Scientists Examine How Networks Are Affecting Their Work

Everyone is using them, and a new study will try to learn how computer networks have changed the practice of science MADISON, Wisc.--The mail, as usual, is waiting for University of Wisconsin mathematician and psychologist Dennis Fryback when he arrives at his office. But it's not exactly "stacked up"; it's more "loaded in." This morning's batch--data from an off-campus collaborator about their research on cancer treatment cost effectiveness, a citation Fryback had requested from an editor in Michigan, and a query from a colleague in Turkey about some classroom materials that Fryback created--arrived electronically. And that's exactly how Fryback--in less than an hour--answers it. Scientists like Fryback increasingly are logging onto computer networks to do their work. The networks help them avoid telephone tag, make it easier to swap files, and provide access to supercomputers, instruments, and databases thousands of miles away. But Fryback wants to do more than simply use the new technology. He also wants to understand how it affects his life and the lives of his colleagues. And a new study that he helped to bring about may provide some answers. As head of a university committee that earmarks money donated by Ameritech, the Chicago-based parent corporation of several regional phone companies in the Midwest, Fryback pushed hard for a study about networks. That led to a collaboration between plant pathologist Jo Handelsman and rural sociologist Jack Kloppenburg, Jr., both of the University of Wisconsin, and Stephen P. Pollak, a doctoral candidate in science policy at the University of Sussex in England. The trio has received $85,000 over two years from Ameritech for what they believe to be the first systematic study of computer networking on an international scale, as well as its potential impact on the scientific process. "The beauty of networks is that you can pull people from a wide range of disciplines or places into a very tightly knit community," says Handelsman. "The question is, is that going to affect the way we interact as scientists or the way we ask questions of each other and the way we gather information?" Computer networks have been threaded through virtually every major university and research institution around the world, and into increasing numbers of smaller ones. And they have become quite popular. The volume of traffic on NSFNET, the largest research and education network in the United States, has tripled in the last year, from 1 billion packets routed in July 1989 to more than 3.4 billion packets in July 1990. (A packet averages about 190 characters.) To study this phenomenon, the Wisconsin researchers will examine interactions among members of the Microbial Strain Data Network (MSDN), which was started by an international group of microbiologists in the mid-1980s. The idea was to link the records of collections of microorganisms housed at laboratories around the world, including the American Type Culture Collection in Rockville, Md. (The Scientist, August 20, 1990, page 1). For about $30, a registered user with the proper software can consult a directory of collections and, via an electronic gateway, tie in to any one of them, or into any other user. For example, a scientist might be looking for strains of bacteria that can digest certain pollutants. Without leaving the system, the researcher could also search the literature available on the organisms and contact the person who has them. Each month the user receives a bill that is based on the amount of time that is spent on the network. The network, administered out of Cambridge University, has received broad financial support from such organizations as the National Institutes of Health, the United Nations Environment Programme, and the European Community. So far MSDN lists 450 registered users, representing about 25 countries, says Pollak. About a third of its users work for pharmaceutical and biotechnology companies. Testing Query Response For the study, the researchers intend to send two queries over the MSDN network and measure the response of users. They will pose the same requests to scientists tied into three other relevant networks, and to those at a scientific meeting. The first query will likely be a request for a microbe with a particular feature. The investigators will keep track of the number of responses received, how quickly they come in, and how wide a range of users respond. The second question will be more complicated, such as how one might solve a common but tricky laboratory problem. This query is designed to elicit more creative and informal responses. "We're curious because of the anonymity of the network," says Handelsman. "Do you get a different type of response than you'd get if you stood up at a meeting and asked this question?" Researchers hope their study will help them test the accuracy of various stereotypes about who uses computers to communicate. Some senior scientists suspect that network users tend to be young. Others say that women scientists are more likely to speak up on a network than in other forums. And there is a feeling among some users that networks enrich scientific debate by attracting "peripheral" scientists from small institutions or from developing or Eastern European countries. Says Pollak, "Even if you're not at a large institution, you may still be integrated into the scientific information flows." The researchers hope that their results will suggest ways to improve scientific discourse on networks and to develop new networks. The design and management of scientific networks promises to be a growth industry, observers say, given the increasing use of existing ones. Already, networks have created a demand for scientists with a knowledge of both computers and the disciplines that the networks will serve. MSDN, for example, is managed by microbiologists who have completed a master's-level curriculum in information science. But not everyone is enamored of this new mode of communication. "It's an absolute disaster for historians of science," says Cornell University's William H. Provine, who chronicles evolutionary biology and genetics. "The problem is no one saves electronic communications for posterity." "I think it's going to make things especially difficult for historians in the future who are documenting modern science," he adds. But even pen-and-paper-loving historians seem to have succumbed to the ease of electronic mail. Eighty percent of his own correspondence, Provine admits, runs over Bitnet. Christine Mlot is a freelance science writer based in Milwaukee.

Scientists Examine How Networks Are Affecting Their Work

Interested in reading more?

Become a Member of