africa blog

July 2, 2007

The ABC of ICT4D

How best to understand information and communication technologies for development

Heloise Emdon, Senior Programme Specialist ICT4D Africa, IDRC

GPS used in a Morogoro village, TanzaniaThrough experience we know people who earn as little as R1000 a month and who also have a cell-phone. Living on this amount a month means people are living above the poverty line of R14 (US$2) a day.

We need to ask: What do we know about how much people spend on these telephone calls, and why do they make such calls on slight incomes?

While most telecommunications and IT market research is about the top spenders and the top earners, little is known about the millions of very small transactions that make up the mass of most of the worlds population.

However, it is this mass group in the market that makes it feasible and commercially worthwhile for service providers and producers. Their statistical distribution is known as the ‘long-tail’ which is “the colloquial name for the low-frequency or low-amplitude population which gradually “tails off” from the big spenders, or the high-frequency or high-amplitude population. In many cases the infrequent usage or low-amplitude events — the long tail, represented here by the bright yellow portion of the graph — can cumulatively outnumber or outweigh the initial portion of the graph so that they comprise the majority.” http://en.wikipedia.org/wiki/Long_tail

A is for Access

//www.signonsandiego.com/uniontrib/20050826/news_1n26phones.htmlUntil cell phones and the market structures in most African countries changed the ability of the poorest of the poor to have access to a phone, resource-poor people only had access to public phones. Now households and individuals use cell phones to enhance their livelihoods, or even turn cell phones into livelihoods by selling time. Fewer people in these economic conditions use the internet, though it is much cheaper for instance to make voice over the internet call. The internet just does not reach that many poor communities, either in rural or urban settings. However, there are several public access internet access points, either businesses or development projects. Using the internet also requires a set of skills, least of which is literacy. But illiterate people can make use of highly visual interfaces, although access and availability are the limiting factors here.

B is for behaviour

 

We do know that resource-poor people use their phones to make various kinds of calls. Some are essential and life-saving, others are to ensure that the family is well, or to check market prices if they are delivering produce and want to get the best prices. Many youths and adults are looking for jobs. (For more on these kinds of behaviours see note at the bottom.) The highest demand for services amongst low income people remains voice calls. However, airtime is sometimes traded for cash, remittances are sent through sending airtime and can be on-sold for cash, or a remittance is sent via someone traveling home and the migrant worker sends a message home to look out for the cash-carrying courier.

A useful approach that gives us understanding of how ICTs are used to improve people’s livelihoods is to consider the way in which people use phones, radio, the internet, television to strengthen their social networks, (social capital). Also how they can improve their skills, training and even education using any of the above (human capital); and gain access to market information, information about natural resources or financial information (market, natural resource and financial capital). These are the five forms of capital that resource-poor people rely on to make a living, and where they experience shortages in times of shock or disaster.

(see http://www.livelihoods.org/info/guidance_sheets_pdfs/section2.pdf).

C is for Cost

Nakaseke Telecentre, UgandaEven though the most popular prepaid units are the cheapest, and resource poor people will continue buying these cheapest units which allow them to make just a few minutes of calls at the highest rate per call, and some “call me” messages per day, the real cost of calls is normally borne by those who do call back. They might be the well-off urban dwelling family member, the employer, but not yet the government service call centre, the hospital service, or business call centre. Here are some ideas for e-government services. Poor people cannot afford the expense of holding on while waiting for a response, but they could just leave their numbers and expect a call back.

Telecommunication services are still extremely high in Africa. The most basic reason is that the cellular companies get assymetrical interconnection agreements, it is usually more expensive to call a cell phone then a fixed line, and more importantly the fixed line operators in most African countries still enjoy the monopoly of voice calls. The cost of the Internet is also largely determined by the cost that the telco charges the internet service provider, the latter adding only a small margin. The most expensive access is satellite communications. Almost every square kilometer of Africa is covered by satellite communications, but governments either control the access or charge extremely high fees for services to bypass their telecommunications services, making this the most expensive model.

