A Revolution in the Tools of Scientific and Engineering Research
“The summation of human experience is being expanded at a prodigious rate, and the means we use for threading through the consequent maze to the momentarily important item is the same as was used in the days of square-rigged ships. But there are signs of a change as new and powerful instrumentalities come into use.” (Vannevar Bush, 1945)
Digital computers and networks are revolutionizing the way we conduct research, just as they are recasting the way we do business and spend our leisure time. It grows easier every day to collect, manipulate, and disseminate information. Researchers everywhere are developing new ways to use these digital systems in every phase of their work. These new tools— both hardware and software—make possible more creative and productive research, including new forms of collaboration on a global scale.
Advanced information technologies have become more than simply faster, more precise, and more automated research tools. They have opened up new ways of learning and communicating for anyone in the business of creating knowledge. No discipline in science or engineering is exempt.
These new ways of conducting research are also challenging many traditions of academic research. All research institutions, including universities, libraries, corporate and government laboratories, publishers, professional associations, and research funding agencies, will need to adapt. At the same time, utmost care must be taken to preserve the core values of research. Applications of technology must be monitored to ensure that they do not threaten researchers' expression of these values.
An equally profound challenge, however, is improving the rising flood of information in digital form and the capacity of the researchers to receive, assess, and act on this information. As the tools get better, the task of research may get harder, since the data is growing more complex and diffuse. In some fields, more information is helpful. In others—especially those involving complex questions at the frontiers of knowledge—it may be overwhelming and distracting. For many fields, what will be needed is better information, and better ways to handle it.
Computers connected by high-speed networks to other devices throughout the world are powerful information systems. Public and private networks permit transmission of nearly instant voice, images, and other information to wide audiences around the world—and all at low cost.
And digital technology continues to advance at a nearly inconceivable rate. For the past four decades, the speed and storage capacity of computers has doubled every 18 to 24 months. Cost, size, and power consumption have shrunk at about the same rate. The data capacity of networks has increased one thousand-fold in just the past decade. Traffic on the Internet is doubling every three months.
The uses that will be made of advanced digital tools are difficult to anticipate. Some of the new applications are already having a fundamental impact on academic life. It is impossible to predict long-term effects, both positive and negative.
In education, teachers use computers to enhance conventional classes, for example, by creating Web sites for course materials, showing demonstrations, and using programmed instructional packages. The new technology has also created a new form of education. In some instances, course materials have been put online, and the number of face-to-face classes has sometimes declined. Most radically, technology permits so-
called distance education, in which students and instructors may never meet face-to-face. These situations raise one of the greatest drawbacks of the digital age—increasing depersonalization. Some can compensate for this, but others can become increasingly isolated. Institutions of learning must take great care in moving into this arena.
Libraries are also changing, as almost all new information is created in digital format and much of it is available online. The need for a student or a researcher physically to be near a library is diminishing. Specialized information previously unavailable to the general public is now available. For example, the Library of Congress is mounting its collections on the World Wide Web; the National Institutes of Health offers medical information to everybody, not only doctors; and the Cornell University Legal Information Institute Web site emphasizes service to professionals who are not lawyers. Stanford University, among others, has established a “digital library.” Many universities have licenses to such digital libraries and offer access to all their faculty and staff. The Association for Computing Machinery has placed all of its publications in a digital library.
Publishing is also changing. Posting research results on the Internet offers researchers a fast and low-cost method of disseminating their results around the world. For example, physicists post preprints of papers in the online archives at Los Alamos National Laboratory. The Association for Computing Machinery and the Institute of Electrical and Electronic Engineers, among others, offer similar services. Publishers are converting their journals to electronic format and are eliminating paper versions of some journals. Since electronic journals are not constrained by the cost of printing, researchers can augment research papers with raw data and examples.
Almost every field of research is making use of computing, and in some disciplines the methods of research are changing fundamentally:
The Internet permits social scientists to conduct surveys more quickly, more cheaply, and in greater detail. For example, data mining is being used to detect patterns in structured and unstructured sets of data.
Remote access to instrumentation, including the remote operation and shared control of robot-like data-harvesting or data-refinement tools, is increasingly the rule. Early examples of this form of “telescience” include network access to an
automated telescope on a mountain in Chile, to an ultrahighvoltage electron microscope in Japan, to a synchrotron in Palo Alto, and to a supercomputer in San Diego. All these instruments are now online, and accessible from digital portals by authorized researchers throughout the world.
Online databases let researchers exchange, analyze, and manipulate protein structures, infrared spectra, recordings of artistic performances, and a vast array of other information. (The Genome Database is at the heart of the Human Genome Initiative.)
Computer simulation has become a standard tool of research in many fields of science and engineering. Some say it has become a “third modality,” as important as theory and experiment. This kind of simulation, never possible before, has made it routine for researchers to design new drug molecules and test their properties “in silico,” understand the complex behavior of networks of brain cells and synaptic clusters in large-scale computer simulations, watch model galaxies collide, and “feel” the forces exerted as a drug docks in a protein.
New ways to represent information have appeared using combinations of written text, dynamic images, and sound. Instructions on how to create new sensory environments are packaged in new, dynamic modes of communication (IEEE Spectrum, 1996). Multimodal methods for representing data are growing rapidly, challenging and expanding users' skills and making use of seldom used abilities of the human nervous system to accommodate broader and more varied visual, auditory, and tactile input. Likely, the most advanced video games and flight training environments, with their heightened interactivity and their density of information, foreshadow the Web-based data exploration environments of the near future.
Desktop publishing, digital video, and Web site development are emerging as new arenas of creativity.
