How Open Source Software and Wireless Networks are Transforming ...

The International

JOURNAL

of

TECHNOLOGY, KNOWLEDGE & SOCIETY

Volume 4, Number 6

How Open Source Software and Wireless Networks are Transforming Two Cultures: An Investigation in Urban North America and Rural Africa David J. Yates, Thomas McGonagle and Anas Tawileh

www.technology-journal.com

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY http://www.Technology-Journal.com First published in 2008 in Melbourne, Australia by Common Ground Publishing Pty Ltd www.CommonGroundPublishing.com. © 2008 (individual papers), the author(s) © 2008 (selection and editorial matter) Common Ground Authors are responsible for the accuracy of citations, quotations, diagrams, tables and maps. All rights reserved. Apart from fair use for the purposes of study, research, criticism or review as permitted under the Copyright Act (Australia), no part of this work may be reproduced without written permission from the publisher. For permissions and other inquiries, please contact . ISSN: 1832-3669 Publisher Site: http://www.Technology-Journal.com THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY is a peer refereed journal. Full papers submitted for publication are refereed by Associate Editors through anonymous referee processes. Typeset in Common Ground Markup Language using CGCreator multichannel typesetting system http://www.CommonGroundSoftware.com.

How Open Source Software and Wireless Networks are Transforming Two Cultures: An Investigation in Urban North America and Rural Africa David J. Yates, Bentley University, Massachusetts, USA Thomas McGonagle, Benjamin Franklin Institute of Technology, Massachusetts, USA Anas Tawileh, Cardiff University, Wales, UK Abstract: This paper explores how open source software and wireless networks enable digital inclusion in the United States and Africa. We begin by measuring the digital divide in these very different regions of the world. Our research demonstrates, both quantitatively and qualitatively, how the digital divide places populations in both regions at a disadvantage. Next we examine the role of these technologies in bridging the digital divide along three complementary dimensions. First, we show that both affordable technology and sound policy are necessary for digital inclusion. Second, we look at how these two technologies are extended, integrated, and customized in information and communication technology (ICT) solutions that are both creative and effective. Third, we describe how the hardware and software in networked systems have been tailored to support applications that are as diverse as the people using them. We also present a series of case studies that highlight specific wireless network and open source technologies and their impact on the education of children, as well as the development of local communities. Keywords: Open Source Software (OSS, FOSS, FLOSS), Wireless Networks, Mesh Networks, Digital Divide, Technology and Policy, Digital Inclusion

Introduction HIS PAPER BEGINS by exploring the digital divide in both the United States and Africa. We analyze how the digital divide can be measured within these very different regions of the world. Because of the contrast in the state of information and communication technology (ICT) in the United States and Africa, it is appropriate to measure the digital divide using different metrics. However, it is important to recognize that some metrics are common to both regions (e.g., percentage of households that have a computer or Internet access). Through both quantitative and qualitative observations, we see how the digital divide disempowers populations in both regions. The main contribution of this paper is an exploration of how open source software and wireless networks enable digital inclusion in the U.S. and Africa. We explore the role of these technologies in bridging the digital divide along three complementary dimensions. First, we show that both affordable technology and sound policy are necessary for digital inclusion. Second, we look at how these two technologies are extended, integrated, and customized in ICT solutions that are both creative and effective. Third, we describe how the hardware and software in networked systems have been tailored to support applic-

T

ations that are as diverse as the people using them. Finally, we present a series of case studies that highlight specific wireless network and open source technologies and their impact on the education of children, as well as the development of local communities.

The Digital Divide Information and communication technology (ICT) is a vital resource in our society. In the United States, more people are going online to conduct day-to-day activities such as education, business transactions, personal correspondence, research, and information gathering. An October 2000 U.S. government report entitled “Falling Through the Net: Toward Digital Inclusion” describes the increasing importance of digital inclusion: Each year, being digitally connected becomes ever more critical to economic and educational advancement and community participation. Now that a large number of Americans regularly use the Internet to conduct daily activities, people who lack access to these tools are at a growing disadvantage. Therefore, raising the level of digital inclusion by increasing the number of

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4, NUMBER 6, 2008 http://www.Technology-Journal.com, ISSN 1832-3669 © Common Ground, David J. Yates, Thomas McGonagle, Anas Tawileh, All Rights Reserved, Permissions: [email protected]

146

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

Americans using the technology tools of the digital age is a vitally important national goal. Although this quote is from a U.S. government report [NTIA00], the need for digital inclusion is clearly a global issue. In December 2003, the World Summit on the Information Society (WSIS) expressed the common desire and commitment to build: … a people-centered, inclusive and development-oriented Information Society, where everyone can create, access, utilize and share information and knowledge, enabling individuals, communities and peoples to achieve their full potential in promoting their sustainable development and improving their quality of life, premised on the purposes and principles of the Charter of the United Nations and respecting fully and upholding the Universal Declaration of Human Rights. [WSIS03]

These quotes emphasize the importance of addressing the digital divide in both the U.S. and Africa. To gain an understanding of the digital divide in these regions, we quantify it state-by-state in the United States and country-by-country in Africa.

Measuring the Digital Divide Figure 1 shows the adoption of information and communication technologies by U.S. households from 1997 to 2003 [NTIA04]. There are several interesting facts that are highlighted by this figure. During the six-year period shown, the percentage of households that owned a computer rose from 36.6% to 61.8%. At the same time, the percentage of households that augmented their computer with Internet access jumped from just over 50% (18.6% of households in 1997) to over 88% (54.6% of households in 2003). This is an almost threefold increase in the percentage of households with Internet access, reflecting the increased utility of the Internet in the daily activities of U.S. residents.

