[Paper presented at the annual meeting of the International Society for
the Systems Sciences, Denver, Colo., July 12 to 17, 1992. Copyright
(c) 1992 by the Society.]
AIMING FOR THE ELUSIVE PAYOFF OF USER NETWORKS:
AN NGO PERSPECTIVE
Stephen R. Ruth, Ph.D., Professor of Decision Sciences, George Mason
University, Fairfax, VA 22030;
R. R. Ronkin, Ph.D., Volunteer, Volunteers in Technical Assistance,
Arlington, VA 22209;
Abstract
Electronic transmission of data over long distances, combined with
the development of user networks, has become a powerful enabling
technology -- and a challenging opportunity -- for nongovernmental
organizations (NGOs). Over 100 nations now have some form of
connectivity, but the poorest nations are the most poorly connected.
These least connected nations represent over half of the world's
population. Yet they are desperately in need of the rapid, low unit-cost
information-transfer capabilities offered by this powerful and robust
technology. The numerous NGOs that work in developing areas of the world
have important potential roles in using these high-yield technologies
for disaster relief as well as routine information transfer. They have
a significant but so far largely unfulfilled opportunity to leverage
this technology toward the needs of the poorest nations. This paper
elaborates the scope of the data communication opportunity and gives
examples of NGOs of various sizes and missions in their implementation
of information technology networks. An agenda for action offers specific
suggestions that can lead NGOs to obtain greater leverage from existing
networks.
Keywords: communications electronic-mail international-development
networks NGOs technology-transfer
Introduction: The Proliferation of Electronic Networking
Computer Mediated Communications Systems (CMCS) exemplified by
electronic mail (e-mail), on line data systems, data-base retrieval
systems, etc. are well known, but their immense capacity, low unit cost
and dramatic growth are not fully recognized. In developed parts of the
world, CMCS are used in businesses, in universities, in primary and
secondary education and in the home. The number of messages sent on CMCS
globally -- often as e-mail -- is very large; the rate of increase per
year is already over 30% and increasing [12]. In the United States alone
the volume of traffic per month on NSFNET has increased 100-fold, from
100 million to 10 billion in only four years [3]. A global, online,
database locator at McGill University served 30 inquiries per day in
November 1990; by September 1991 it was serving 2,600. The appeal of
these networks lies in their quick, cheap and relatively error-free
connectivity among millions of sites around the world and a compelling
sense of immediacy. A part of this structure has been used primarily to
connect researchers. Research networks include BITNET in the United
States, EARN in Europe, GULFNET in the Persian Gulf, JANET in the United
Kingdom, NETNORTH in Canada, and SWIFT in Switzerland. Some of them have
operated for over a decade and offer very low unit costs to their users.
Internet, the giant of them all, is a global network of networks with
some 770,000 multi-user nodes and an exponential growth rate.
The reasons for the low unit cost of e-mail are easy to understand.
First, the major tasks of network analysis and planning are already
completed. Thus, investments in software and cadres of specialists at
key node locations have been carefully rationalized over a period of
several years. Stability of services as well as increasingly efficient
system performance are common results [16]. Second, as usage increases,
the large capacity of electronic channels tends to serve increasing
volumes of message traffic, offering the economic advantages of scale.
For example, the BITNET membership fee for a large university (usually
in the range of $5,000) together with an allocation for some user
services personnel may approximate $100,000 per year. Computer costs are
relatively low since the minicomputer that handles the message traffic
is often used for many other tasks too. If this large institution sends
5,000 messages a day, the unit cost is about one cent per message.
Moreover, the saturation level of network users is growing: at Syracuse
University it is over 50% and at least 4,000 professors and students
have network connections.
A third reason for the relatively low cost of e-mail is that it is
so far not viewed by the general public as a service that competes on
the basis of usefulness with telex or fax transmission. Thus, it is not
perceived as a threat to the pricing structure of those technologies. In
actuality, the networks are as dependable as any other type of service.
But most users of fax and telex do not consider e-mail as an equally
acceptable medium of messaging. Rather, it is viewed as a contrivance,
an artifact that is still under development. Studies in Latin America
[17] and Eastern Europe [18] showed that e-mail services were at least
one and often two orders of magnitude lower in unit cost than fax or
telex, at similar levels of service. That is, sending a five-page
message from Prague to New York or Santiago to New York costs perhaps
$.10-$.25 compared to $5 to $15. Moreover, sending e-mail messages is
relatively easy: the sender in Prague uses a few simple keyboard
commands; the file is sent within seconds to New York.
