Scientists discover a new planet that might be able to support life. A drug that targets certain cancer receptors. Evidence of a previously unknown branch of human ancestors. Extreme weather. Dinosaurs had feathers? Who the hell knew ravens were so clever! And you saw the story on Facebook, on Twitter, in an email from your mom. Then you shared it to your coworker via Slack, your boss over lunch, from a barstool during your date that evening.
Science news gets around. But a lot of it---you probably didn't know this---comes from the same place. A website called EurekAlert gives journalists access to the latest studies before publication, before those studies are revealed to the general public. Launched 20 years ago this week, EurekAlert has tracked, and in some ways shaped, the way places like WIRED cover science in the digital era.
Yes, of course the Internet was going to change science journalism---the same way it was destined to change all journalism. But things could have been very different. EurekAlert gathered much of the latest breaking scientific research in one easily accessible place.
You probably know the basic process of science: Researcher asks a question, forms a hypothesis, tests the hypothesis (again and again and again and again), gets results, submits to journal—where peers review—and if the data is complete and the premise is sound, the journal agrees to publish.
Science happens at universities, government institutions, and private labs. All of those places have some interest in publicizing the cool stuff they do. So those places hire people—public information officers—to alert the public of each new, notable finding. (OK, maybe some aren't so notable, but whatever.) And the route for that notification is often via journalists.
And, much like the way journalists compete with one another for scoops, journals compete with one another for the attention of journalists to publicize their research. After all, Science, Nature, JAMA, and so on are interested in promoting their brands so they can attract more smart, impactful research. As someone smart once said, science is a contact sport.
So how did EurekAlert become the one clearinghouse to rule them all?
In the sepia-tinted epoch before digital communication, journalists received mountains of letters and faxes. "One of the major tools of the trade was a letter opener," says Richard Harris, who is currently on book leave from his job covering science for NPR. And each journal would send their notices separately. Sometimes journals would publish without any expository press releases---quelle horreur!—and reporters would be left to their own devices to sift through the articles and figure out what was newsworthy.
Now, if you think that this is either easy or, you know, their job, consider that one of the landmark papers proving that ulcers were caused by bacteria and not excess stomach acid---a germinal piece of research with huge implications---was published in 1994 in the New England Journal of Medicine under the title "Helicobacter pylori infection and gastric lymphoma." Doesn't exactly scream front page news. If that paper hadn't come with a news release explaining its significance—overturning decades of gastrointestinal dogma—only the most dedicated biology beat writers would have picked up the story. That cryptically-titled paper would eventually help its authors earn a Nobel Prize.
In 1995, a group of an university PIOs had an idea for a centralized, online hub for science press releases. They soon realized that developing such a platform on their own would be prohibitively expensive without at least a dozen different shareholders chipping in. "After a couple of meetings it became clear that it was unlikely that we would be able to convince such a critical mass to put up the money to proceed," writes David Salisbury in an email. So instead a member of the group named Dennis Meredith approached Nan Broadbent, then communications director for the American Association for the Advancement of Science, who agreed to give the idea a go1. But the site would not only include press releases from studies published in the AAAS publication Science. It would have all new research from every journal. Which might seem trite on today's hyper-aggregated web. Remember, this was an era when anime nerds on Geocities were still struggling to organize their competing Ranma 1/2 fansites into webrings*.
Remember, scientific journals are way more competitive than anime nerds. And AAAS publishes Science, one of the most dominant publications on the planet. "I had to convince the others that we were not going to steal their stories," says Broadbent, who coordinated the collaborative EurekAlert effort. She did so by organizing the service as an independent arm within AAAS.
"You have to credit Nan for her persistence and creativity," says Ginger Pinholster, the AAAS' current communications director. "She brought together all the right players, and organized a meeting in spring 1996 to set forth editorial and access policies." Such as: EurekAlert employees would not get sneak previews of Science's upcoming stories, and they would not share articles from other journals with their coworkers in the AAAS. Every reporter would have their credentials verified in order to get access. For accountability, EurekAlert would have an advisory board independent of the AAAS.
Broadbent applied for grants to pay for it. Monsanto donated, so did Genentech. One of the most important came from Sun Microsystems, which connected her with developers at Stanford University. "The embargoed releases would come to us from the journals and universities via web submission or FTP," says Tim Torgenrud, a Stanford technology lead who was in charge of building the EurekAlert platform. Each file was put into a locked folder, and timestamped for when its embargo lifted. When that time came, the file would automatically move itself from the locked down folder to another one, from which reporters with the proper credentials could access it via the EurekAlert website.
The site launched on May 20, 1996. "EurekAlert was like a phase change," says Charles Petit, a freelance journalist who was working at the San Francisco Chronicle when EurekAlert came out. Before then, journalists weren't just acquiring mountains of paper spam. They had to hustle for access to studies. Reporters at regional outlets, with less of a travel budget, were at a disadvantage to large national papers and magazines that could send staff to meetings to develop relationships with individual scientists or university PIOs. "It used to be a pain in the ass, that’s for sure, to try to keep track of all the journals that EurekAlert offers," says David Perlman, who has been reporting for the Chronicle since the 1940s.