Figure 1: Adoption of ICT by U.S. Households from 1997 to 2003 Although the trends in Figure 1 show increased adoption of ICT by U.S. households, we see troubling signs of the digital divide when we break down the data gathered for the two most recent samples (i.e., in September 2001 and October 2003). We first examine these data state-by-state, and then by family income and race. Figure 2 shows the percentage of each state’s population that had access to the Internet in September 2001 (on the left) and October 2003 (on the right). During this two-year period the number of states that had 60% or more Internet users within their population increased from 24 to 31. However, in 2003, more than a decade after the Internet was generally available in the U.S., 19 of the

50 states had 40% of their population without access to the Internet. Table 1 shows demographic data taken at the same time as the data presented in Figures 1 and 2. We summarize these data first by family income, and second by race of head of household. Table 1 again shows that although availability of ICT to the population at large is improving over time, there are several disadvantaged groups for which this is not true. For example, in October 2003 only 37-46% of Blacks and Hispanics in the United States had access to the Internet. This is significantly less that the 63 (or more) percent of Internet users found among other races. The digital divide is even greater if we view

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

these same data across family incomes. In October 2003, only 31.2% of individuals living in families that earned less than $15,000 per year had access to the Internet. An individual in a family earning

$75,000 or more in 2003 was 2.65 times more likely to be able to go on-line easily (i.e., 82% or more of these individuals had Internet access).

Figure 2: Distribution of Internet Use across the States in 2001 and 2003 (Percentage of State Population Age 3 and Over) Table 1: Internet Use in the United States by Individuals Age 3 and Over [NTIA04] Internet Users (Percent) Total Population

Live in a Broadband Household (Percent)

September 2001

October 2003

October 2003

55.1

58.7

22.8

Race / Ethnicity of Head of Household White

61.3

65.1

25.7

Black

41.1

42.2 – 45.6

13.9 – 14.2

Asian American & Pacific Is- 62.5 lander

63.0 – 63.1

34.2 – 34.7

Hispanic (of any race)

33.4

37.2

12.6

Less than $15,000

25.9

31.2

7.5

$15,000 - $24,999

34.4

38.0

9.3

$25,000 - $34,999

45.3

48.9

13.4

$35,000 - $49,999

58.3

62.1

19.0

$50,000 - $74,999

68.9

71.8

27.9

$75,000 and above

80.4

82.9 – 86.1

45.4 – 57.7

Family Income

Clearly the data in Figure 2 and Table 1 illustrate the digital divide in the United States, and show some of the populations that are disempowered by its impact. A specific example of this is captured by an on-line quote about the Wireless Philadelphia project:

Fewer than half of the roughly 650,000 households in the City of Philadelphia have any Internet access at all, and more than 70% of those that are connected still use dial-up. This presents tremendous obstacles for low-income

147

148

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

families. One cannot apply for even entry-level jobs in any major hospital or hotel in Philadelphia without Internet access, and over 80% of middle and high-school aged students across the country have homework every night that requires them to use the Internet. [Nico08] An important part of addressing the digital divide is empowering people by educating them and giving them ongoing access to information [Norr01]. Our discussion of the digital divide in the United States has therefore mostly focused on Internet access. This is because we know that the majority of people in the U.S. have access to more basic infrastructure such as electric power and telephone service. Unfortunately, the digital divide in Africa is a much larger problem, and therefore requires us to view it through a wider lens. We begin, however, by a meaningful comparison that sets the stage for the contrast between the digital divide in the United States and Africa. Figure 3 presents a world map that shows the percentage of the population in each country that had Internet access in 2002 [UCSC04]. In this figure, the U.S. is among the countries that had the most available Internet access. More than 45% of the U.S. population had Internet access at this time. This fact is confirmed by the data in Table 1, which suggests that the actual number is between 55.1% and 58.7%. Sadly, most countries in Africa had less than 1 person in 100 that were Internet users at this time. This is

true for almost all counties found in the poorest regions of Africa, i.e., central and eastern Africa. Only a handful of countries had populations in which more than 5% of the people had access to the Internet. Two such countries were Tunisia and South Africa. Another measure of the Internet digital divide is how much bandwidth is delivered to each country. Figure 4 presents per capita GDP (as shades of blue) and per capita bandwidth (in bits per second per person) for African countries in 2002 [IDRC03]. Consider what the information in Figure 4 tells us about the digital divide within Africa. Looking at the “bits per capita” pie charts associated with each country on the mainland, we see that only a select few have more than 5 bits per second (bps) per person, e.g., Tunisia, Gabon, and South Africa. At the other extreme, we see that many countries have an order of magnitude less bandwidth per person. These include Niger, Chad, and Sudan in central Africa, as well as Somalia, Tanzania, and Mozambique in east Africa. Now compare the data that we can derive from Figure 1 with the data in Figure 4. In 2002, conservative bandwidth estimates of the U.S. Internet backbone averaged about 40,000 bps per connected household. With approximately half the nation online, and approximately four people per household, the per capita bandwidth was at least 5,000 bps per person. Thus, at this time, the average person in the United States was at least three orders of magnitude richer in terms of Internet bandwidth than the people living in the vast majority of countries in Africa.

