CONTACT: Stanford University News Service (650) 723-2558
HighWire Press a pioneer in moving scientific journals online
STANFORD -- Biology Professor Robert Simoni has set himself a test: He wants to write a scientific review article without going to the library to look up papers or references.
Two years ago, that would have been unthinkable. But now it's just a matter of time, due in large part to the efforts of HighWire Press, a fledgling division of Stanford Libraries that is a key player in shifting distribution of scientific ideas from printed journals to online publications.
According to advocates, online publication holds the potential to help scientists better manage increasing volumes of scientific literature; speed communications among scientists; reduce the cost of distributing scientific information; and even improve the quality of science news available to the general public.
"This is a transforming technology," declares Michael Keller, Stanford University Librarian and publisher of HighWire Press.
Simoni, an editor of the Journal of Biological Chemistry (JBC), the first journal produced online by HighWire Press, says that at first he was skeptical. "I frankly took [Keller's] claims that we could revolutionize how science is published and disseminated as hyperbole. But now I'm really a convert. I think the revolution has started."
As to his test: "I just finished a review [for which] I only had to go to the library once or twice. It is quite amazing."
The number of academic journals going online has been growing steadily for five years. According to a study by Philip F. McEldowney at the University of North Carolina at Greensboro, in 1991 there were fewer than 30 electronic journals, as measured by the Association of Research Libraries' Directory of Electronic Journals, Newsletters and Academic Discussion Lists. By 1995 the number had grown to more than 300. Of these, about one-third deal with scientific topics, he reports.
HighWire Press caught this wave in January 1995 when it began producing the online version of JBC for the American Society for Biochemistry and Molecular Biology (ASBMB). Last month HighWire also began producing a full text electronic version of the journal Science for the American Association for the Advancement of Science, one of the nation's oldest and largest scientific societies.
These projects have propelled the group into the vanguard of the post-Gutenberg shift from ink on paper to electrons. As word has gotten around to other scientific societies, additional proposals have been flooding in. Already HighWire Press has signed contracts with more than a dozen other journals, according to director John Sack, who says he is particularly eager to talk to Stanford scientists who edit journals and are interested in going online.
In a little more than two years, HighWire Press is nearly self-supporting, Keller says. It employs 15 people (not all full time) and brought in a gross income of about $800,000 in fiscal year 1996. Next year he projects that its income will double.
Too much information to mail
Although HighWire Press had been a gleam in Keller's eye for nearly a decade, its conception dates to 1993 when Simoni and his fellow editors realized that their journal was getting too large. Although they had switched to weekly publication, volumes had grown to the size of phone books (800-plus pages) and were approaching the U.S. Post Office's four-pound weight limit for second-class distribution. For three years, the society had tried quarterly publication of the journal on CD-ROM, but subscribers disliked the additional time delay.
"At a Faculty Senate meeting I asked some of our networking people if they could put the CD-ROM version of the journal on a university system so it could be distributed electronically," Simoni recalls. After the meeting Keller approached him and proposed the alternative idea of putting the journal on the World Wide Web.
Keller put together a team consisting of librarian Michael Newman, Sandra Senti from Network Services and Ann Mueller from the Stanford Data Center who developed a technical model and cost estimates. After a year of discussion, the ASBMB board agreed to a joint venture with Stanford Libraries. But the board insisted that the prototype be finished in time for the society's annual meeting only three months away.
To head the project Keller tapped Sack, then director of the Stanford Data Center. Sack came up with the name HighWire, which he says was appropriate because the group had taken on a very ambitious project and was working without a net.
"We were trying to put the second-largest peer-reviewed publication in the world online full text in three months when vastly larger organizations had worked for 18 months and were still experimenting. We knew we had to be successful, but there were no guarantees."
Despite such knotty problems as reproducing Greek letters, mathematical symbols and equations, and the other non-standard characters that are used in scientific literature, the group did in fact produce the first online issue of the journal in time for the society's annual meeting.
"These people have done what everyone at Stanford does," says Simoni. "They looked at this as a giant, exciting experiment, which it really is."
Simoni says his fellow biologists and biochemists have greeted JBC Online with "a curious mix of enthusiasm for the technical accomplishment and reluctance to abandon the print in favor of the electronic."
The society provided the online version free to subscribers for the first year. Recently it has adopted a fee schedule that requires libraries to pay a significant amount extra to receive both the electronic and print versions. The goal is to get them to begin switching to the electronic version.
"The shift is going more slowly than we had hoped. I think everyone recognizes that it is the future. But there is institutional reluctance to make the switch," Simoni says.
This reluctance represents one of the major uncertainties facing online publishers. It is not yet clear how they are going to recoup the costs involved in going electronic. At present, most publishers appear likely to limit access to full text and some value-added services to paid subscribers while keeping their tables of contents, abstracts and searchable archives available for free, Sack says.
