Managing Scientific Inquiry in a Laboratory the Size of the Web
By ALEX WRIGHT
Published: December 27, 2010
Hanny van Arkel had been using the Galaxy Zoo Web site less than a week when she noticed something odd about the photograph of IC 2497, a minor galaxy in the Leo Minor constellation. “It was this strange thing,” she recalled: an enormous gas cloud, floating like a ghost in front of the spiral galaxy.
Adrie Mouthaan
THE NAMESAKE Hanny van Arkel , a Dutch schoolteacher, discovered the body known as Hanny’s Voorwerp by studying deep-space images.
Get Science News From The New York Times »
Readers' Comments
Share your thoughts.
A Dutch schoolteacher with no formal training in astronomy, Ms. van Arkel had joined tens of thousands of other Web volunteers to help classify photographs taken by deep-space telescopes. Stumped by the unusual image on her computer screen, she e-mailed the project staff for guidance. Staff members were stumped, too. And thus was christened the celestial body now known to astronomers worldwide as Hanny’s Voorwerp (Dutch for “object”).
Stories like Ms. van Arkel’s are becoming more common, as the Internet opens up new opportunities for so-called citizen scientists. And as millions of people get involved in these participatory projects, scientists are grappling with how best to harness the amateurs’ enthusiasm.
Some critics argue that citizen science projects are often little more than ploys to stimulate public interest rather than advance scientific knowledge. Others fret over the quality of data generated by nonspecialists. But scientists must weigh such risks against the benefits of a powerful new research tool: a vast computer network that can parcel out complex projects into small tasks that can be completed by individuals with relatively limited training.
Many got their first taste of citizen science withSETI@Home, which enlisted more than five million users in the search for signs of extraterrestrial life. Volunteers downloaded a program that used their computers’ idle processing cycles to sift through data from radio telescopes.
The success of SETI@Home has inspired a number of other grid-computing initiatives, like Grid Republic, a consortium of more than 50 projects relying on the same screensaver-based software, and IBM’s World Community Grid, which is being used by Chinese researchers to investigate efficient water-filtering techniques using nanotubes.
Now researchers are starting to look for ways to engage contributors in more substantive ways, taking advantage of what the Internet pundit Clay Shirky calls the “cognitive surplus” of online brainpower.
Like Galaxy Zoo, the Herbaria@home project turns users into volunteer taxonomists, classifying images from collections of herbarium specimens across Britain. To date, the site’s 284 registered users have cataloged more than 75,000 specimens.
Other citizen science projects enlist users as data gatherers, inviting them to contribute field observations about meteorological data, wildlife behavior and other phenomena.
Cornell University’s Nestwatch provides a platform for amateur ornithologists to share observations about the nesting habits of bird species across North America. Recently, organizers have focused on tracking the effects of the BP oil spill on local birds’ nesting patterns. And Australia’s ClimateWatch project invites users to track seasonal variations in plant and animal life cycles across the Southern Hemisphere.
Given the open-door participation policies, project administrators face major challenges in ensuring the integrity of their data. Staff members at Herbaria@home perform manual quality checks on about 5 percent of the incoming data, while ClimateWatch uses an algorithm to look for unusual data points, which can then be manually checked for accuracy.
Quality control aside, a larger question remains: Does the act of data gathering really constitute science?
“These people are not doing the work of scientists,” said David Weinberger, a senior researcher at the Berkman Center for the Internet and Society at Harvard, who is writing a book about the changing shape of human knowledge in the online era. “They are doing the work of scientific instruments.”
Stephen Emmott, head of computational research at Microsoft Research, agrees that most citizen science projects tend to treat participants as high-functioning cogs in a distributed machine. “Certainly this is participatory,” he said, “but is it science?”
Dr. Emmott believes that before Web users can claim the mantle of citizen scientists, they will have to be given more meaningful roles. “Participants should be able to make a genuine contribution,” he said, “and get something back.”
Dr. Emmott’s team is exploring new models that would involve users in the research process without compromising academic rigor.
Other researchers are looking to tap the higher brain functions of Web users. Foldit turns protein research into a game, offering Web users a puzzle in the form of a multicolored knot of spirals and clumps. Each puzzle represents an amino acid, which the user tries to fold into the most efficient shape possible. By combining computational algorithms with the visual problem-solving skills of more than 100,000 Web users, the researchers hope to pioneer a new approach to solving computational problems using supplemental human brain power.
Readers' Comments
Share your thoughts.
While some researchers are focused on expanding the contributions of average Web users, others want to cultivate people who already have some level of scientific training.
InnoCentive offers an online marketplace where organizations can pose problems to a network of more than 200,000 “solvers” who compete for cash rewards for the best solution. Recent InnoCentive contests include a $1 million prize for a biomarker to measure the progression of Lou Gehrig’s disease, as well as a $7,500 prize for better-smelling cat litter.
Other citizen science projects are taking shape at the grassroots level.DIYBio is a two-year-old organization with more than 1,500 members interested in conducting their own collaborative science projects. The organization has now spawned three physical lab spaces in Boston, Brooklyn and San Francisco , where members can participate in projects with polymerase chain reaction tests, centrifuges and DNA sequencing equipment.
“We’re trying to get people more involved in hands-on science,” said Jason Bobe, a founder of DIYBio. Members are exploring opportunities to turn genome sequencing into a game in which participants pull in data from crosswalk buttons and the like to model the spread of micro-organisms in urban environments. In a recent experiment, members collected dollar bills from 50 volunteers, then extracted the DNA for sequencing to identify all the micro-organisms.
As new self-organizing research communities emerge online, the long-term impact of citizen science projects may have less to do with the citizens than with the scientists.
When Web users can review research in progress and engage scientists in open discussions, the traditional process of peer review based on the model of academic publishing will face some new challenges. “It opens up another layer of participation,” Dr. Weinberger said, “and that’s where the deep impact is felt.”
Meanwhile, out on the open Web, hundreds of thousands of nonscientists continue to look for opportunities to take part in scientific research, inspired by the example of people like Ms. van Arkel, who has been enjoying the perks of Internet microstardom. She now blogs, maintains a Twitter feed and speaks at conferences and recently took a star turn in her own community-generated comic book. In January, her place in the annals of science will finally be cemented when her name appears as co-author of a scientific paper documenting her discovery.
Still, Ms. van Arkel is under no illusions about her place in the scientific establishment. “I like to tell people that you can do science without being a scientist,” she said. “It’s a life-changing thing.”
No hay comentarios:
Publicar un comentario