Figure 3: Distribution of Internet Use across Countries in 2002 [UCSC04]

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

Figure 4: Per Capita GDP and Per Capita Internet Bandwidth for African Countries in 2002 At the time the data in Figure 4 was being gathered and analyzed, an understanding of the need to address the global digital divide was emerging. National governments and international organizations began to call for information and communication technology for development (abbreviated ICT4D). This drew public attention in November 2002, when Kofi Annan, the Secretary General of the United Nations at the time, made the following public statement: “We need to think of ways to bring wireless fidelity (Wi-Fi) applications to the developing world, so as to make use of unlicensed radio spectrum to deliver cheap and fast Internet access.” This quote is from a speech that Kofi Annan made as part of a challenge to Silicon Valley (and the United States as a whole)

to develop and contribute to ICT4D in less fortunate countries [Anna02]. Later in this paper, we explore the role of Wi-Fi and other wireless network technologies in ICT4D. Beginning in 2002, the International Telecommunications Union (ITU) and United Nations (UN) spent more than two years working to measure the extent of the global digital divide. The result of this work was a measure of the digital opportunity in each country of the world, and an assessment of the so-called digital opportunity index based on data available in 2004 [ITU05]. Figure 5 shows the digital opportunity index (DOI) for most global regions, as well as the increase in the DOI from 2004 to 2006 [ITU07].

Figure 5: Average DOI by Region and DOI Growth by Year for 2004-2006

149

150

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

First consider the average DOI in 2006 by region, shown on the left hand side of Figure 5 [ITU07]. The highest regional DOI of 0.58 is found in Europe for this year. In North, Central and South America, the regional DOI is 0.45. The lowest regional DOI by far is in Africa, which has a DOI of 0.22, 45% lower than the world-wide average. The data in the right hand side of Figure 5 paints a similar picture of the world in 2004. In other words, a DOI of 0.51 in Europe during this year was still the highest regional average. Similarly, a DOI of 0.18 in Africa was still the lowest regional average. Also, the DOI in the Americas was slightly higher than the world-wide average of 0.35. Note also that regional digital divides persist in spite of positive global trends. In this section we have measured and analyzed the digital divide in the United States and Africa. We have even seen aggregate data that suggests that “digital opportunity” is increasing in the United States (e.g., as shown in Figure 1) as well as Africa (e.g., as in Figure 5), even though a digital divide persists in both regions of the world. We now look at policies in the United States and Africa that move disempowered populations towards digital inclusion. We then examine how wireless networks and open source software enable the same.

Towards Digital Inclusion Digital inclusion is important to residents of many countries in several areas of their lives. Perhaps most easily measured are the economic benefits. However, to the extent that digital inclusion educates, informs

and empowers individuals, it also improves their participation in their family, community, and government [Norr01]. For digital inclusion to be most effective it should consider all individuals, including those in disadvantaged populations. Doing so ensures the greatest impact on the economy and community, both in the present and the short term, as well as in the future. Thus, most successful policies and practices in the United States and Africa have addressed digital inclusion for men and women, for children and the elderly, as well as for minorities and the poor [FCC06, Huds06]. One of the most visible digital inclusion projects in the U.S. in recent years is Wireless Philadelphia, a project with the mission to “transform Philadelphia’s neighborhoods by making high-speed Internet access more available and affordable through Digital Inclusion – the initiative that helps people who are not online gain access with hardware, software, tech support/information, and broadband Internet service, so they can begin to use this technology to improve their educational, employment, health and life opportunities.” This project is managed by a non-profit corporation of the same name (see http://www.wirelessphiladelphia.org/). Figure 6 shows a map of the coverage for wireless Internet access in early 2008 [WPC08]. As of January 2008 approximately 80% of Philadelphia had Wi-Fi-based service available. Since this time, ownership and operation of the Wireless Philadelphia network has been transferred from Earthlink to the Network Acquistion Company.

Figure 6: Map of Wireless Philadelphia Coverage

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

There are many interesting aspects to the Wireless Philadelphia project. From a policy perspective, the City of Philadelphia has worked to serve its low-income population. The digital inclusion program that offers qualified Philadelphia residents Internet access has a discounted rate of less than $10 per month. In order to be eligible for this rate, customers can have income up to 150% of the federal poverty level, or participate in certain supportive programs such as Transitional Assistance to Needy Families (TANF), Medicaid, or Supplemental Security Income [WPB07]. Wireless Philadelphia has also added programs in two important areas: 1) accepting and refurbishing donated computer and networking equipment for individuals and families that need it; and 2) training users to operate computer hardware, computer applications, networking gear, and operating system software. From a technology perspective, Wireless Philadelphia has demonstrated a proof of concept for large-scale municipal wireless network access and wireless mesh networking. Wireless mesh networking is an important emerging technology since it both distributes and scales network access. Finally, as Wireless Philadelphia goes live (as we write this paper) it has validated that in the U.S., the cost to roll out and begin operating a municipal wireless network in an urban area exceeds $100,000 per square mile [HSA05]. The Wireless Philadelphia project and similar projects elsewhere have also triggered policies that

may be unfavorable towards digital inclusion. The threat of low cost municipal wireless caused the telecommunications and cable industries to seek legislation in Pennsylvania to keep government out of the wireless Internet access business. A communitybased effort was able to obtain an exemption that let Philadelphia’s plan proceed, but most other parts of the state of Pennsylvania are blocked from developing similar networks. Today there are more than 20 states that have legislation that defines the extent to which municipalities may provide wireless Internet access. Some of this legislation has proposed to define or restrict municipalities’ ability to provide wireless Internet service. For example, some bills require municipalities to undertake feasibility studies, long-term cost-benefit analyses, public hearings, or referendums [LSG06, TMS06]. Other bills go as far as prohibiting municipal wireless Internet provisioning altogether [Hend08]. It remains to be seen how the dynamic between public and private wireless Internet access in the United States plays out. Meanwhile, the Personal Telco Project lists more than 50 community or municipal wireless networks in the U.S. and Canada [Pers07]. Just as there is a digital divide in the availability of information and communication technology between the U.S. and Africa, there is also a “policy divide” when considering ICT. Table 2 shows that the development of ICT policies by African countries has escalated in importance in recent years [FI07].

