Against All Odds, The Internet in Bangladesh[1]

 

Larry Press, Jamilur Reza Choudhury, Sy Goodman

 

There are few nations in which it has been such an uphill battle, but Bangladesh is on the Internet.  Bangladesh is one of the poorest, most densely populated, least developed countries in the world. A population of 127 million lives in an area roughly the size of Wisconsin, and the GNP per capita is only $270 ($1,330 when adjusted for purchasing-power parity).  Impediments to development include frequent cyclones and floods, the inefficiency of state-owned enterprises, a rapidly growing labor force that cannot be absorbed by agriculture, delays in exploiting natural gas resources, inadequate power supplies, and slow implementation of economic reforms.  Add minimal telephone infrastructure, a high rate of illiteracy, and poverty, and Internet connectivity seems hopeless, but, like a weed popping through a crack in a sidewalk, it is there.

 

This article begins with a description of the Internet in Bangladesh, and then summarizes some of the factors, that have constrained its growth -- factors typical of developing nations.  We conclude with a look at village connectivity, the area in which Bangladesh may become an important Internet innovator.

 

The Internet in Bangladesh

 

The Internet came late to Bangladesh, with UUCP email beginning in 1993 and IP connectivity in 1996.  By July 1997 there were an estimated 5,500 IP and UUCP accounts.  A year later, that number had grown to 13,000, and today there are 25,000 accounts and an estimated 100,000 users.[2]  The number of IP accounts is estimated at 10,000.[3] A number of Cyber cafes have opened recently in Dhaka, and many Bangladeshi have email accounts with Hotmail or other advertiser-supported email services.  One ISP reported that these free accounts make it more difficult to collect service bills since a customer who is unable to pay can shift to another ISP and keep the same email address.

 

Bangladesh has 15 active ISPs.  Larger ISPs connect to the Internet using 64 or 128 KBPS VSATs, and the smaller ones are downstream from those.[4]  International links are obtained through the Bangladesh Telegraph and Telephone Board (BTTB) of the Ministry of Posts and Telecommunications, and, in 1997, a 64-k half-circuit to Singapore cost $58,600 per year.  The BTTB recently began offering Internet service, and they have plans for public-access facilities and a fiber backbone.  There was an opportunity to obtain an undersea cable landing in 1991, but the government did not feel the cost could be justified at that time.  The State Railways have an underutilized fiber network, but have been unwilling thus far to let others share it.

 

There is no .bd top-level domain because the ISPs and other interested parties have been unable to agree on an impartial administrator.  BTTB applied to APNIC, but was turned down as not being representative of the networking community.  (They were not an ISP at the time).  A seven-company ISP Forum of Bangladesh was formed and met with an APNIC delegation in November 1997, but they were unable to agree on a representative organization to manage the .bd domain.  A new organization is being formed now, but this sort of mistrust, coupled with a lack of funds, is reminiscent of the delays in forming network information centers in Africa and Latin America.  In the interim, Bangladeshi register in .com, .org, or .net, and Matrix Information and Directory Services estimates that there are roughly 400 hosts and 30 domains in the country.

 

Universities often pioneer networking in developing nations, but that was not the case in Bangladesh. The reluctance of universities is illustrated by the case of AGNI Systems, an early ISP.  AGNI founder Nawab Kabir offered to establish UUCP service for his university while a business student there.  When they refused, he dropped out to found an ISP, and the university is now a paying customer.  Private firms and NGOs led initial efforts in Bangladesh, and four of the larger ISPs are today affiliated with NGOs.

 

We have used a six-dimensional framework to categorize the state of the Internet in a nation.[5]  Our dimensions are shown in Table 1.  Using this scheme, the Bangladeshi Internet is just beginning to emerge, with a ranking of level 1 on each dimension, Table 2.

 

Constraints on Internet Growth

 

The explanations for this slow start are typical of emerging nations.  Internet growth depends upon telecommunication infrastructure, computing and networking equipment, human resources, and an active, supportive government.[6]  As in many developing nations, the Bangladeshi Internet is hobbled by poor telecommunication infrastructure, a lack of computing and networking equipment, few human resources, and an indifferent, bureaucratic government.