Network Connections: Haves and Have Nots
Unfortunately, the spread of e-mail technology has followed the
path of many other technologies -- wealthy nations are quick to utilize
it and gain benefits, while poor countries must wait. Table 1 gives a
summary of the levels of connectivity of nations of the world,
classified according to the complexity and scope of their e-mail
services [15]. We include under the heading Excellent the countries that
have the capacity to use four major connectivity services: Internet,
UUNET, BITNET and FidoNet. Under Good we include the nations that have
two or three of four target capabilities; countries listed as Poor have
only one. The category No Service includes countries that are unable to
connect using the simplest modalities. About three-fourths of the
countries of the world and over half of its population are characterized
by poor or nonexistent services. It is not surprising that most of the
countries that are poor in the context of telecommunications are also
classified by the UN Human Development Report as "Low Human Development"
nations [20].
Table 1 Quality of Connectivity By Nation
Type of Conectivity Number of Countries Aggregate Population,
Billions (Percent of
Total)
Excellent 31 1.220 (22.4%)
Good 28 1.463 (26.8%)
Poor 48 1.662 (30.5%)
None 125 1.102 (20.3%)
NGOs: Many Types, Many Tasks
People know of the existence of NGOs, although they may not appreciate
what NGOs do that neither government nor business seems to do as well. They
vary greatly in purpose, size, and sponsorship. Many of them are known for
helping the poor and unorganized to improve or take control of their lives.
NGOs are numerous and uncounted: there are an estimated 11,000 in the Latin
America-Caribbean region alone, most of recent origin [4]. A recent European
compilation [19] shows that the NGOs specialize in agricultural issues,
alternative trading, appropriate technology, arts and culture, audiovisual
methods, children, cooperative movements, coordination, consumer issues,
consumer unions, decentralized cooperation, development education, disabled
persons, documentation . . . only through the letter D in the alphabet.
Examples of NGOs, from Local to Global
The Women's Water Committee (Mukusi, Zambia) was started by a group
of women who were dissatisfied with their village's only water supply:
it was inconvenient to access and hazardous to health [14]. They sank
dozens of boreholes and installed reliable handpumps; thousands of
persons lined up to fill their buckets and plastic jerry cans. The
status of women in the village has improved because they manage the new
water system. Now the committee plans to improve sanitation by
eliminating open garbage heaps in favor of burial pits and build
improved, ventilated latrines.
The Junior Engineering and Technical Society (Alexandria, Va.) is a
nonprofit, Nationwide organization for precollege students interested in
engineering and technology. It sponsors student competitions, a National
Engineering Aptitude Search, and engineering design contests. Its paid
staff of three persons produces a newsletter for 35,000 student
subscribers at some 2,200 schools and responds to 10,000 requests for
career guidance each year at no charge.
CARE (New York), a global NGO, has 65 field offices in Latin
America, Asia, East Africa, and West Africa. After Bangladesh's 1991
cyclone, the 1,600-member CARE staff in that country distributed food
supplies to 400,000 people and 55,000 tons of wheat to 2,000,000
farmers. After the Persian Gulf War, CARE set up refugee camps at the
Turkish border and provided vital supplies to 240,000 people. CARE
projects feature primary health care, population and family planning,
agriculture and natural resources, small economic activity development,
training, food assistance, and emergency response.
Using Data Communications-
Some NGO Success Stories
Can NGOs, small and large, use electronics to break down the
barriers of isolation from peers, clients, and patrons? Here are
examples of how NGOs are using modern communications technology.
A woman who leads a conservation data center in Indonesia uses a PC
and a modem to send critical environmental data by e-mail to The Nature
Conservancy and other organizations quickly and at low cost. The Nature
Conservancy, a U.S.-based NGO, is implementing an e-mail network which,
together with its local-area network (LAN), will allow the staff to
share proposals and planning documents with NGOs in other countries to
make better and faster funding decisions.
SatelLife (Boston), a not-for-profit organization, distributes
medical information in sub-Saharan Africa by low-Earth orbit satellite
and e-mail. An international health official says, "Establishing
reliable communications may be one of the most important priorities for
improving health in Africa." [6] Telephone service is poor in quality
and high in cost; sending a 6-page fax from Benin to Nairobi can cost
$300. The Boston office communicates with its African staff through
PeaceNet and FidoNet, both global, nonprofit networks, taking advantage
of the dramatically lower unit costs of this service compared to more
traditional methods of communication. HealthNet, the satellite-based
message service, now operates in Kenya, Tanzania, Uganda, and Zimbabwe;
countries are being added. The New England Journal of Medicine allows
free electronic distribution on HealthNet of articles requested by
African doctors. Electronic distribution of a medical newsletter started
in March 1992 using a variety of technologies. HealthNet's clients are
universities and, through them, doctors and medical societies.