Table 2: ICT Policy Development by African Countries from 2000 to 2007 National ICT Policy Development by Country

2000

2005

2007

Policy in place

13

28

36

Policy under development

10

15

12

No development underway

30

10

5

Total

53

53

53

In 2000, only 23 countries within Africa had an ICT policy either in place or under development. Fortunately this situation improved by 2007. At this time 48 of 53 nations had a policy either in place or under development. Sadly, five countries still had neither.

Wireless Networking Affordable technology and sound policy are both critical to advancing digital inclusion. In this section of our paper we show how not one, but two wireless “revolutions” have enabled affordable ICT in the U.S. and ICT4D in Africa. In the next section, we show how open source software, most of which is free, has done the same. Affordable wireless networking technologies have had an enormous impact on societies in both the

United States and Africa. In the last decade, cellular telephones (which operate in radio spectrum licensed by the FCC in the U.S.) have gone from being a luxury item to a commodity item, although they are still not affordable for poorer families. Similarly, wireless local area networks or Wi-Fi networks (which operate in unlicensed spectrum), have empowered many businesses and households to expand their access to the Internet. In Africa, cellular telephones have also been sold in great enough numbers that they are giving many individuals and families easier access to communication with others. The impact of cellular telephones in Africa has been credited with helping bridge the digital divide and drive local economies [Econ05]. Wi-Fi has also played a critical role in providing Internet access in

151

152

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

Africa. Not only does Wi-Fi provide access to clusters of people on the same local area network (LAN), but the same technology has been adapted to build metropolitan and even wide area networks (MANs and WANs). Extending the reach of networks by building these long-haul wireless networks has given many Africans access to the Internet for the first time, often in community telecentres [JE01, Hoe06, UN04]. Perhaps the most important factor in making wireless networking affordable is the effective development of open international standards in two areas. The first area is wide area cellular networks. The second is wireless LANs. Even though the revolution in cellular telecommunication began more than two decades ago, it continues today as we begin taking advantage of third-generation (3G) technologies in the latest “smart phones.” Likewise, within the suite of Institute of Electrical and Electronics Engineers (IEEE) 802.11 wireless LAN standards, rapid innovation is still occurring in technologies that take advantage of unlicensed radio spectrum. IEEE 802.11a/b/g wireless networks are state-of-the-art today. However, 802.11n is increasing the bandwidth of wireless network nodes in 2008. Not far behind this standard, IEEE 802.11s is likely to offer the first commercially available open standard for wireless mesh networking. Even though wireless LAN standards have been developed more recently, they have had a significant impact. For example, the Wireless Philadelphia project implements an IEEE 802.11b-based wireless mesh network that uses unlicensed spectrum to distribute and scale Internet access over 100+ square miles in Philadelphia. The One Laptop Per Child (OLPC) laptop implements a pre-standard version of IEEE 802.11s so that XO laptops within a school can form a mesh network, allowing children to share resources, ideas, content, etc. with each other [Raws06].

Open Source Software Open source software (OSS) has also enabled affordable information and communication technology in the U.S. and Africa. OSS is eminently affordable since it typically can be distributed and used without license fees. This software is developed by contributors from all over the world, and is released under license agreements that enable users to access, study, and modify the source code. These characteristics ensure that open source software plays an important role as a catalyst for information-driven development and change [Webe04]. By capitalizing on the vast amount of available open source software, development projects can leverage their resources and significantly increase their impact [Heek08]. The model

proposed by Gerster and Zimmermann for the utilization of information and communication technologies for poverty reduction also acknowledges the potential of open source to create local value, stimulate local language, reduce costs, and improve security [GZ03]. The philosophical concept of openness has been extended beyond software to include areas as diverse as open content, open education, open access, and even open hardware. These factors have made technology more affordable and information more available to development initiatives, and spawned an increasing number of ICT4D projects all over the world [BDS03]. One important use of OSS that impacts all of us is the use of web servers on the Internet. Most web servers today run an open source web server called Apache [CB07]. Furthermore, Apache often runs on an open source operating system such as Linux [Siev05]. So even if you do not use Firefox to browse the Internet or OpenOffice to create documents, you probably “touch” open source software as you access most content on the Internet. Since Microsoft and other commercial software companies consider OSS a threat, they have worked hard to market and sell their own products where they see OSS success stories. Most of these OSS success stories represent grass roots efforts to address a particular issue. For example, Brandon Elementary School in Atlanta, Georgia, found that as their computers aged, they became slow, frequently frozen, and “fraught with hardware, software, and malware issues” [Gasp06]. Since a Linux operating system is better able to be installed and run on older machines with slower processors and less memory, this school district evaluated OSS solutions for their computers. Parents and administrators worked together on this issue, and found readily available K-12 Linux Terminal Server Project (K12LTSP) software to be the best choice. With a proper migration strategy, this software transforms old hardware into thin clients that boot off the network without hard drives. A thin client-based network architecture has the added benefit of centralizing administration of client systems on a small number of servers (sometimes just one server). OSS success stories like the one at Brandon Elementary have encouraged larger and more ambitious projects. In 2007 Atlanta Public Schools completed a pilot project that installed over 2,000 computers in seven of their schools. This ICT roll-out was also made possible by employing Linux thin client computers that run OSS from centralized servers. The “book-sized” thin clients are compact and efficient laptops, with no moving parts like disk drives or fans that wear out. These computers provide Internet access, plus hundreds of other open source educational applications, to more than 4,000 Atlanta elementary