Other questions that have been raised by some observers are whether online products can maintain the same quality as their print predecessors, and whether they can achieve the stability and longevity of print journals.
Taking Science magazine on line
Nevertheless, scientific journals like Science are forging ahead. "Once we realized that it was possible to deliver figures, graphics and further data electronically, we felt we had to exploit the new medium in a way that would serve our community better," says Science editor Ellis Rubinstein.
The journal had already started a web page before hooking up with the Stanford operation. But, when Rubinstein and his fellow editors heard what HighWire Press was doing, they decided to join forces. "We were very impressed with [HighWire's] plans for JBC," he says.
The Science editors were concerned that taking on a major new project, with so many unknowns, might overextend their staff, resulting in a decline in the quality of their work. HighWire Press helped reduce this strain, Rubinstein says.
"HighWire is . . . not just a vendor. They are a collaborator. They bring a sense of the market, based on their contact with the scientists at Stanford, and their library background allows them to bring interesting ideas to the table."
The leadership at Science has been surprised at how popular its site has proven, even before the addition of full text of the journal articles. According to their surveys, the site was averaging more than 25,000 unique visitors per week. About two-thirds of those do not subscribe or read the magazine regularly, they have found. Activity at the site has grown by 40 percent since the full text of articles was added, Sack says.
"In the past, we pushed information out to our readership," Rubinstein says. "With this new technology, we need to rethink what we do. We need to become more of an information interchange."
Sack calls this a shift from a publication to a communications model. "Maybe these journal articles that scientists publish are an artifact of the scientific communication process. Maybe if we evaluate what the scientists are really trying to communicate, there are better ways to do so, given the more interactive, non-broadcast kind of technology that is now available," he says.
Simply reducing the barriers of geography and time, and the vagaries of national and institutional postal services, are very important to scientists in many parts of the world, Rubinstein points out. On a recent trip to Asia, for example, he found that the scientists there were "very emotional and excited" about the magazine's online efforts, which allow them to read Science articles when they are published, rather than weeks later when the magazine arrives.
Simoni emphasizes that electronic journals will help him and his colleagues cope with the tremendous information explosion that has been taking place in the scientific literature.
"I have to read or at least peruse four or five journals [including the 800-page per week JBC] regularly," Simoni says. "A great advantage of the electronic version is that it allows you to search for specific topics and even to preprogram your computer to download copies of articles on specific topics. So it is not just another form of dissemination."
Another potential benefit for scientists is getting their findings into print faster. Going online at JBC beats mail delivery by two weeks, Simoni says. Moving from paper to electronic communications for the peer review portion of the process will skim another two weeks off the process, he estimates. "So we should be able to get the information out a month sooner. Some people find this enormously important," Simoni says.
Rubinstein agrees that going electronic will have a major impact on how scientists do their work, but adds that it also will have an important effect on how the public gets its information about science in the future.
"This technology makes it easier to adapt scientific material for lay audiences," he says. Publications like Science will be able to provide the public with news about scientific developments directly, rather than relying solely on the media.
Reduced distribution costs
If HighWire Press is successful, it also may help substantially reduce the costs of distributing scientific information.
"For some time I have been concerned about the treatment of scientific information as a commodity," Keller says. "Commercial publishers like Elsevier, Springer, John Wiley and Academic Press are getting rich by getting information from scientists for free and selling it back to their university libraries for large and annually skyrocketing fees. This technology allows us to provide an economically viable alternative, and we have a certain evangelic fervor about it."
A 1994 study by the Association of Research Libraries showed that average journal subscription costs had more than doubled since 1986. And a recent article in Forbes magazine reported that the Amsterdam-based Reed Elsevier earned a pre-tax profit of nearly 40 percent on its scientific journals.
Commercial publishers are spending millions of dollars to bring their journals online. The companies can afford to do so because they are making profits of 25 to 30 percent per year and have very large cash reserves, Keller says. All 1,100 of Elsevier's journals, for example, are now available electronically.
"Scientific societies, even societies as large as the AAAS, don't have millions of dollars to invest in electronic publishing," Sack says. "Somehow we've come up with an approach that keeps the not-for-profit societies from being disadvantaged in adopting new technology relative to the for-profits. By working with us, they have in a sense formed a collective to co-invest in the development of electronic technologies that they can use."
HighWire Press intends to broaden this base even further by developing and selling software that will allow non-profits to produce electronic versions of their journals by themselves.
While this will not eliminate the problem represented by the largest for-profit science publishers, Keller and Sack maintain that it will help keep them in check by providing lower-cost, alternative distribution methods for scientific information.
Despite all the activity, Simoni cautions that the final results are far from clear. "It would be a mistake to lose sight of the fact that this remains an experiment," he says.