All of these things are happening now. Foreseeing their full effects on research is impossible. As researchers use these technologies, and as they organize their lives and careers around them, personal and professional incentives and social interactions may change.
The rapidly expanding availability of primary sources of data in digital form may be shifting the balance of research away from working with secondary sources such as scholarly publications. Researchers today struggle to extract meaning
from these masses of data, because our techniques of searching, analyzing, interpreting, and certifying information remain primitive. New automated systems, and perhaps new intermediary institutions for searching and authenticating information, will develop to provide these services, much as libraries and scholarly publications served these roles in the past.
It is clear that profound changes are taking place in the way research is conducted and disseminated. The value researchers place on open communication about methods and results and the diligent search for truth, however, appears to remain constant.
CHALLENGES TO THE INSTITUTIONS OF RESEARCH
The institutions of research include organizations such as universities and laboratories, libraries, museums, publishers, and professional associations, as well as the formal and informal norms of professional behavior. Together they define the boundaries of what is possible and permissible. Those boundaries are shifting strongly, and cannot be discerned with confidence, but some generalizations are possible.
The evolving applications of information technology offer opportunities for cooperative relationships among widely dispersed people who are interested in sharing new experiences and intellectual pursuits. The resulting opportunities for investigation and communication will multiply far beyond the bounds of the physical campus. Research careers in the future may be built independent of today's institutions, unless those institutions can offer attractive infrastructure and resources not available elsewhere.
Libraries will expand their functions, offering users remote access to the world's information through text, images, sound, and other forms of information not yet envisioned. Libraries are moving beyond their traditional role as selective stockpilers of paper documents, exploiting and developing new tools for identifying, retrieving, and disseminating information in all of its forms. They are certain to be integrated with the Internet
through the successful and flexible World Wide Web interface (see, for example, Fox, 1999).
One function of libraries that will remain, and perhaps increase in importance, is helping users navigate, analyze, and evaluate the integrity and reliability of information. The very wealth of information available already makes it hard to find a particular piece of data. The techniques of information retrieval will need to keep pace with the large and complex distributed databases that are increasingly housing the world's information. Libraries, as organizations, may change dramatically, but the need for information professionals to support research is unlikely to diminish. They will require increasingly technical expertise and training to play the pivotal role that researchers and institutions demand of them.
Scientific and engineering journals are being supplemented, and in some cases supplanted, by digital forms of publication, both formal and informal. Most journals today have electronic counterparts on the Internet, and some have abandoned paper circulation altogether. In the past decade or so, some researchers have come to use press releases and Web sites to announce results before they are published in peer-reviewed journals. A vast literature of uncertain authority has arisen.
Electronic publication has great advantages: low cost, speed, and the potential for great depth of reference and cross-reference (i.e., through hyperlinks to source material and other documentation). But it challenges many of the customs of research. The academic professions and institutions will need to develop responses that preserve traditions of open inquiry and communication but do not stifle these important digital tools.
Electronic publication allows articles to be constantly revised to reflect new findings. But those who cite results from such a source must be able to identify just which version they are referring to. There are techniques for tracking changes over time, but they are not used uniformly today (Brown and Duiguid, 1995).
Peer-reviewed publication has always played an important role in linking research results with particular researchers or teams. Through conventions of peer review and publication, results and discoveries are credited to researchers. First publi-
cation typically leads to attribution of priority for results and discoveries. Electronic publication offers an enormous variety of publications, including peer-reviewed online journals, internal Web-based research notes, and individual Web pages, many of which are not peer reviewed. It also makes the discovery of previously published work more difficult. Thus, electronic publication may weaken existing conventions for attributing credit and priority (Nissenbaum, 1998). As with any form of “gray literature,” quality control is the responsibility of the author who chooses to cite such information.
Electronic publication, because it can be evanescent, also requires new forms of archiving. If a journal goes out of business in the paper world, all of its back issues normally remain available and securely indexed in libraries. If the same thing happens in cyberspace, its contents and their cataloging information vanish forever unless systematically archived. Standards for preserving electronic data and publications are under development (See http://www.arl.org/preserv; Waters and Garrett, 1996, also discuss these issues.)
Profound changes are taking place in academic publishing as publishers and researchers struggle with the opportunities and challenges of electronic information, but the underlying need has not changed. Researchers will always need ways to publish their research that are timely, accurate, and available to the entire academic community.
VIRTUAL COMMUNITIES OF RESEARCHERS?
Research is a process by which knowledge is created from information, disseminated to others, and preserved for future generations. These activities include hypothesis, observation and inquiry, data collection, analysis, discussion, dissemination, and reanalysis. Historically, most research has been conducted in universities, institutes, laboratories and similar organizations that provide access to books, facilities and instruments, and professional colleagues. Now, digital networks have greatly enhanced the ability for researchers to work “virtually” with colleagues around the world, regardless of their physical location. Distance education for graduate student researchers promises the same freedom from geographic constraints. While electronic communication can provide students with access to expertise beyond their immediate advisors and mentors, it cannot replace the close mentoring relationship that many
consider to be at the heart of today's graduate education. While no one expects the imminent demise of university campuses, the technology is advancing so rapidly that future researchers may organize themselves around global digital networks. Individuals will differ in the extent to which they may rely on these evolving networks. The world's universities and their laboratories and other facilities over the next few decades will be joined by new types of institutions, able to exploit the evolving information technology and sometimes competing with them. (Noam, 1995; Wulf 1995; Lenzer, 1997; O'Donnell, 1998; Duderstadt, 1999).