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

and middle school students. The 2,000+ machines in the seven pilot schools are run using just 31 servers, including one data server at each school which stores working files that can be securely shared among students and teachers. The educational applications include Internet and office tools like web browsers, word processors, spreadsheets, and presentation builders. However, the client computers also provide users with free access to hundreds of other powerful packages that do everything from graphics and publishing, to programming, drafting and 3D animation. It is expected that using OSS will help bridge the digital divide in Atlanta by giving more students better access to the Internet as well as educational applications. The need for affordable ICT has motivated even larger scale open source projects in Africa. One of the most successful of these projects is the tuXlab project, which focuses on schools in South Africa. The vision of this project is to develop and maintain an ICT model that yields replicable solutions for schools, which can be extended into other community development programs. The tuXlab project began in 2002 and six years later consists of over 200 school community partnerships throughout South Africa, serving over 160,000 students and teachers. There are many factors that have been critical to the success of the tuXlab deployments; however, we focus on two important factors: local content and local languages. As in most countries, educational material in South Africa is usually locally developed and proprietary in nature. It is therefore important that the tuXlab software base include a foundation on top of which content management software and customizable educational content can be layered. The tuXlab software includes a range of language options provided by the translate.org.za project (see http://translate.org.za). This means that many of the applications included in tuXlab are available in Afrikaans, Xhosa, and Zulu. The tuXlab software distribution continues to add applications and languages as demand arises and the project evolves. Even though local languages are used whenever possible, some high-quality educational material is available in English first. For example, tuXlab includes a mini-Wikipedia distribution compiled by the SOS Children’s project. The articles in this compilation are adapted from the main Wikipedia project and represent a 15 volume encyclopedia with 24,000 pictures, 14 million words, and articles on 4,625 topics. We turn our attention from schools to communities in the two sections of the paper that follow. The first section is a detailed case study on a wireless mesh network serving residents of a low-income housing complex in Boston, Massachusetts. The second section describes the features and impact of a telecentre

and radio station in Karagwe, Tanzania that together serve over 400,000 people.

Castle Square Tenants Organization, Boston, Massachusetts The Castle Square Tenants Organization (CSTO) located in the South End of Boston has built a large wireless mesh network that provides broadband Internet access to 500 families. The project uses inexpensive consumer Internet connections, commodity Netgear access points, an OpenWRT Linux distribution (http://openwrt.org), and an open source mesh networking software developed at MIT called Roofnet [BABM05]. This combination of high-speed Internet bandwidth coupled with low cost hardware and free OSS makes the network useful and resilient, and the project affordable. The CSTO network is maintained and supported through a mixture of volunteerism, community activism, and stipends for local college students. This effort is led by Gabriel Fishman, who is a CTC Americorps VISTA member. In addition to keeping the network running, Gabriel and his small staff refurbish the hardware and software inside computers (including donated equipment), and train the residents of Castle Square on computer and Internet use. The original rollout of the network was enabled by Stephen Ronan, one of Boston’s most prolific wireless volunteers. Steve spends most of his time building and managing wireless networks in housing developments in Boston and Cambridge. The decision to deploy a wireless mesh network at the CSTO instead of a conventional wireless LAN is significant. Each access point in a wireless access network requires a connection to an Ethernet switch (or an IP router). Such a network design would require Ethernet cabling to all 500 apartments in Castle Square. A wireless link in a mesh network penetrates walls and connects devices on the same radio frequency without cabling. A wireless mesh network can form an interconnected topology of network nodes and organize traffic flows so that data is directed to and from the Internet properly. This is accomplished by the Roofnet mesh software, which automatically creates a map of which nodes are connected to an Internet connection, and how packets should traverse intermediate nodes to get to an Internet-connected node. This wireless mesh network allows the CSTO to provide Internet access to all residents using only two high-speed Internet connections. Another major advantage of the Roofnet software is the minimal configuration and administration that it requires. Wireless mesh networks in general, and Roofnet deployments in particular, determine the network topology and configure each node as it

153

154

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

powers on. The main tasks in planning and designing the wireless network at the CSTO were choosing where to physically locate the nodes, mounting and powering them, and ensuring that the radio transmission between enough nodes worked. Every time a node is added it discovers its neighbors, adds itself to the network, and learns one or more routes to the Internet. There is no need to assign IP addresses, use directional antennas, or require other network users to reconfigure or change their equipment for a new node to join the mesh network. Routine failures such as loss of power, radio interference, and ISP problems can for the most part, be routed around. The CSTO network shows that low cost hardware, free open source software, and surplus IEEE 802.11 bandwidth are all that is required to build a community wireless mesh network. To learn more about the Roofnet implementation visit http://pdos.csail . mit.edu/roofnet/doku.php.

FADECO Telecentre and FM Radio Station, Karagwe, Tanzania FADECO is a small center that provides computer training, Internet service, and radio broadcasting to local communities in Karagwe, Tanzania, one of the most remote locations in the country. The center is making tremendous efforts to bridge the digital divide for 428,000 people in this rural part of the African continent through several initiatives and activities. These include an education program, a resource telecentre, and HIV/AIDS awareness activities. In its pursuit of leveraging the limited resources available and maximizing impact on people’s lives, FADECO focuses on the consolidation of its information resources and on implementing the most effective method for content distribution. Internet connectivity is crucial for the center to be able to achieve its purpose and to deliver its services to the local community. FADECO taps into the resources available on the web to collect information that benefits the local community, and makes it available in alternative formats to reach the largest possible audience. Additionally, the Internet simplifies the distribution of locally produced content to other parts of the continent that speak Swahili, the official language in Karagwe. The cost of Internet connectivity, which includes ongoing monthly subscriptions, electricity charges and staffing, in addition to the fixed cost of the expensive equipment required, is a substantial burden on the center. In order to afford the high cost of Internet access, FADECO developed innovative solutions that were feasible given the strict resource limitations under which the center operates. FADECO has a VSAT Internet connection provided by SATCOM based in Darussalam. The