Masyoi Vilage CantennaeWe have supported projects that seek other ways to aggregate the cost of those internet connections, especially the high cost of satellite connectivity and through sharing the signal with others using a wireless access point. These WiFi signals use the least cost interventions, such as DIY “cantennae” enable several users and especially rural NGOs, households, schools and clinic to gain access to the signal and to share the costs. Some satellite and some leased line services can cost community projects, access points or clinics between R4 344 and R7 240 (US$600 and $1000) a month. See www.fmfi.co.za

D is for digital content and open access

When referring to the cost of owning a computer, connectivity, the software and length of use that the hardware and software make possible, the term Total Cost of Ownership is useful. Consider that all content, not only the dialup cost, is costly. One of the ways to further lower the cost of access is to choose intellectual property rights that allow the author to share, ask for attribution, allow for remixing, sharing any improvements back with the author and others who would be using it, allowing one to generate an income or not from it. This is choosing a particular kind of intellectual property approach that gives the author and user more rights than the normal default restrictive copyright procedures. We have supported the Creative Commons getting established in South Africa and promoting the use amongst other early adopters.

See http://www.commons-sense.org/index.htm# for several of the tools available to create open access content. We also support the use of free and open source software which makes a whole lot of sense for a global community of developers constantly improving and a whole lot of users constantly proving the use.

See www.OpenMRS.org or www.AVOIR.uwc.ac.za.

E is for efficiency

Finally, in this ABC of ICT4D, the most compelling reason for governments and NGOs to be using information and communication technologies is that they can deliver their services more efficiently. Consider the benefit to those patients on lifelong Anti-retroviral therapy whose files are kept electronically. This means every visit to the clinic or complication that needs to be reviewed by a doctor, the patient’s record is up to date, the vitals have been monitored and entered, any drug interactions, reactions or other symptoms are monitored. Those health services that are able to maintain this kind of service are able to ensure better quality of care, improved management information systems, better reporting about the effects of the treatment on patients and the management of the treatment project-wide, province-wide or nation-wide.

PDA for patient monitoring and electronic medical recordThe household surveys that are conducted to establish the burden of disease in any region or country, facilities-based information in a health system, or extension of microcredit loans are more efficiently managed on handeld computers. They may be synchronized or dialed up from remote rural settings to the data-base management servers held in capitals. Here the data is managed and analysed, processed and fed-back into the field or to national or international level decision-makers.

PDA forms software for electronic medical records, survey toolsActually the algorithms and protocols that can be coded into most forms-based software can provide the fieldworker with enough feedback to enable that fieldworker to manage patients on lifelong treatment, manage loan portfolios, give advice in remote settings and when smart phones get cheaper, this could all happen in real time.

ICTs are for Development?

Are you convinced? Cell phones are not mere luxuries or fashion items, they are tools in the employ of people development. If we consider Maslow’s hierarchy of needs the idea of communication fits snugly into the basic human need for security. There is no longer a debate about whether communications infrastructure and gaining access to it fulfills a basic human need. The challenges we face in development are improving access, decreasing cost of access and total cost of ownership, access to open content and open source software as well as making use of all of this to improve health and education services in resource-poor countries, and several attendant services that can improve livelihoods, such as agriculture, primary production in order to improve our economies.

Governments fear liberalizing their telecommunications markets because they fear loosing the goose that lays the golden egg. What they don’t realize is that cheaper and improved access to communication spawns many, many golden geese!

The slideshow is available at http://www.slideshare.net/kdiga/idrcpeople-media-i-t-22-jun07-ict4d/

Some resources. See: “Towards An African e-Index: SME e-Access and Usage in 14 African Countries” 2006 http://researchictafrica.net/modules.php?op=modload&name=News&file=article&sid=518 as well as at the “Towards an African e-Index: Household and Individual ICT Access across 10 African countries” 2005. http://researchictafrica.net/modules.php?op=modload&name=News&file=article&sid=504.

Both of these research reports were funded by the International Development Research Centre http://www.idrc.ca/Acacia or www.idrc.ca/acacia

August 15, 2003

Stuck in a Jakarta Traffic Jam? Find your way onto the Information Highway!

Heloise Emdon, Jakarta, 15 August 2003

Pragmatic innovation:

The day the lights went out from Ottawa to New York. Frank Tulus and myself found ourselves stuck in Jakarta traffic jams. It is not these traffic jams I will here describe, apart from mentioning the smoggy, pushy, hooting, humid, clamour of cars and mopeds, there is nothing remarkable about the Jakarta traffic jams: except the huge inadequacy of the infrastructure to carry the people.

It’s the other highway I want to tell you about: The more than 50 ISPs in Indonesia would not settle for the high cost of 2Mb lines that the Indonesian Telkom is selling them at Rupea 40million per year ($5m). In Indonesia we learn how the ISPs and the community have been pragmatic in “getting around” this high cost.

jakarta_woman.jpgNurlina, Onno’s wife who met us because he was on a “roadshow” in Bali, drops us near the high rise buildings in the Mampang district, a newer section of what seems to be downtown of sprawling Jakarta. We are closer to the JW Marriot that got bombed just two weeks earlier. It is hard to locate a downtown because even where these multinational towers scrape the sky, there are low-rise commercial and domestic buildings, a labyrinth of small streets that bottleneck into large intersections, and then dual carriages that cut between this planning stew with footbridges to cross over. Clearly some thoughtful planning did go into solving how to get people across the busy highways.