The site also helped journalists work on stories besides those being promoted. "It was the first online forum where I could easily access and search for articles on a topic," says Deborah Blum, director of the Knight Science Journalism Program at MIT. Blum was a Pulitzer Prize-winning investigative reporter at the Sacramento Bee when EurekAlert debuted in 1996. And she was lucky. Many news outlets didn't even have Internet at the time. Some had a single console, and reporters had to queue up for access. "You’re talking about a period of digital dark ages," Blum says. "If you wanted to know what had been published on a topic, you had to go down to university library and trawl the information desk for literal stacks of articles."
Broadbent's group came up with a business model, too. Journals and institutions sustain EurekAlert through annual fees to post their studies. (Reporters get free access.) The price is a bargain compared to the way universities once toiled to connect with journalists. Jeff Grabmeier, longtime PIO for Ohio State University, remembers how his boss had spent years developing a massive contact list of reporters, and what kinds of research they were all interested in covering. "He protected that list like it was gold," Grabmeier says. EurekAlert made that golden list obsolete.
But while EurekAlert democratized journalists' access to papers, and PIOs' access to journalists, those who had the resources to develop their own connections—like Grabmeier's boss, or reporters at big, national outlets—suddenly found themselves competing with, well, everyone. "EurekAlert is kind of like a giant virtual press conference, in that it pulls everybody to the same spot," says Cristine Russell, a freelance science writer since the 1970s.
That centralization, coupled with the embargo system (which has existed way before EurekAlert), has contributed to a longstanding tension within science journalism about what gets covered—and what does not. Embargoes prohibit scientists and journalists from publicizing any new research until a given date has passed, specified by whatever journal is publishing the work.
Embargoes serve several purposes. First, they ensure that any new research has been properly peer reviewed before being presented to the public. Second, embargoes give reporters enough time to report the science accurately, without worrying about beating their competition to press. Inevitably, the system also creates PR for the research itself. Ever wonder why some days, in the middle of the morning, eleventy-one different news organizations will suddenly spam Google News, Facebook, and Twitter with the same groundbreaking discovery? An embargo has lifted, unleashing a flood of coverage.
Even restricted to publishing at the same time, journalists are competing. "Embargoes are hard to resist," says Paul Raeburn, media critic at the online science news site Undark.org. Sometimes a reporter's instinct tells them that a study is not worth reporting. "But refusing it is harder to do when you have the possibility of facing down your editor after an onslaught of your competitors all release that same story," he says. EurekAlert is not responsible for this problem. But the fact that it exists as a centralized hub for science research makes it easier for many individual reporters encounter the same story, and experience the same challenges to their news judgment.
EurekAlert is no longer the only embargoed-research portal. The UK-based Nature family of journals operates its own password-protected site, as does the Royal Society. But even though those and certain other journals do not partner with EurekAlert, their papers can still end up on the site via university PIOs promoting the work of their scientists. And just about every study ends up on Google Scholar after it publishes. But EurekAlert still has the most diverse collection of pre-publication papers.
And those other embargo services are, in some ways, cautionary tales of what access to research would have looked like without EurekAlert: Fragmented. "A reporter isn’t going to visit 50 to 100 sites every week," says Broadbent. "That’s where the idea came from, for a single site for protected exchange of information and involve more than one organization."
The service has stayed mostly the same. A few years after launch Torgenrud and his team updated the file system to a proper database, and AAAS eventually migrated EurekAlert from Stanford's servers to a commercial host. And of course the website itself has kept up with contemporary web design standards.
EurekAlert opened up science in a way that it had never been open before. The site has 12,000 registered reporters from 90 different countries (Getting embargoed access is a minor rite of passage for new science writers). It receives around 200 submissions a day, from 10,000 PIOs representing 6,000 different institutions all around the world. Once an embargo lifts, anyone is free to read the same press releases as the journalists (access to the original papers is a trickier ordeal). EurekAlert gets about 775,000 unique visitors every month. Its articles are translated into French, German, Spanish, Portuguese, Japanese, and Chinese.
Sure, the site is not perfect. It is arguably no longer even necessary—modern journalists are web native-or-die self-aggregators. But that's the thing. EurekAlert was never trying to be much more than a convenience. Which turned out to be its greatest gift: Making science easy to access.
* For the record, this author did not have a Ranma 1/2 fansite. It was a Gunsmith Cats fansite.
1 UPDATE: 05/16/2016 8:00 PM — Previously, this article had neglected to mention contributions from the university PIOs and journalists who conceived and presented the EurekAlert idea to the AAAS. Credit goes out to Joe Palca, Dennis Meredith (who came up with the EurekAlert name), David Salisbury (who arranged for EurekAlert's initial home on the Stanford University servers) and others instrumental in developing the platform.