VSAT provides 128 Kbps of bandwidth downstream and 64 Kbps upstream via a Ku-band connection. This link costs $195 per month. The center has decided that the best option to sustain its Internet connectivity is through cost sharing with other institutions. In this scenario, FADECO splits the burden of the Internet connection costs by giving others the right to use the connection in exchange for a share in the cost. A significant challenge faced by FADECO in sharing its Internet connection was the sparse geographical distribution of potential clients in Karagwe. The locations of client sites were hundreds of meters away from FADECO’s main building. Installing appropriate cabling was virtually impossible due to the nature of the terrain and the prohibitively expensive cost of fiber-optic cables in the region. Wireless networking was considered the best solution to address these issues. Using commodity hardware and locally assembled antennas and masts, FADECO was able to install two dedicated connections to remote sites. The first connection (over 12 kilometers of terrain) connects FADECO to a school building. This school uses the wireless link to connect more than 15 computers in its local area network to the Internet. The second wireless connection is used by an organization that utilizes it to connect 10 computers to the Internet. This organization’s building is 10 kilometers away from FADECO. These two connections are implemented using a point-to-point topology between the main building and each of the client sites. Wireless Internet connectivity was also extended to the local area via a hot-spot designed to provide Internet access to people in the local community who have computers with a wireless interface. In another attempt to address the needs of its local community, FADECO has installed a community radio station to extend its information services to communities that suffer from high levels of illiteracy. The new radio station equipment is built using low cost materials sourced locally and assembled by local staff members. Innovation is clearly visible in several components of the system. For example, plastic pipes are used as antenna masts to prevent lightning strikes to the equipment. The equipment is also designed to be portable and lightweight so that it can be easily used in the field to broadcast live events. The radio station went live on July 22, 2007, and provides local content for education and sustainable development to contribute to the local government’s efforts to fight and eradicate poverty. This story is an exemplary illustration of the potential offered by emerging technologies such as wireless networking and open source software to bridge the digital divide. Tackling the challenges to building a global information society no longer re-

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

quires significant investments. Innovative solutions can be designed and deployed using low cost materials and information that is freely available on the world-wide web. FADECO has been highly successful in exploiting these trends in technology development. Its low cost, locally designed and built network now serves the information needs of a population of 428,000 in ways that could not otherwise be possible. The service provided by the ICT deployed at FADECO serves farmers, business people, livestock keepers, housewives, and children in schools. The center is also playing a crucial role in preserving local culture through participatory content creation and publication. The ability to successfully design and implement low-cost ICT solutions depends on the availability of open information sources that provide the required know-how without the excessive restrictions of traditional copyright protection. Many African institutions do not have sufficient resources to train and build the capacity of their staff to manage their technology needs. The open content movement (e.g., as seen in action on Wikipedia) addresses this issue by providing more people with access to the information sources that are vitally important for building a global, equitable, and inclusive information society.

metrics. Through both quantitative and qualitative observations, we show that the digital divide has visible costs to society in both regions. The main contribution of this paper is an exploration of how open source software and wireless networks enable digital inclusion in the U.S. and Africa. Because the digital divide is wider in most regions in Africa than it is in the U.S., bridging the digital divide in Africa often requires more creative ICT solutions. We have observed that in both regions of the world, affordable technology and sound policy are both critical to empowering individuals by providing digital inclusion. We surveyed wireless networking and open source software, and analyzed their important role in bridging the digital divide. We presented more detailed investigations into a few important technologies, specifically IEEE 802.11 wireless networks, wireless mesh networks, and GNU/Linux-based open source software. We have seen the immediate and widespread impact that these technologies have had in communities as different as Castle Square in Boston, Massachusetts and Karagwe, Tanzania. We have also seen how these technologies are being used to educate children today, so that they will acquire the knowledge and develop the ingenuity needed to design global ICT and ICT4D in the future.

Conclusions

Acknowledgements

This paper explores the digital divide in both the United States and Africa. We analyze how the digital divide can be measured within these very different regions of the world. Because of the contrast in the state of ICT in the United States and Africa, it is appropriate to measure the digital divide using different

The authors would like to thank Gabriel Fishman, Stephen Ronan, Ian Howard, and Joseph Sekiku for their generous help in making this work possible. This research was funded in part by the International Development Research Centre and the Bentley Laboratory for Intelligent System Software.

References [Anna02] “Kofi Annan’s IT challenge to Silicon Valley”, CNET NEWS.com, November 2002, available at http://www.news.com/2010-1069-964507.html. [BABM05] John Bicket, Daniel Aguayo, Sanjit Biswas and Robert Morris, “Architecture and Evaluation of an Unplanned 802.11b Mesh Network,” ACM MobiCom Conference, August 2005. [BDS03] Carlos Primo Braga, John A. Daly and Bimal Sareen, “The Future of Information and Communication Technologies for Development”, Washington: Development Gateway, 2003. [Benv06] Christian Benvenuti, Understanding LINUX Network Internals, O’Reilly Media, (Sebastopol, CA, USA), 2006. [BMO06] BMO Book Sprint Team, How to Accelerate Your Internet, Rob Flickenger, (Seattle, WA, USA), October 2006. [CB07] Ken Coar and Rich Bowen, Apache Cookbook: Solutions and Examples for Apache Administrators, O’Reilly Media, 2nd edition, December 2007. [Econ05] “The real digital divide; Technology and development,” The Economist (U.S.), 374.8417, March 12, 2005. [FCC06] Federal Communications Commission, “Expanding Telecommunications Access in Indian Country,” FCC Consumer & Governmental Affairs Bureau Publication, July 2006. [FI07] Glen Farrell and Shafika Isaacs, Survey of ICT and Education in Africa: A Summary Report, Based on 53 Country Surveys, infoDev / World Bank, (Washington, DC, USA), 2007, available at http://www.infodev.org/en/Publication.353.html. [Flic07] Rob Flickenger, Corinna Aichele, Carlo Fonda, Jim Forster, Ian Howard, Tomas Kraq and Marco Zennaro, Wireless Networking in the Developing World, Lulu (Morrisville, NC, USA), 2nd edition, December 2007. [Gasp06] Tina Gasperson, “Brandon Elementary transformed by K12LTSP,” Linux.com Migration Case Study, July 2006, available at http://www.linux.com/articles/55632.

