|United Nations University - Work in Progress Newsletter - Volume 15, Number 1, 1998 (UNU, 1998, 12 pages)|
By Chris George and Xu Qiwen
From its offices in downtown Macau, the UNU centre for software development looks out over the Pearl River Delta and the South China Sea, one of the fastest growing high-tech areas in the world today. Here where the industrialized and developing worlds stand cheek by jowl is an appropriate setting for the UNU International Institute for Software Technology (UNU/IIST), a place which is helping the developing world to keep pace with the 21st century.
Personal computers are increasingly available at lower and lower prices. Technological advances in, for example, the small cellular phone with computer access may also soon help get around the power outages that have plagued many poorer countries. The problem is not the hardware, but the software, the instructions telling the computer what to do: compile a payroll, store data, solve abstruse equations, schedule a flight, perform word-processing or whatever. But most developing countries have few software companies and little experience in its creation.
As the price of hardware has plummeted, the potential of the computer remains frustratingly just out of reach - cheap but of little use without the proper software. The Macau centre responds directly to that need, strengthening the capabilities of the developing countries to produce their own software, designed specifically to meet their needs. The following article, by two scientists on the UNU/IIST academic staff, was written specially for Work in Progress. - Editor
Information has been called a resource that is not only renewable but self-generating. Running out of it is not a problem - drowning in it is. At its new institute for software technology, the UNU is seeking to help Third World communities put the new technology to use in improving their daily lives - and to stay afloat in the mounting tide of data.
The small personal computer is an ideal tool in helping promote the decentralized growth of self-reliance at the regional and village level in the developing countries. Properly designed, computer programmes have enormous potential for the developing countries in nearly every field of relevance to their growth and success - including agriculture, industry, transport, natural resources, science, medicine, education, and the management of financial and human resources.
The work we have under way at the UNU's International Institute for Software Technology pays particular attention to software products that speak directly to the kinds of problems faced daily in a developing country - such as management of natural disasters like floods or earthquakes, rural and mobile health care, small business inventories or customs and cargo clearance, or even making the trains run on time. These are essential aids to full participation in the benefits of the information revolution.
Most software companies, however, are located in the affluent nations and they tend to see their markets in affluent terms games, multimedia displays, and the like. Local software development, that speaks to local Third World problems, is therefore badly needed.
UNU/IIST was made possible by financial support from the Governor of Macau and the Governments of Portugal and China. A multipurpose endeavour, it has been characterized by its first Director Dines Bjørner as a cradle, a showroom, a bridge and a channel. As a cradle, the institute nurtures young developing world scientists from the information field. As a showroom, it helps to demonstrate workable approaches to them in various aspects of their field - application modelling, requirements engineering, programming, software engineering and software technology management techniques and tools.
UNU/IIST serves as a bridge between theory and practice, between university and industry, between consumer and producer, and between the developing and the industrialized countries. Finally, it is a channel which, we hope, brings to international attention the little-known achievements of developing countries in software research and technology.
The staff of the Macau institute is not large. Currently, there are 7 academic and technologist staff, 8 administrative and technical people, and some 20 Fellows. UNU/IIST does not award degrees, though many Fellows arrange for their work here to be part of work for postgraduate degrees. Fellows come from partner institutions, and typically spend 9 to 12 months at the Macau centre.
UNU/IIST is involved in a number of activities - including advanced development projects, research, postgraduate and postdoctoral courses, events such as workshops and symposia, and dissemination, primarily to universities and institutes in developing countries.
Advancing Development Know-how
The advanced development projects are concerned mainly with engineering aspects of system construction (UNU/IIST research, by contrast, aims at understanding foundational matters of more sophisticated software technology). The development work is basically cooperative - efforts carried out jointly with industrial companies, government departments, universities or research institutes from developing countries. Since advanced development is one of UNU/IIST's major "showrooms" (to repeat the Bjørner analogy), we intend to devote the bulk of this article to work in this area. Our aim is to show something of the range of real-world concerns on which IIST is focused.
First, however, a note about methodology. The use of mathematical modelling and reasoning about contemplated designs is at the centre of any of the development activities that UNU/IIST promotes. To the laymen, the precise mathematical formulations used can seem a bewildering blur of symbols and numbers. To the software professional, mathematics is a tool for expressing abstraction - and abstraction is the engineer's most important way of capturing the essence of matter before delving into details.
A number of disciplines come into play - including software engineering, programming, management and computational science. Also very important is a relatively new field of study, rigorously adhered to, which is known as "application domain modelling and requirements capture."
