"There is perhaps no area of modern technology that combines so many potential difficulties and problems of transfer and implementation with such high promise," began Dr. Philip F. Palmedo, chairman of the International Resources Group, in introducing information and communications technologies and their applications in the fields of agriculture, health care, and education.

"Information technology" is defined broadly to include conventional communications systems (telephones as well as satellite and fiber optic communications), radio and television, microprocessors, computers, and all the software, applications, and combinations of these technologies, including, for example, robotics and computer/video systems.

According to Palmedo, information technology is perhaps most valuable when used to provide people in developing countries with the information they need to make a correct decision - be they farmers trying to determine when they should plant their crop, or health workers trying to diagnose an illness. The timely availability of information and know-how is one of the critical barriers to development. Palmedo cautioned, however, that the application of post-industrial state technologies to preindustrial or industrializing countries "raises profound issues of both desirability and feasibility."

What are the issues surrounding use of these technologies in developing countries? According to Palmedo, they are:
· The dazzle effect. It is not safe to assume that dazzling and exciting new information technologies can solve major problems. There is already a tendency to overestimate the role of hardware, which traditionally is considered first in importance, followed by software, adaptation and training, and institutional development. This list should be reversed, argued Palmedo. For example, in one educational program the fascination with television interfered with the progress being made with use of the radio as a tool. The result was a deterioration in the services provided because the television system absorbed all the attention and funds. The lesson to be learned is that applications of information technologies should be driven by the needs, not by the technologies. Technical assistance agencies must understand where and how (in detail) to apply the appropriate technology and must implement accompanying programs in policy and institutional development so the new system will work.
· Dual sector problem. A national capability in information technologies is useful only if it contributes to the development of other national priorities, such as improved health services, increased agricultural production, and advances in education.
· Structural realities. Information is a valuable resource. In some countries power is derived from the withholding of information. Thus, in the transfer of information technologies the structural characteristics of developing country governments must be considered. Two-way flow of information is essential. For example, technical information must flow from the government to the farmers and information on requirements and results must flow from the farmers to the government.
· Quantification of benefits. Although the benefits attached to the use of information technologies are usually indirect, it is important that they be measured no matter how difficult that may be. According to Palmedo, "it may well be that from a development perspective we have seriously underinvested in communications systems in the past."
· Role of the private sector. Much of the technology in this area resides in the private sector, leaving the door open for closer ties between foreign assistance agencies and private industry than has been the case in the past. For example, the U.S. Telecommunications Training Institute, an organization funded by AT&T, MCI, COMSAT, and others, has been used to train developing country professionals. USAID makes it possible for developing country technicians to take advantage of this training by funding their travel.

Microcomputer Applications in Developing Countries: A Project of the Board on Science and Technology for International Development

Microcomputers are an increasingly important tool in all aspects of development as the need to handle and assimilate vast quantities of information becomes ever more critical for both the international development community and the developing countries.... Unfortunately this new technology represents not only an opportunity if properly exploited but a threat if ignored. The widespread and increasing incorporation of microcomputers into all aspects of the developed countries represents a major technological advance and an inevitable social change. If a developing country fails to take advantage of the opportunity that microcomputer technology represents, its level of development in relation to developed countries will be significantly lowered.

- Microcomputers and Their Applications for Developing Countries (Westview Press, 1986)

This volume is the first publication emanating from a BOSTID project of the same title. In 1983 the U.S. Agency for International Development asked BOSTID to convene a series of four symposia to assess the implications of microcomputers for developing countries. The first symposium was held in Sri Lanka in 1984, and it addressed applications in agriculture, health, and energy (the above volume is a product of that meeting). The second symposium, which looked at applications in education, was held in Mexico in 1985, and the third was held in Lisbon in the fall of 1986 to examine expert systems and cutting edge technologies. Policies, priorities, and economic and public administration issues were the subject of the final symposium held in Washington, D.C. in January 1988. The proceedings of all four symposia are being published by Westview Press.

Applications to Agriculture

Although there have been dramatic improvements in agricultural science during the past 25 years, the transfer of research results to applications has been a problem. This finding was reported by Dr. James W. Jones, professor of agricultural engineering at the University of Florida. Recent advances in information science and computer technology have narrowed the gap between research and application in developed countries. Similar success is needed in developing countries. For example, if a country is importing much of its cooking oil and wishes to expand peanut production to meet national needs, this technology can suggest where and how the expansion should occur and which proposed agricultural projects should be considered. By using computer technology to integrate, organize, and store information and make it available to users in a timely and efficient manner, scientists, planners, and farmers can increase their ability to predict the outcomes of agricultural systems under various conditions. This approach is illustrated in Figure 1, which depicts the framework used by the IBSNAT (International Benchmark Sites Network for Agrotechnology Transfer) project.