155

156

THE INTERNATIONAL JOURNAL OF TECHNOLOGY, KNOWLEDGE AND SOCIETY, VOLUME 4

[GZ03] Richard Gerster and Sonja Zimmermann, “Information and Communication Technologies (ICTs) for Poverty Reduction?”, Discussion Paper, Swiss Agency for Development and Cooperation (SDC), March 2003. [Heek08] Richard Heeks, “ICT4D 2.0: The Next Phase of Applying ICT for International Development,” IEEE Computer, vol. 41, no. 6, pp. 26-33, June 2008. [Hend08] Dewayne Hendricks, Personal Communication, February 2008. [Hoe06] Nah Soo Hoe, “Breaking Barriers: The Potential of Free and Open Source Software for Sustainable Human Development - A Compilation of Case Studies from Across the World”, UNDP Asia-Pacific Development Information Programme, 2006. [HSA05] Jay Horwitz, Ina Sebastian and Julie Ask, “Municipal Wireless: Partner to Spread Risks and Costs While Maximizing Benefit Opportunities,” Jupiter Research Report, June 2005. [Huds06] Heather E. Hudson, From Rural Village to Global Village: Telecommunications for Development in the Information Age, Lawrence Erlbaum Associates, (Mahwah, NJ, USA), 2006. [IDRC03] International Development Research Centre, “The Internet: Out of Africa”, November 2003, available at http://www.idrc.ca/en/ev-6568-201-1-DO_TOPIC.html. [ITU05] “Measuring Digital Opportunity”, ITU Report, June 2005, http://www.itu.int/itu-wsis/2005/DOI%20V2.pdf. [ITU07] “World Information Society Report 2007: Beyond WSIS”, ITU Report, http://www.itu.int/osg/spu/publications/worldinformationsociety/2007/report.html. [JE01] Mike Jensen and Anriette Esterhuysen, “The Community Telecentre Cookbook for Africa: Recipes for Self-Sustainability,” UNESCO Report CI-2001/WS/2 (Paris, France), 2001. [John08] Robert Johnston, “Open Letter to Nicholas Negroponte: Its Education Not Laptops,” January 16, 2008, available at http://www.olpcnews.com/people/negroponte/. [LSG06] William Lehr, Marvin Sirbu and Sharon Gillett, “Wireless is Changing the Policy Calculus for Municipal Broadband,” Government Information Quarterly, vol. 23, no. 3-4, pp. 435-453, 2006. [Morr07] Chris Morris, “First Mile, First Inch,” Meraka Institute Research Report, (Pretoria, South Africa), 2007. [Moy94] John Moy, “OSPF Version 2,” IETF Internet Request for Comments 1583, March 1994. [Nico08] Ryan Nichols, Statement on Community Need addressed by Wireless Philadelphia Project, January 22, 2008, available at http://www.phillyfellows.org/placements0809/wp.html. [Norr01] Pippa Norris, Digital Divide: Civic Engagement, Information Poverty, and the Internet Worldwide, Cambridge University Press, (New York, NY, USA), 2001. [NTIA00] “Falling Through the Net: Toward Digital Inclusion”, U.S. NTIA Report, October 2000, available at http://www.ntia.doc.gov/ntiahome/fttn00/contents00.html. [NTIA04] “A Nation Online: Entering the Broadband Age”, U.S. NTIA Report, September 2004, available at http://www.ntia.doc.gov/reports/anol/NationOnlineBroadband04.htm. [Pers07] Personal Telco Project, “Wireless Communities,” http://wiki.personaltelco.net/index.cgi/WirelessCommunities, accessed December 21, 2007. [Raws06] Alice Rawsthorn, “One Laptop Per Child: Computer designed for those who can least afford them,” in the International Herald Tribune, November 19, 2006. [Siev05] Ellen Siever, Aaron Weber, Stephen Figgins, Robert Love and Arnold Robbins, Linux in a Nutshell, O’Reilly Media, 5th edition, July 2005. [TMS06] Andrea Tapia, Carleen Maitland and Matt Stone, “Making IT work for municipalities: Building municipal wireless networks,” Government Information Quarterly, vol. 23, no. 3-4, pp. 359-380, 2006. [UCSC04] “Global Digital Divide: The Issue”, UC Atlas of Global Inequality, July 2004, available at http://ucatlas.ucsc.edu/communication/digitaldivide.php. [UN04] United Nations, “Supporting Africa’s Efforts To Achieve Sustainable Development: Dialogues Of The Economic And Social Council,” United Nations Publications (New York, NY, USA), June 2004. [Webe04] Steven Weber, “Open Source Software in Developing Economies”, Social Science Research Council, New York, 2004, available at http://programs.ssrc.org/itic/publications/ITST_materials/webernote2.pdf. [WPB07] Wireless Philadelphia Blog, “City of Philadelphia Approves Municipal Wi-Fi Proof of Concept,” May 24, 2007. [WPC08] Wireless Philadelphia Coverage Map, “Wireless Areas – Coverage Areas,” http://wirelessphiladelphia.org/coverage_area.cfm, accessed January 21, 2008. [WSIS03] World Summit on the Information Society, “Building the Information Society: A Global Challenge in the New Millennium”, (Geneva, Switzerland), December 12, 2003. [Zhan07] Yuwei Zhang “Connect Africa Summit Commits to Bridging the Digital Divide,” UN Chronicle Online Edition, November 2007.