Frank and I walk across a footbridge spanning the dual carriage way, view the remains of an informal settlement that was moved from the area, drop down onto a dirt road track that leads to a small, unassuming road where the Indonesian Internet Exchange building stands. There are no serious security checks, or so it seems, we go up in a lift to the Apjii offices, the secretariat that serve all the Indonesian ISPs. Their offices are all located in this building. It is around lunchtime and the lift is loaded with young men, none of them a day over 30 years. I later discover that when the Indonesian government started licensing ISPs one of the conditions for registration was that the licensees be younger than 35 years. No one can explain why the government thought of this investment in the youth, but it has paid off in getting innovators to invest in Indonesia’s future. It seems, however, that the government stifles a lot of other innovation.

Demand for internet bandwith has increased dramatically, and could be ascribed to the both corporate and multimedia demand. Everyone is mum about the effect of VOIP on the system because of the telco’s monopoly on voice traffic, and despite magnitudes of compression available for VOIP, one must safely assume that in a country where there is less than 10% penetration of fixed lines, a total of 6 million fixed lines and 7 million GSM connections, the dramatic demand for data, around 1Gbps could be ascribed to at least some voice traffic.

jakarta_window.jpgWhen large ISPs experienced similar traffic problems, and Telkom would only commit a further 2Mbps if the ISP were to pay for the construction and terminal equipment at huge expense of $12 000 for installation and the further $40 000 per month, they agreed to establish a “carrier neutral data centre” to interconnect with each other at no cost. They located the ISPs servers, routers and switches on the same floor of the Elektrindo Building.

By this peering they were able to increase their shared bandwith. By physically locating each of the servers as well as all the access infrastructure such as the international gateway on these premises, the servers are peered running Ethernet cable between the peering ISPs in the same building so that they could run 100Mb Ethernet cables (which is restricted to 100m lengths) amongst the racks in the same building and so doing increasing the bandwidth they have available in their local peering point:.

http://www.apjii.org.id shows the following logical graphic of its peering point:

Since the peering point was established in 1999 the bandwidth demand has grown from:

—————————

Based on the annual report of the Indonesian ISP Association (APJII) that can be downloaded from here, the estimated Internet users and subscribers up to the end of 2002 is as follows:

Subscribers Users
1998 134.000 512.000
1999 256.000 1.000.000
2000 400.000 1.900.000
2001 581.000 4.200.000
2002* 1.000.000 8.000.000

*Estimated up to the end of 2002
Table: Growth Indonesian Internet Subscribers and Users
Source: APJII (www.apjii.or.id)
From: Onno W Purbo,

————————————-

The Entrepreneur behind the “only physical” peering point Johar Alam, describes himself as a real-estate manager. The building we are in houses the physical peering point of all the ISPs, and once there the international gateway providers Indosat, and other satellite and cable service providers landed their ports into the same building, as did Telkom which provides the fibre into the rest of the Indonesian archipelago. However its service is not impressive: serving a mere 6 million of the 240 million Indonesian people. It is believed that 50 million Indonesians have never heard dial tone.

jakarta_smile_man.jpgFrom his chapter in the “Digital Review: Asia Pacific” on Indonesia Onno Purbo reports that Indonesia, despite its capacity constraints, is using a peak of 1Gbps.

“As reported by Johar Alam, the administrator of Indonesian Internet Exchange, in the year 2002, the total IIX in country peak bandwidth is in access of 250Mbps. Since the international traffic is normally about three (3) times of local bandwidth, the peak Indonesian international bandwidth is estimated about 800Mbps. The peak bandwidth is normally about 80% of the maximum bandwidth. Thus, it is safe to estimate a maximum bandwidth of 1Gbps from Indonesia to the Internet The ratio of in-coming and out-going Internet traffic volume is about 1:10 as Indonesian is still consume more information rather than produce information.”

Battling for independent regulator

jakarta_two_men.jpgMs Rahayu Juniarti Soehardo and Taru Wisnu show off the publications produced by Mastel, the Indonesian Infocom Society. Representative of all the service providers, the society is working hard at proposing policy on forming an independent regulator in Indonesia.

Blog at WordPress.com.