 

The Internet requires underlying telecommunication infrastructure, and a teledensity of .4 phone lines per 100 people (95% of which are in the capital city, Dhaka) is an obvious constraint. For example, in June, 1998 Grameen Cybernet, one of the largest ISPs, had over 4,000 users sharing 64 dial-up lines.  Their central office was saturated, and they stated that if another were added, it too would be immediately saturated. Average waiting time for a telephone is over ten years. This situation is the product of bureaucracy as well as poverty.  BTTB is a government monopoly,[7] and the approval and tender process for a new exchange takes one and a half years.  BTTB also licenses ISP’s satellite links.  Grameen applied for a VSAT license in 1992, but did not get it until 1995.  This delayed their entry in the market to July, 1996. The situation may improve, since an independent Telecom Regulatory Commission is being established and could be operational by mid-1999, and a board has been created to work on the privatization of state-owned enterprises.

 

The Internet also requires computing and networking equipment.  BTTB sub-contracts the purchase of satellite equipment, and contracts for connectivity to Singapore, Hong Kong or the US.  BTTB sets the rate for the outbound half-circuit, but an ISP is free to negotiate the second half circuit and connection charges.  The equipment and uplink subcontracts add an estimated 25% to the networking cost. The situation could be improved if the ISPs shared a common high-speed uplink, and BTTB is considering such a service as well as eventual connection to a marine cable.

 

While the cost of networking equipment is significant in a developing nation, PC scarcity is a more difficult constraint to overcome.  Bangladesh has made recent progress in this area.  In June, 1998, the Government eliminated import duties and value-added tax on computer hardware and software.  As a result, PC prices have dropped dramatically, and the number sold rose from 40,000 in 1997 to 120,000 in 1998.  An 80-90% annual growth in the number of PCs sold is expected for 1999.  Still, only a small fraction of organizations and individuals can afford a computer, so public-access facilities will play an important role.

 

Human resources -- technicians and trained, demanding users -- are also critical.  Networking has not been a priority in Bangladeshi universities, but that may be beginning to change.  In June 1997, the government established a Task Force on Export and appointed a Committee to review the problems and prospects of software export from Bangladesh and formulate recommendations.[8]  They included several recommendations for improving IT education, and the government has placed top priority on human resource development in the IT field.  Today there are around 500 IT graduates per year, but the government target is to produce 10,000 programmers annually by the year 2001. What Bangladeshi IT education lacks in quantity, it makes up in quality.  The team from Bangladesh University of Engineering and Technology (BUET) won their regional competition, and placed 24^th (above many highly reputed US universities) in the ACM Programming contest last year in Atlanta, and they have repeated as regional champions this year.  A university research and education network is also being established.

 

Large numbers of trained, demanding users will be more difficult to achieve than networking technicians.  The small personal computer installed base and poverty in primary and secondary education pose sever impediments to user access and training.  Illiteracy and a dearth of software or content in the Bangla language are also barriers.   While the educated middle class and technicians speak English, pervasive use of the Internet is constrained by the language barrier.

 

To date, the government has been largely indifferent toward the Internet and information technology; however, the commissioning of the Report on Export of Computer Software may signal change.  The report made 45 recommendations, many of which would have a positive effect on the Internet if they were carried out.  In some nations, government caution with respect to the Internet is partially explained by fear that it will undermine national security, cultural values or the political regime.  Bangladesh is a Moslem nation, but not a theocracy, and the fundamentalist party gets only 3-5% of the vote, though their impact is greater than that. The government is aware of religious, political and pornographic censorship, but these issues have not been critical in decisions regarding the Internet.

 

The obstacles in Bangladesh are among the most formidable faced by the Internet in any nation, yet it has taken root and begun to grow.  Bangladesh has been atypical in the degree to which NGOs have been instrumental in establishing the Internet there, but if they are to move to a higher level on any of our dimensions, the government, universities, and industry will also have to become involved.