About the Authors Dr. David J. Yates David is an Assistant Professor of Computer Information Systems at Bentley University. David’s research areas include computer networking, data communications, sensor networks, embedded systems, operating systems, and computer architecture. Before joining Bentley, David held research and academic positions at the University

DAVID J. YATES, THOMAS MCGONAGLE, ANAS TAWILEH

of Massachusetts and Boston University. His work has been published and presented at international symposiums and conferences, and appeared in refereed journals. In the corporate arena, he was a co-founder and vice president of software development at InfoLibria – a startup that grew to become a leading provider of hardware and software for building content distribution and delivery networks before it was acquired. With various colleagues, he holds several U.S. patents for processes and systems related to computer networking, content management, and mobile computing. He holds a PhD and MSc from the University of Massachusetts. He also holds a BSc from Tufts University. Thomas McGonagle Thomas is a passionate open source advocate and Boston area wireless activist. He has built and operated several community wireless networks. He has also developed and deployed large-scale, open source based systems for the Federal Aviation Administration (FAA) and a high traffic, market leading social networking site. He also works as an Adjunct Instructor teaching advanced Linux and networking courses at the Benjamin Franklin Institute of Technology. He holds an MS and BS from Bentley University. He expects to complete a second MS at Brandeis University and plans to continue his graduate studies at Emerson College. Anas Tawileh Anas is a researcher and consultant for Information and Communication Technology for Development (ICT4D). Anas worked on many projects to bring technology to developing countries and has designed, developed and delivered several training and capacity building programs and workshops. He is currently working as consultant for the International Development Research Centre’s ICT4D project in the Middle East. His experience also includes more than seven years of working with international organizations including the European Commission, the Open Society Institute and the Association for Progressive Communications.

157

EDITORS Bill Cope, University of Illinois, Urbana-Champaign, USA. Mary Kalantzis, University of Illinois, Urbana-Champaign, USA.

EDITORIAL ADVISORY BOARD Darin Barney, McGill University, Montreal, Quebec, Canada. Marcus Breen, Northeastern University, Boston, USA. G.K. Chadha, Jawahrlal Nehru University, India. Simon Cooper, Monash University, Australia. Bill Dutton, University of Oxford, United Kingdom. Amareswar Galla, The University of Queensland, Australia. David Hakken, University of Indiana, Bloomington, Indiana, USA. Michele Knobel, Montclair State University, New Jersey, USA. Jeannette Shaffer, Edtech Leaders, VA, USA. Ravi S. Sharma, Nanyang Technological University, Singapore. Robin Stanton, Australian National University, Canberra, Australia. Telle Whitney, Anita Borg Institute for Women and Technology.

Please visit the Journal website at http://www.Technology-Journal.com for further information about the Journal or to subscribe.

THE UNIVERSITY PRESS JOURNALS International Journal of the Arts in Society Creates a space for dialogue on innovative theories and practices in the arts, and their inter-relationships with society. ISSN: 1833-1866 http://www.Arts-Journal.com International Journal of the Book Explores the past, present and future of books, publishing, libraries, information, literacy and learning in the information society. ISSN: 1447-9567 http://www.Book-Journal.com Design Principles and Practices: An International Journal Examines the meaning and purpose of ‘design’ while also speaking in grounded ways about the task of design and the use of designed artefacts and processes. ISSN: 1833-1874 http://www.Design-Journal.com International Journal of Diversity in Organisations, Communities and Nations Provides a forum for discussion and builds a body of knowledge on the forms and dynamics of difference and diversity. ISSN: 1447-9583 http://www.Diversity-Journal.com International Journal of Environmental, Cultural, Economic and Social Sustainability Draws from the various fields and perspectives through which we can address fundamental questions of sustainability. ISSN: 1832-2077 http://www.Sustainability-Journal.com Global Studies Journal Maps and interprets new trends and patterns in globalization. ISSN 1835-4432 http://www.GlobalStudiesJournal.com International Journal of the Humanities Discusses the role of the humanities in contemplating the future and the human, in an era otherwise dominated by scientific, technical and economic rationalisms. ISSN: 1447-9559 http://www.Humanities-Journal.com International Journal of the Inclusive Museum Addresses the key question: How can the institution of the museum become more inclusive? ISSN 1835-2014 http://www.Museum-Journal.com International Journal of Interdisciplinary Social Sciences Discusses disciplinary and interdisciplinary approaches to knowledge creation within and across the various social sciences and between the social, natural and applied sciences. ISSN: 1833-1882 http://www.Socialsciences-Journal.com International Journal of Knowledge, Culture and Change Management Creates a space for discussion of the nature and future of organisations, in all their forms and manifestations. ISSN: 1447-9575 http://www.Management-Journal.com International Journal of Learning Sets out to foster inquiry, invite dialogue and build a body of knowledge on the nature and future of learning. ISSN: 1447-9540 http://www.Learning-Journal.com International Journal of Technology, Knowledge and Society Focuses on a range of critically important themes in the various fields that address the complex and subtle relationships between technology, knowledge and society. ISSN: 1832-3669 http://www.Technology-Journal.com Journal of the World Universities Forum Explores the meaning and purpose of the academy in times of striking social transformation. ISSN 1835-2030 http://www.Universities-Journal.com

FOR SUBSCRIPTION INFORMATION, PLEASE CONTACT [email protected]