Out in the Open: Some MIT Scientists Sharing Results, Posting Unpublished Data
Barry J. Canton, a 28-year-old biological engineer at the Massachusetts Institute of Technology, has posted raw scientific data, his thesis proposal, and original research ideas on an online Web site for all to see.
To young people primed for openness by the confessional existence they live online, that may not seem like a big deal.
But in the world of science — where promotions, tenure, and fortune rest on publishing papers in prestigious journals, securing competitive grants, and patenting discoveries — it’s a brazen, potentially self-destructive move. To many scientists, leaving unfinished work and ideas in the open seems as reckless as leaving your debit card and password at a busy ATM machine.
Canton is part of a peaceful insurgency in science that is beginning to pry open an endeavor that still communicates its cutting-edge discoveries in much the same way it has since Ben Franklin was experimenting with lightning. Papers are published in research journals after being reviewed by specialists to ensure that the methods and conclusions are sound, a process that can take many months.
“We’re a generation who expects all information is a Google search away,” Canton said. “Not only is it a Google search away, but it’s also released immediately. As soon as it happens, the video is up on YouTube and on all the blogs. The old model feels kind of crazy when you’re used to this instant information.”
Openness has always been an integral part of science, with scientists presenting findings in journals or at conferences. But the open-science movement, with many of its leaders in the Boston area, encourages scientists to share techniques and even their work long before they are ready to present results, when they are devising research questions, running experiments, and analyzing data. In such open forums, the wisdom of the crowd could offer the ultimate form of peer review. And scientific information, they say, should be available without the hefty subscription fees charged by most journals.
It is an attempt to bring the kind of revolutionary and disruptive change to the laboratory that the Internet has already wrought on the music and print media industries. The idea is that opening up science could speed discoveries, increase collaboration, and transform the field in unforeseen ways.
On the other side are people who see the benefits of the status quo. For centuries, scientific discoveries have occurred at a steady clip, without the help of wikis or Web tools. Journals publish papers that have been scrutinized by specialists, ensuring that bad research doesn’t mislead other scientists or the public.
Scientists who plunge into openness also risk giving a competing lab a leg up.
“Maybe somebody has discovered some interesting gene and doesn’t want to blab to the whole world about why it’s interesting,” said Michael T. Laub, an assistant professor of biology at MIT. He says his lab is not overly secretive, but does not post “all the gory details of what someone is working on, because I don’t want my grad students necessarily to be scooped by someone else.”
More broadly, the entire system of credit in science is based on being the first to publish a finding in a reputable journal; there’s no incentive to post on blogs or community Web sites. Scientists try to get their findings published in the top journals in their fields, and major scientific prizes are awarded to those who make breakthroughs.
Despite these concerns, the counterculture scientific movement is gathering steam, and not just among junior researchers.
For example, OpenWetWare.org started out in 2005 as Endipedia, a Web site that scientists in Thomas F. Knight ’79 and Andrew D. Endy’s labs at MIT used to share information. But today the Web site is backed by a National Science Foundation grant, and more than 4,000 biologists and bioengineers from across the world have signed up to share techniques, get practical tips, and even detail their day-to-day work if they choose.
Science Commons, a nonprofit group based at MIT, works to Web-enable the scientific enterprise by working on other aspects of openness: trying to find ways to make inaccessible journals broadly available and developing Internet tools to ease sharing of information.
“In the same way you couldn’t get to Facebook until you had the Web for 10 years — all sorts of stuff had to happen to the Web itself to support the emergence of something like Facebook,” said John T. Wilbanks, executive director of Science Commons. “I think the tipping point will come when scientists look at someone next to them using the open system and getting more discoveries, and saying ‘I want that.’”
Another local effort, Somerville’s Journal of Visualized Experiments, is an open-access video journal that seeks to increase transparency in the how-to part of science, since researchers often waste time trying to replicate another team’s experiment.
The Web site, with tutorials from top researchers on subjects from basic stem cell techniques to dissecting mosquitoes’ salivary glands, is informed by the experience of its cofounder, Moshe Pritsker. Pritsker recalls how he as a graduate student spent more than a month unsuccessfully trying to replicate a two-year-old stem cell technique; eventually he flew to Scotland to learn firsthand.
It’s hard to say which, if any, of these forms of openness will gain traction in the wider community. But the ethos of the Internet, where people are used to getting everything from television shows to news articles without paying, is already challenging the scientific publishing industry.
There are open-access journals, such as those published by the Public Library of Science, but scientific journals usually require a paid subscription to get access. But in February, Harvard’s largest division, the Faculty of Arts and Science, voted unanimously to make scholarly papers authored by faculty available free in an online repository, which will begin beta-testing this fall. The National Institutes of Health began an open-access policy this year requiring that NIH-funded research be posted online for free, within a year of publication.
Just as giving content away for free on the Internet has proved troublesome for newspapers as they try to adapt to a new business model, scientific publishers worry that open access could undermine the foundation on which scientific communication is built. Journals typically make money through a combination of subscription fees paid by individuals or by universities and advertising, which support its editorial and peer-review process.
“The bottom line is it’s a wonderful experiment, but it needs to be approached carefully, or you go out of business,” said Fred Dylla, executive director of the American Institute of Physics, which publishes 11 of its own journals.
Eventually, the success of open science hinges on utility: If research improves, scientists will have to adopt it or fall behind.
Canton, working in the relatively new field of synthetic biology, has seen the benefits firsthand. He and colleagues devised a bit of genetic material that could be inserted into a cell to let it communicate with other cells.
They posted their work online, but also submitted it to a journal over a year ago to be formally presented to the world’s scientific community. Meanwhile, their work was incorporated into 18 different projects by other labs. Canton was invited to workshops.
Last month, it came out in the journal Nature Biotechnology.