It was about 8:30 pm on December 30, and Marjorie Pollack, a physician and epidemiologist, was working in her home office in Cobble Hill, Brooklyn.

Her email pinged. In her inbox was a note from a frequent and reliable contributor to ProMED, an email list of disease alerts for which Pollack serves as deputy editor. The contributor, who speaks and reads Chinese, wanted her to know about a new post that was getting some attention on Weibo, a within-China social media network. The post said that a few hours earlier, the Wuhan Municipal Health Committee had issued “an urgent notice on the treatment of pneumonia of unknown cause.”

Read all of our coronavirus coverage here.

It was the sort of note that ProMED gets every day. The low-tech list and site relies on a worldwide network of readers who funnel notices and tips–social media chatter, health department announcements, stories from small media outlets–to its roughly 50 very part-time employees. Almost all of them are medical or public health or research professionals in real life, and they share an ethos of strict standards of evidence, never publishing anything for which they cannot find confirmation through at least one other source.

So Pollack swung into action. She alerted the rest of the network. Within a few hours, they found the confirmation their standards call for, a story on a Chinese business news site that confirmed the existence of the Wuhan authorities’ notice and included more details. They wrote up a post, vetted it through ProMED’s structure of copy editors and moderators, and hit the button to publish it at one minute before midnight on December 30.

All the time, even hammering on her keyboard, Pollack could not shake a back-of-the-neck chill: “I said to myself, until proven otherwise, this is SARS revisited.”

She was right: The Wuhan health department’s notice was the first warning of Covid-19, which, like SARS in 2003, is caused by a novel coronavirus. And she was in a position to know–because ProMED also published the first bulletin, in February 2003, that alerted the world outside of China to the existence of the SARS epidemic. It also published the first notice of MERS, another transnational coronavirus epidemic, when it arose in Saudi Arabia in September 2012, and of numerous additional outbreaks of Zika, Nipah, Ebola, and other terrifying diseases.

Consistently, this human-curated, barely funded, open-access list–avidly followed in the realm of public health and almost unknown outside it–has been an alarm bell for the world.

ProMED (it stands for “Program for Monitoring Emerging Diseases”) began in 1994, the personal project of John Payne Woodall, an entomologist and virologist who died in 2016. Its cofounders were Stephen Morse, a professor of epidemiology at Columbia University’s Mailman School of Public Health, and Barbara Hatch Rosenberg, a biological weapons expert and former professor of microbiology at SUNY Purchase.

Woodall, universally known as Jack, had an unconventional career. He was born in China to a British family who arrived as missionaries, stayed to run important schools, and were interned in a Japanese prisoner camp from 1941 to 1945. (He once told an interviewer that he dated his interest in insects to the bugs he saw while wandering around the camp’s weedy grounds.) He worked on several continents for almost every organization that was important in the post-World War II push to improve global public health, from the Centers for Disease Control and Prevention to the World Health Organization to the Rockefeller Foundation.

Maybe because of that much-traveled background, he believed passionately in the usefulness of everyday people’s reports of events–what the WHO now calls “epidemic intelligence from open sources”–to augment the official data gathered by governments and nongovernmental entities. He also believed in the power of the internet to get that information to people without governments interfering. He founded ProMED at about the same moment that internet use was opening up to everyone, and the site and its email format retain an endearingly clunky text-forward design.