Simulation models, data bases, and expert systems are the computer tools used to accomplish these tasks. The first tool, crop simulation models, integrates what is known about the crop-soil-weather-management processes so that crop yields can be predicted. Changes in temperature, humidity, precipitation, wind, solar radiation, and soil conditions cause tremendous variability in crop yields - over time and space. Crop simulation models cannot, however, be used without testing, cautioned Jones. These models should be tested and modified by researchers as needed for their own cultivars. Experiments can then be simulated on the computer and analyzed to help design critical field experiments to evaluate model results. This more efficient approach saves time and reduces the cost of field experiments.

Crop simulation models need databases, the second computer tool, to work. Computers can quite effectively organize and store data on soils, weather, crops, management, and economics for predictive models. With the advent of the more powerful microcomputers, spatial data (data on soils over a region) can be used to predict crop yields for a region and display them on maps.

The third computer tool, expert systems, has considerable potential for developing countries. Expert systems are computer programs that can search stored knowledge for answers to specific problems. Much of our agricultural knowledge is based on experience that has not evolved far enough for direct integration into simulation models. For example, cited Jones, diagnoses of plant diseases or soil fertility problems, or recommendations of crop varieties or pesticide types and the amounts to use, do not lend themselves well to simulation. An expert system can store this information in the computer, thereby making expertise available when real experts are not. These systems can be adapted to specific locations and can be used by extension workers, with the appropriate training, to advise farmers. Of the various expert systems that have been developed, one of the most notable is for managing soil pH in Indonesia. According to Jones, this methodology is only in its infancy, but it shows considerable promise and should be developed further. The IBSNAT project is evaluating the capabilities of these technologies.

In concluding, Jones called for science and education programs to encourage the integration of this technology into existing research and technology transfer programs for developing countries. "Emphasis should be placed on research to develop and test crop simulation models and the expert systems needed to help analyze critical problems, and to develop spatial databases.... In this way the tools of the information age can be used by scientists around the world to help integrate, organize, and store agricultural knowledge in a form that can be easily transferred to and used by others."

IBSNAT International Benchmark Sites Network for Agrotechnology Transfer

IBSNAT, a program supported by the U.S. Agency for International Development and carried out by the University of Hawaii, uses advanced computer technology to overcome the bottlenecks that prevent new or alternative crops, cultivars, products, and practices from being quickly integrated into Third World farming systems. A decision support system, devised by combining soil' crop, weather, and farm management data bases with simulation models and expert systems, allows long-term strategic planning and day-to-day tactical decisions by farmers.

Participants in this program - scientists from developing country research centers, planning agencies, and universities, as well as international research centers - hope to accelerate the transfer of agrotechnology and to maximize its successes and minimize its risks. They will also assess the long-term effects of agricultural practices on the agroecosystem.

With the recent advances in computer technology and artificial intelligence, the costly and time-consuming trial and error method of technology transfer can be supplemented by a systems-based approach. This approach can predict the performance of a crop, product, or practice in any location during any season and for a wide range of management alternatives. The advances in information technology include personal computers with near-mainframe capabilities, large-capacity data-base management systems that are easily accessed, and new developments in the field of artificial intelligence.

The scientists participating in this program have already made considerable progress using these advanced technologies. Five two-week workshops have been held in various countries to train agricultural scientists in these concepts, and crop simulation models for corn, soybean, wheat, and peanut have been developed and are being tested (models for other crops are being developed). Work on expert systems is just beginning.

Applications to Health Care

In trying to apply information and communications technologies to health care systems in the twenty-first century, developing countries will face many difficulties. The most obvious, of course, is the lack of the requisite hardware, software, and skills. Less obvious difficulties are long-term development policies that frequently do not include plans to immediately use this technology (developing country policymakers are frequently unaware of what is available in artificial intelligence technologies); inappropriate cultural backgrounds in informatics; and managers who expect fast results with low investment, resulting in the use of black box, inappropriate, or obsolete applications.

