid off, she did the most direct thing possible: she recruited her sister Mary Lindsey and former colleague Anna Eshelman, and the three formed a core team that began collecting historical backups of climate datasets from public sources.
Then it snowballed.
Around 80 volunteers joined — former NOAA scientists, university researchers, science communicators, programmers. No office, no government budget — just GitHub collaboration, mailing lists, and Zoom calls. More than 2,500 people donated, totaling over $320,000, covering roughly a third of the project’s startup costs. The rest came from an anonymous donor.
On June 24, 2026, Climate.us went live. Its homepage is a real-time dashboard showing CO₂ concentration, Arctic sea-ice extent, global surface temperature, and ocean heat content — nearly all of the most-visited Climate.gov indicators were back. Teaching resources, regional climate maps, and El Niño explainers were restored too.
This happened not through a technological miracle, nor through anyone’s heroism. It happened because the data was designed from the start so that “the government’s left hand can’t shut down the right hand’s copies.”
Raw Data vs. Usable Information — Separated by an Entire Layer of Fired Experts
Up to here, the story sounds pretty complete. But on Hacker News, the argument went in a completely different direction.
One user raised a sharp question: “Climate.gov was never the only place climate data lived. There are dozens of petabytes of climate data scattered across the place — NOAA, NASA, university servers, everywhere. Want data? It’s all over.”
Another user replied — quoted and upvoted repeatedly: “I, myself, don’t want the data. I want a service built on reliable data and expert validation.”
That sentence hits the core contradiction of the whole affair. Hand an ordinary person a pile of raw observations — satellite imagery, temperature readings, CO₂ curves — and they can’t read it. They need someone to tell them: what does this number mean? Is it anomalous on a 10-year timescale? Is this trend real or just noise within the margin of error?
That was Climate.gov’s original core function — what its 10 full-time staff did every day. Translate. Validate. De-noise. Explain the science in language the public could understand.
80 volunteers can rebuild the website framework, restore datasets from historical backups, and put a PayPal link on the donation page. But how many of them can, long-term, full-time, and in an organized way, keep explaining each day’s new data?
Climate.us currently runs on donations. Its founders have said publicly that this isn’t a sustainable long-term model — because maintaining a public data service is the job of taxes, not crowdfunding.
Who’s the “Villain”? Two Levels of Conflict
This article has two layers of conflict, not one.
The first is obvious: government shutdown vs. the public’s right to know. A public resource built over 15 years with taxpayer money, deleted by executive order in one click. That’s the brute exercise of power — but precisely because the data was designed from the start around the “public domain” principle, the brute force was offset by the law. You kill the homepage; I rebuild one.
The second layer is subtler but more important: raw data vs. usable information. Climate data was never truly “hidden” — observations of the atmosphere, oceans, and ice sheets are scattered across institutions worldwide. For professional researchers, Climate.gov was just one entrance among many. But for everyone else — farmers, teachers, journalists, small-town planners — Climate.gov was nearly the only entrance. What the shutdown destroyed was the layer that turns data from “machine-readable” into “human-usable.” The data itself survives, but the bridge to it is broken.
To use an analogy from the HN discussion: you can download Wikipedia’s database backup to a hard drive, but that doesn’t mean you can use Wikipedia. You still need indexing, search, formatting, community governance — and a server that keeps running.
Climate.us has rebuilt the latter’s framework, but whether it can sustain that “translation and validation” layer long-term is far from clear.
This Is Not a “Community Saves the World” Story
Writing this, I had a strong feeling: this story is easily told as a “civil society triumphs over the bureaucracy” narrative. But after reading the original and the 140-plus HN comments, I lean toward seeing it instead as a warning about the fragility of public infrastructure.
If U.S. law hadn’t declared government data part of the public domain, this story would have no second act. If NOAA’s layoffs had gone any deeper, and datasets had stopped updating even at the raw-observation level, the rebuild would have been just a historical snapshot. If those 2,500 donors hadn’t opened their wallets, Climate.us would have been just an unlaunched domain.
Every one of those “ifs” is a governance choice, not a technical problem.
Climate data is a public good, like weather forecasts, water-quality monitoring, and earthquake early-warning. Its value peaks at the moment every dollar is cashed out as public interest — not at the moment it’s shut down and then picked back up by well-meaning volunteers. The latter deserves praise, but the former deserves to be fought for.
Reference Links
- Werd I/O: Ben Werdmuller’s commentary analyzing how, after Climate.gov’s shutdown, open data became a firewall against administrative-order destruction
- The 19th: an in-depth report by Jenae Barnes documenting in detail how Rebecca Lindsey’s team rebuilt the climate data platform after being laid off
- My Modern Met: a timeline of Climate.gov from launch to shutdown to rebuild, with background on NOAA’s mass layoffs
- Climate.us: the rebuilt independent climate data platform, maintained by former NOAA scientists and run entirely on donations
- HN discussion: Hacker News discussion of the event, including a deep debate on “raw data vs. usable-information service”
- BizTech Weekly: a technical-architecture analysis of how Climate.us achieves distributed data management, data-provenance verification, and open-source collaboration