Volunteers in Technical Assistance (Arlington, Va.) has pioneered
the development of satellite-based packet-radio technology for remote
areas of the world. VITA uses an opportunistic mix of communication
methods for its own field projects. Its small-business development
project in Chad (where phone service is poor) operates from two cities
linked by packet radio. A FidoNet link between Arlington and the
capital city of N'Djamena is planned. Elsewhere in Africa, its solar-
energy project in Djibouti loads technical questions onto VITA's
satellite; the questions are referred to Sandia Laboratories in New
Mexico by satellite for quick response. An agricultural infrastructure
project in Afghanistan is headquartered in Peshawar, Pakistan; data are
being exchanged with Arlington by a low-cost, U.S.-initiated modem call,
but a satellite ground station has come on line. As a service, this link
will also carry UN refugee affairs messages between Geneva and the UN
High Commissioner for Refugees in Peshawar via Arlington. Most
significantly for VITA operations, all message traffic is centered on
its bulletin-board system, which is freely accessible to VITA's staff
and Volunteers.
Recording for the Blind (Princeton, N.J.) produces and distributes
audio tapes of current textbooks for cost-free distribution to 27,000
active users, with the help of 4,800 volunteers working in 31 recording
studios throughout the United States. Borrowers typically are students
at the secondary, college, and professional levels. The studios use
modem-to-modem communications to access the central databases at
Princeton and ensure that borrowers' requests are promptly assigned to
studios that have the needed volunteer skills.
The Famine Early Warning System (FEWS) includes state as well as
nongovernment organizations; the main performer is Tulane University in
Louisiana [3]. The purpose of this large, 7-year old effort is to reduce
the risk of famine in an east-west strip of African countries from
Mauritania to Ethiopia and Eritrea. Technical organizations in the
United States and the United Kingdom provide remote-sensing image data,
in digital form, related to cloud cover and precipitation. The data are
interpreted by experts in Africa; the information is then sent to the
United States and translated into recommendations for African farmers,
NGOs, local government agencies, and lending institutions. But people
carry the data on diskettes in both directions. The express company DHL
International carries a diskette to a remote area of the world for $70
with 4-day delivery guaranteed.
The present plans of FEWS to put FidoNet nodes in the targeted
countries can bring dramatic improvements in project effectiveness. For
example, the present ten-day information cycle can be shortened. Costs
will drop and experts located in different targeted countries can easily
consult each other.
Impediments to NGO Use of Networks
In many parts of the world, voice telephone does not work well and
the cost of a postage stamp for international mail is two days of salary
of a university professor. Many NGOs that need telecommunications prefer
the most basic methods of communication: voice phone, fax and telex.
These methods are the most costly and they are not useful for data
transfer: databases, spreadsheets, technical graphics, and word-
processing manuscripts. In view of the savings of money and time, as
well as other advantages, there must be important obstacles to the
greater use of electronic-communications technology by NGOs. Assuming
motivated operators and managers, the principal obstacles to adoption of
better technologies are as follows:
-- Poor local infrastructure, including a broad and variable list of
factors (considered in a later section) that impede the transfer and
endanger the viability of any advanced technology.
-- Inaccurate perception that startup costs are high.
-- Lack of information on how to acquire the technology.
-- Fears that introducing the technology will divert attention from
high-priority tasks, or disturb work flow or staff relations.
-- Inaccurate perception that the technology is not needed.
Gaining the Potential Advantages
Examples of Low-Cost, High-Yield Approaches
Many NGOs already have the equipment they need to get started. The
minimal configuration for realizing benefits includes a PC or other
microcomputer, a modem, and ready access to an enthusiastic, computer-
literate person. The approaches within reach for low-cost data
transmission are usable individually or in combination; they are
summarized as follows:
Approaches requiring telephone line and modem.
-- Modem-to-modem data transmission. At pre-arranged times, a
computer in an industrial country telephones its partner in a developing
country and carries out a two-way exchange of data [10]. The success of
this method is due to the comparatively low cost of originating the
phone call in an industrial country, and to the recent development of
automatic error-checking methods for file transfer over noisy phone
lines.