After identifying some of the difficulties in utilizing cutting edge information and communications technologies, Dr. Jose Negrete, director of the Institute of Biomedical Research at the National Autonomous University of Mexico, proposed that if these technologies are to be used in developing countries, their indirect benefits should be maximized. That is, they should be used to increase standardization of medical care procedures, to lower the cost of health care, to increase the quality of patient care, and to help both physicians and teachers keep current with advances in medical science. For example, according to Negrete, the direct benefit of the use of microcomputers to provide decision theory and cost-effectiveness calculations for medical decision making would be lower health costs - because the information needed for the otherwise simple calculations is not available. The indirect benefits of this technology would be both short and long term. One short-term benefit would be the acceptance that decision making in medicine can be openly and rationally discussed, using the appropriate statistical or other data. This indirect benefit will lead eventually to the desirable standardization of medical care procedures and to a reduction in health care costs.

The long-term indirect benefit would be obtained after the short-term benefits have permeated medical and paramedical training in the form of tutorial programs that work with local data and locally constructed decision trees.

The use of computers in medical diagnosis and patient management would also have large indirect benefits, such as the unique opportunity to take a close look at the local medical expertise, a necessary step before feeding useful knowledge into the computer, and the eventual diffusion of this reviewed and revised expertise into the classroom. Such an examination of medical knowledge will significantly improve the quality of the patient care.

Applications to Education

Well over a billion persons worldwide are affected by issues of literacy and education, stated Dr. Kurt D. Moses, director of service systems of the International Division of the Academy for Educational Development. Over 600 million people lack basic abilities in literacy and numeracy. But, in a sense, said Moses, applications of new information and communications technologies should concentrate on the estimated 340 million students who are receiving a primary education.

Using gross data from UNESCO, the World Bank, and the International Monetary Fund, Moses reported that the developed countries, excluding Eastern Europe and the Soviet Union, spend an average of $2,O55 per student per year on educational operating expenditures, while the developing countries, very broadly defined, spend an average of $210 per student per year for all levels of education (kindergarten through university). Africa, a special case, spends an average of $88 per student per year in operating support for I education.

Another measure that relates directly to the impact of information and communication technologies is based on that very small portion of the operating budget dedicated to materials and supplies (about 2-10 percent of the operating budget in most countries). In developed countries about $270 per student per year is spent on materials and supplies, while the respective figure for the developing countries is $15 and for Africa, $8.

Moses noted that the expectations of political leaders play a big role in the acceptance of technology and innovation designed to enhance education and literacy. These expectations are that new technologies will: (1) reduce costs and permit expansion of services (the strongest expectation), (2) upgrade the quality of education, (3) provide unique opportunities and services where none existed before, and (4) lead to human resource development and economic growth. The "dazzle effect" is responsible for many new technologies - particularly computers, satellites, and certain telecommunications innovations - being viewed as symbols of modernization. '´In my experience," said Moses, "[the latter] counts for at least 40 percent of the reasons for implementation of a variety of efforts."

Examples of Specific Applications

By giving examples of how information and communications technologies are already being applied to education and literacy in developing countries, and their respective cost structures, Moses was able to show why the cost-benefit structure often determines the use of many of the newer technologies.

In Indonesia, 13 universities located on various islands in the more than 13,000-island archipelago are linked by a single-channel, narrow-band telephone circuit. This telephone line is being used for teacher and student training, some administration for a newly established open university, and short-term training assistance for health and agricultural workers (see Figure 2). This USAID Rural Satellite Program has served 3,500 students per semester in 17 courses, and its "up time" averages about 92 percent. Many of these activities would not have been underaken without this satellite effort, partly because of cost. The basic cost of this system is about $10 per course per student (or about $100 per student per year) versus an estimated cost of $64 per course per student using the visiting lecturer approach. Comparison of this figure to the average support per student per year shows that in fact there is an implicit cost-benefit ratio in operation as to whether or not some efforts will be undertaken; some are appropriate for higher education but not for lower levels, where funds are less.

A second example offered by Moses is the use in Kenya of a more mass-oriented technology, instructional radio, to teach English in grades one through three. Carrying the core of the curriculum through daily broadcasts, this USAID radio project provided 98 instructional hours per grade per year. The basic cost of this program, assuming that the FM stations were in place, was between $.22 and $.40 per student per year, including programming. Here then is a technology that is to a certain extent affordable within current cost structures.

Thus, Moses concluded, in applying information and communications technologies "we will have to be extremely careful in the political environment to find the time when the cost curves and the benefit curves are at the right point for a lot of Third World countries."