Domain analysis involves careful study of the domain or the "environment" in which a given software is to serve. The domain, say, of transport in a study of railway computing systems. Establishing a domain is considered an absolute prerequisite for any meaningful software development. Once the domain is modelled - its essential aspects described and understood - the requirements can also be expressed or "captured" using precise mathematical formulations to state, in, for instance, a rail system, the relationships of rail networks, lines and stations.
Advanced development projects generally focus on software support for infrastructures, activities which support other essential systems: transportation, manufacturing, telecommunications, the financial industry and its services, document processing, and the like. Research into the infrastructure concept is a distinguishing, unique feature of the Macau programme.
Helping China's Trains Run on Time
Software research at IIST has focused on infrastructure concerns in a number of developing countries. In huge, sprawling China, efficient railway utilization is an obvious economic factor. The project there has involved a domain analysis of train dispatching for the Chinese Government. The results are being implemented at 28 train dispatch centres along the 500 kilometers of rail line between the cities of Wuhan and Zhengzhou in northeastern China. Seven Fellows from the Chinese Railways worked on this project in Macau and produced a prototype "Running Map" tool for use by train dispatchers. The project demonstrated that developing country railway software does not always have to come from the industrialized countries.
In another developing country, an UNU/IIST project on Road Management Systems reverse-engineered the functions and behaviour of a tollbooth system for Indonesia's Toll Way System Company. The system is being refined to fit the electro-mechanics of Mitsubishi Electric Corporation's toll way system equipment. A side benefit of the project is the new light it may shed on Asian mega-city traffic congestion problems; it is thereby aiding another UNU project concerned with urban development and sustainable cities.
Viet Nam Budgeting
Elsewhere in Asia, the Vietnamese Government asked UNU/IIST to assist in the architectural design of its budget computerization. Effective economic management can make a big difference to a country's economic performance; computerized information systems are an important useful management tool, ensuring information is accurate, comprehensive and up-to-date. In just over a year, a comprehensive model of Viet Nam's taxation, budget, treasury and external aid and loans was developed by Fellows from the Vietnamese Ministry of Finance, Hanoi's Institute of Information Technology and Hanoi University.
Fellows from Viet Nam Airlines are working on the so-called ABC 2000 project - for Airline Business Computing. The aim is to develop software support for the strategic planning and management of aircraft, route networks, staff, ticket agencies and other facilities and finance. The software will also support scheduling, including timetables and daily operations: ticketing, passenger and freight check-in, gate control, flight dispatch, etc.
Other problems being explored include microwave telephone systems in the Philippines and port management in India. The UNU/IIST umbrella covers a wide variety of problems faced by the developing countries in their efforts to join the modern industrialized sector.
Software for All Scripts
A particularly important computer consideration in many developing societies is software that speaks the local language. The magnitude of that need can be appreciated when one realizes that two out of three citizens of the Third World do not use the Roman script to write their languages. This can give particular headaches to the software developer in Asia where scripts in daily use can read, in addition to English usage, from right to left (Arabic, Hebrew) or from top to bottom (Chinese, Japanese).
The UNU/IIST project known as Multiscript focuses on this linguistic chaos. Increasing international communication generates the need for contracts and other documents in pairs (or even triplicates) of different direction text, and requires dramatically new forms of text systems. The Multiscript projects is prototyping a common software architecture that allows reasonably arbitrary word processing system to cooperate on input, editing, formatting and communication in combination of scripts. Fellows from Mongolia are contributing to the study, research and development of Multiscript as part of their training.
Space does not permit fuller discussion of other UNU/IIST activities in training and curriculum development. Training courses in software development have been organized in many places around the world. Here in Macau, the courses are provided in liaison with the Faculty of Science and Technology at the University of Macau. Elsewhere, we have offered courses to date in Argentina, Brazil, Cameroon, China, Gabon, India, Indonesia, Democratic People's Republic of Korea, Mongolia, the Philippines, Romania, Russia, Thailand, Turkey, Ukraine and Viet Nam.
As the world of the new information technology is wide, so are the topics dealt with by UNU/IIST. Last year, for example, we had a one-day event on "Trading over the Internet," thus predating the intellectual recognition given stock trading in this year's Nobel Prizes. A vast and varied world is out there for those with the right software. As the developing societies enter "cyberspace," they badly need to be able to unlock the potential of the computer themselves and direct it towards the resolution of their own special problems.
Taking the Bus in Minsk ...
To the layman, the software engineer's precise formulations can seem a bewildering blur of symbols and numbers. To the professional, they are a tool for expressing important abstractions. Above, fragment of Minsk Metropolitan Transport System (prepared by Gueorgui Satchok, UNU/IIST Fellow, Minsk, Belarus).