-- Electronic networks, many of them carrying e-mail, can transmit
data and files. Of special interest are the nonprofit Association for
Progressive Communications, whose global networks provide low-cost
service to 14,000 subscribers in 92 countries, and CGNET Services
International. Many of their subscribers are NGOs. Among networks,
FidoNet deserves special mention because of its low cost, availability,
and effectiveness. It is a noncommercial, telephone-linked network of
computer hobbyists, totally user supported; the world node list is
updated every week [9]. Through "gateways", FidoNet operators can access
the major global networks. Many operators also run bulletin boards,
which may provide public access to global networks for persons who lack
network accounts. The software for setting up a Fidonet node is free or
inexpensive. There are hundreds of thousands of Fidonet nodes in the
world; may with multiple users; in all of Africa outside of South
Africa, there are about 9 nodes.
-- Bulletin board systems (BBSs) are operated by microcomputer
owners. Excellent software for operating BBSs is available free. A BBS
can be dialled on the public telephone system by other computer users.
The caller can read or leave messages, and transfer files to or from the
BBS. Through many BBSs, any computer user can access international E-
mail networks for a fee or no fee; one recent listing shows 116 U.S.-
based BBSs that offer open access [7]. There are relatively few BBSs in
developing countries, but Jensen and Sears [13] report their recent
rapid increase in Africa.
Approaches not requiring a telephone line.
-- Packet-radio systems. These transmit digital information with
extremely high accuracy and often at very low cost. The power of packet
radio to reach remote areas is vastly increased through satellite
technology. A complete, solar-powered ground station (including
microcomputer) for communicating with a satellite now costs about
$7,000.
-- Amateur ("ham") radio technology provides the oldest and best
established public telecommunications network [1]. There are 450,000
hams licensed in the United States and 1,574 licensed in 37 countries of
Subsaharan Africa. Ham radio is probably underused by NGOs.
Agenda for Action
We have described some of the ways NGOs now benefit from the low-
unit cost, high-yield capabilities of the emerging networks. We now
summarize some of the most successful implementations from the
perspective of management support, hardware and software, training and
viability.
Management Support
Since most NGOs are small, focused, and local rather
than global in scope, their leaders must become aware that the steps
needed to connect to a broader group of users are relatively simple,
less expensive than expected, and capable of offering massive leverage
from a small investment. Thus, ways must be found to explain to key NGO
personnel the rudiments of connecting to networks, including many
positive examples of the types we have described. This objective can be
reached in several ways. For example, a larger, successful NGO with
experience and competence in the region, acting as an agent of
technology transfer, sets up training for managers and offers to connect
with smaller NGOs on a phased basis. The more experienced organization
can be nearby or in a distant country. It is crucial to focus on the
specific tasks and needs of each NGO and on assuring that a fully
functioning cadre is left behind. Some nonprofit NGOs specialize in
transferring communication and other information management technologies
to smaller organizations at cost. Examples include Telecommunications
Cooperative Network (Washington, D.C.), TechnoServe (Boston, Mass.)
Massachusetts, and Volunteers in Technical Assistance.
Hardware and Software
Few hardware or software problems in network
connectivity have not been solved in some location or another. There is
a vast amount of information about what works and what doesn't in
hardware, software and network services. For example, a new connection
to Internet and BITNET is being developed by the Escuela Politecnica del
Litoral in Guayaquil, Ecuador, under a grant from the Mellon Foundation
for Value Added Services (VAS). VAS aims to make so much information
available to the key leaders of this polytechnic and its 900 faculty
members that there are few unsolved problems at implementation time. In
Brazil, NGOs buy microcomputers for $1,000 to $2,000 and join Alternex,
a Brazilian network that connects indirectly but quickly and cheaply to
peer organizations in 70 countries [2].
Training and Follow-up
Usually, the most crippling infrastructural shortage in developing
countries is of technically trained persons and technical services. Two
steps are needed. First, training must be supplied by visits or short
courses conducted in the client country or, if abroad, by persons
familiar with conditions in the client country. The second essential
step consists of follow-up activities that address local technical
problems as they arise, with a short response time, and provide
continuing advice and encouragement. The follow-up activities should be
planned so that they do not foster dependence on the donor.
Forming Partnerships
In addition to training, there are often shortages of equipment,
foreign exchange, literature resources, and planning tools. Other
impediments may include slow or uncooperative customs procedures, bad
telephones, a poor dealer network, and failure of new software and
hardware to perform as expected. It may happen that local policy makers,
unaware of the high return from investment in communications, cannot be
persuaded to support it, or even to license modems [11]. Many
infrastructural deficits take time to overcome, but are often best
addressed by forming an enduring, voluntary, professional partnership
with a peer in another country. The result of this arrangement is that
both sides share some of the infrastructural resources of the country
that is better off. There is often a two-way flow of technical
information that benefits both sides. Once low-cost, international
electronic messaging is launched, it becomes easy to join international
groups of peers who share interests through electronic messages and
documents, and can help each other. On BITNET alone, there are more than
2,000 such groups.