 

An Opportunity for Innovation

 

The sixth of our Internet-diffusion variables is sophistication of use.  A high score on that variable requires that a nation has gone beyond the substitution of Internet services for others like FAX or telephone.  A highly innovative nation will have developed applications that alter existing organizations, institutions and procedures and will have caused new technology to be developed.  Necessity is the mother of invention, and nations will typically innovate in areas in which they have needs.   Thus the need of the US computer science community to communicate and share resources led to the development of packet-switching networks, and the demands of our consumer economy are generating innovation in electronic commerce.  Nations with unique needs will also develop vertical innovations, as, for example, in the invention of logging technology in the Chilean forests.  Similarly, Bangladesh may innovate in bringing connectivity to rural populations of developing nations.

 

Bangladesh, like many developing nations, has an obvious need to connect its rural population.  They also have an outstanding example of innovation in rural development in the Grameen Bank, www.grameen.com.  Since 1976, Grameen has been offering “microcredit” -- loans of a few hundred dollars -- to the poorest people in Bangladeshi villages.  This has allowed village entrepreneurs (95% women) to increase their income and accumulate capital.  Grameen’s decentralized organization and pragmatic evolution is reminiscent of the Internet, as is their growth rate (see Figures 1 and 2).  In 1976, they began by loaning 10 people a total of $498, and by August 1998 had cumulative loans of $2.44 billion to 2,357,153 members.  Along the way, members had used Grameen loans to build 443,636 houses for their families. 

 

Grameen has been successful in reducing poverty, and is self-sustaining, and replicable.  While initially dependent upon development loans and grants and the energy and charisma of Dr. Muhammad Yunus, the founder, Grameen stopped negotiating for new grants or soft loans in 1995, and has been replicated in 30 nations.  For further information on worldwide microcredit, see http://www.soc.titech.ac.jp/icm/icm.html. 

 

Table 1:  Internet diffusion dimensions

Dimension	Description
Pervasiveness	This is based on the number of hosts and users per capita.
Geographic Dispersion	Over 200 nations now have IP connectivity, but in many of these, access is restricted to one or two large cities.  This dimension measures the concentration of the Internet within a nation, from none or a single city to nationwide availability with points-of-presence or toll free access in all first-tier political subdivisions and pervasive rural access.
Sectoral Absorption	While widespread access is desirable, the payoff is in who uses the Internet in a nation. This dimension assesses the degree of Internet utilization in the education, commercial, health care, and public sectors. These sectors are seen as key to development, and were suggested by the measures used by the United Nations Development Programme Human Development Index.
Connectivity Infrastructure	This measure is based on international and domestic backbone bandwidth, exchange points, and last-mile access methods. A highly rated nation will have high speed domestic and international backbone connectivity, public and bilateral exchange points, and a high proportion of homes with last-mile access using CATV, xDSL, or some other technology that is faster than analog modems.
Organizational Infrastructure	This dimension is based on state of the ISP industry and market conditions. A highly rated nation would have many ISPs and a high degree of openness and competition in both the ISP and telecommunication industries. It would also have collaborative organizations and arrangements like public exchanges, ISP industry associations, and emergency response teams.
Sophistication of Use	This variable ranks usage from conventional to highly sophisticated and driving innovation. A relatively conventional nation would be using the Internet as a straight forward substitute for other communication media like telephone and FAX, whereas in a more advanced nation, applications may result in significant changes in existing processes and practices and may even drive the invention of new technology.

 

 

Table 2:  Dimension values for Bangladesh

 

Dimension	Level	Explanation
Pervasiveness	1	There are very few users, accounts or domains per capita.
Geographic dispersion	1	There is connectivity in only two cities.
Sectoral absorption	1	No sector (government, commercial, education or health) has the 10% adoption rate necessary for a higher rating.
Connectivity infrastructure	1	Total international bandwidth is below T1, and there is no domestic backbone, Internet exchange or high-speed connectivity.
Organizational infrastructure	1	Local organizations are not cooperative.
Sophistication of use	1	The primary applications are email and FAX as substitutes for mail and telephone.  This has not led to qualitative organization or cultural change.