* * *
Space does not allow us to thank individually the many persons consulted
during preparation of this paper. The references listed below exclude
annual reports and other organizational prospectuses.
[1] American Radio Relay League, 1991. ARRL Handbook for Radio Amateurs:
1992. 69th ed. Newington, Conn.: ARRL. M. Hertel (ed.), 1989, Radio
Amateur Callbook, International Listings Exclusive of North America and
Hawaii. Lake Bluff (Ill.): Radio Amateur Callbook.
[2] S. Annis, 1991, "Giving Voice to the Poor." Foreign Policy Vol. 84
(Fall 1991), pp. 93-106.
[3] E. Aupperte, 1992, "USA NSF Backbone," Internet Society News Vol. 1,
No. 1, p. 6
[4] A. Bebbington, 1991. "Sharecropping Agricultural Development, the
Potential for GSO-Government Cooperation." Grassroots Development Vol.
15, No. 2, pp. 21-30. Diane B. Bendahmane, 1991, "Performance Review
for NGOs," loc. cit., pp. 31-37. Inter-American Foundation, 1991, A
Guide to NGO Directories; How to Find Over 11,000 Nongovernmental
Organizations in Latin America and The Caribbean. Rosslyn, Va.: Inter-
American Foundation.
[5] R. Barad, 1992, Applications for FidoNet Technology in Famine Early
Warning: Tests Conducted 27-31 Jan uary 1992, prepared for the Tulane /
Pragma Group, Washington, D.C.: US Agency for International Development.
US Agency for International Development, 1991, Famine Early Warning
System, Project Paper, Amendment No. 2, No. 698-0466, 6 September 1991,
Washington, D.C.: U.S. AID.
[6] C. Clements, 1992. "HealthNet connects Africa to vital medical
data." Satellite Communications, 1992 (January): pp. 18-21. R. Pool,
1991, "Opening Up Communications." Nature Vol. 353, p. 293.
[7] Diskmuncher [pseud.], , 1992, RE: Publicly
Accessable [sic] Internet Sites. Electronic posting on Internet,
February 14.
[8] D.P. Dern, 1992, "Applying the Internet, Corporate, Research,
Educational, Governmental, and Other Real-World Issues." Byte 1992
(Feb.) pp. 111-114, 118.
[9] Fido Software, 1992, FidoNet Nodelist ... A Listing of the Systems
Within FidoNet. San Francisco: Fido Software; electronic posting on
FidoNet, March 20.
[10] G.L. Garriott, 1991, "Africa: Better Data Communications,"
DevelopNet News (electronic publ. ) Vol. 1, No. 4 (July).
[11] G.L. Garriott, 1991, "Packet Radio in Earth and Space Environments
For Relief and Development." Paper presented at the 34th Annual Meeting
of the African Studies Association, St. Louis, Mo., November 23-26,
1991. Anon., 1992, "Four New Ground Stations Talk to VITA's Satellite,"
DevelopNet News (electronic publ. ) Vol. 2, No. 3 (March).
[12] V. Gurbaxani, 1990, "Diffusion in Computing Networks: The Case of
BITNET", Communications of the Association for Computing Machinery Vol.
33, No. 12 (December): pp. 65-75.
[13] M. Jensen, and G. Sears, 1991, "Low Cost Global Electronic
Communications Networks for Africa." Paper presented at the 34th Annual
Meeting of the African Studies Association, St. Louis, Mo., November 23-
26, 1991.
[14] D. Kinley, 1991, "Zambia: Women Organize for Safe Water," UNDP
Source (June) pp. 6, 7, 9.
[15] L. Landweber, 1992, "Internet Connectivity Report", April 18, 1992,
Reston, Va.: Internet Society.
[16] E. Puliatti, 1990, "Computer-Mediated Communications Systems and
Developing Countries," Development Vol. 1, No. 2: pp. 60-65.
[17] S. Ruth, F. Utreras and R. Brescia, 1990, "The South American
Network Within One Year: An Attainable, Low Cost High Yield Reality,"
Interciencia Vol. 15, pp. 259-264.
[18] S. Ruth, 1992, "Using Powerful, Low Cost High Yield Academic
Networks in Developing Nations: A Value Added Services Perspective,"
International Information Systems Vol. 1, No. 3, in publication.
[19] Unesco, 1991, Guide to the Information Activities of European
Development Networks, 1st ed. Paris: Unesco.
[20] UN Development Programme, 1990, Human Development Report, New York:
Oxford University Press, p. 185.