Page 7 of 13INFRASTRUCTURE
So far, we have discussed the analytic lenses that are important in ICTD. At the level of infrastructure, connectivity has continued to dominate ICT discussion in the Asia Pacific region. Between 1999 and 2006, the number of Internet users in Asia and the Pacific increased five-fold, from two to 12 per 100 inhabitants. But this is still below the world average of 17 and far below the figures of 69 in North America and 43 in Europe (UN ESCAP 2007). Nevertheless, economies of scale and the increased number of Internet users pushed the demand in this sector from "no or limited connectivity" to broadband and a level of "bandwidth redundancy". However, in 2006, there were still only three broadband subscribers per 100 people in Asia and the Pacific, compared with 20 in North America and 16 in Europe (UN ESCAP 2007).
It is not only the throughput (usually measured in bits per second) of the Internet connection that is important. The latency of commonly used satellite-based bandwidth makes it unsuitable for many services such as Voice over Internet Protocol (VoIP). One solution for these countries has been to join consortia to install and use under-the-sea fibre optic cables. The first South East Asia-Middle East-Western Europe cable system (known as SEA-ME-WE) was introduced in 1985. Fibre optic cable was laid to establish SEA-ME-WE 2 connecting the three zones in 1994. SEA-ME-WE 3 introduced "Wave Division Multiplexing" (WDM) technology in 1999, connecting 39 landing points in 33 countries from Germany to Australia. In November 2005, SEA-ME-WE 4 would carry 1.2 terabytes per second (Tbps) of bandwidth. This cable system has connected 16 landing points in 14 countries in the three continents (Undersea Cable 2006).2 Most of the companies that formed this consortium are government- or state-owned enterprises.
However, the submarine cable system does not seem to be adequate for the demand and it is susceptible to disruption of services. Thus, governments have been looking to the private sector to develop alternative or additional submarine cable systems. For example, Bharti Airtel, a private company in India, is joining five other companies in Japan, Malaysia, Singapore, and the US to build a high-bandwidth, undersea fibre optic cable linking Asia and the US, to go live in 2010. The Bangladesh government has also decided to allow a second submarine cable financed by the private sector to maintain uninterrupted overseas voice and data communications and to back up its existing undersea cable.
Google is involved as a service provider and non-telecom investor in all three additional under-the-sea fibre optic networks that connect the US with the Asia Pacific region, marking a significant shift in funding models for data. The Trans-Pacific Express Cable System is going to connect the US with China, the Republic of Korea, and Taiwan. The Asia-America Gateway Cable System, which is being planned to come on service in the first quarter of 2009, will connect the US and several South Asian countries. The third cable is being planned by Reliance FLAG with a speed of 2 × 1.28 Tbps. When operational, these three cables will change not only the present landscape of bandwidth capacity, but also the price regime of connectivity in the Asia Pacific region, leading to further multimedia- and connectivity-based services in the region. The unknown economic question is whether the region is a net consumer or producer of such services.
In parallel to these external connectivity opportunities, there is a push by national governments to add capacity, and share or build up backbone infrastructure. In some landlocked countries such as Nepal and Cambodia, government-owned telecom entities are laying out fibre optic backbone connecting to the nearest country that has access to under-the-sea cable. Afghanistan is building its national fibre optic backbone following the national ring-road infrastructure and is planning to connect this with under-the-sea cable through Iran, Turkmenistan, Uzbekistan, Tajikistan, and Pakistan.
Some governments are making it a licencing obligation for service providers to rollout to rural areas. For example, in Pakistan, the private sector has led the expansion of three new nationwide optical fibre systems. One private telecom entity in Sri Lanka already owns a nationwide fibre network that is supplemented by Worldwide Interoperability for Microwave Access (WiMAX) technology for broadband to the door. The Indian government has directed the private incumbents to extend the network to rural areas (Samarajiva and Zainudeen 2008).
As the chapters on individual economies in this volume show, investment vehicles are being developed so that costs can be shared by various entities using the infrastructure. Australia has developed a hybrid plan where private companies would provide the infrastructure to the populous areas and government funding would make feasible the rollout of services to most regional, remote, and rural communities. The Bangladesh government has recently signed a deal with the Power Grid Company of Bangladesh Limited to provide backup fibre optic network to the existing one.
Different governments in the region are also making substantial political commitments to broadband expansion, which is making investment in backbone infrastructure more viable in areas that were not commercially viable earlier (Samarajiva and Zainudeen 2008). For example, the Indian government is expecting 20 million broadband connections by 2010 and plans to ensure broadband connectivity in every school, health centre, and Gram Panchayat (local government units). Digital Subscriber Line (DSL) users in Pakistan now number 100,000, and the target is to reach 1.6 million in 2009. The National Telecommunications Commission of Thailand has already granted 12 licences for operators to conduct commercial trials of broadband wireless access and allocated frequency for this. The Malaysian government's Information, Communication, and Multimedia Services 886 Strategy (MyICMS 886) talks about eight new services to build up eight essential infrastructures that includes high-speed broadband. In the Republic of Korea, the IT839 strategy consists of the introduction of eight new services that it is hoped will prompt investment in the building of three essential networks. The synergies here are aimed at stimulating nine new sectors, including intelligent services and home networks. Notable is the link between enabling infrastructure and technology and the clear identification of the economic sectors to be stimulated, even if such outcomes are not always determined in advance (Shin 2007). Some of the roles governments can take in stimulating infrastructure are taken up further in the overview chapter on ICT policy.
At the logical layer of Internet infrastructure - between the hardware and end-user applications - changes are also occurring as many Asian countries are introducing next generation Internet protocol, Internet Protocol version 6 (IPv6), which allows greater flexibility in assigning addresses. IPv6 can support a bigger set of 3.4 × 1038 (340 undecillion) unique addresses while Internet Protocol version 4 (IPv4), which is still widely used, was designed to provide about four billion unique Internet Protocol (IP) addresses only. China, Japan, the Republic of Korea, and Taiwan have been at the forefront of the first wave of IPv6 deployment, while the second wave has been led by Australia, India, Indonesia, the Philippines, Thailand, and others. One of the motivations for Asian countries to move to IPv6 was that Asian countries control only 9 percent of the allocated IPv4 addresses while they have half of the world's population. However, even though the protocol has been ratified for some time, IPv4 remains widely used and the challenges in stimulating widespread uptake point to the unusual governance questions that arise in an Internet environment with decentralized authority, as there are no incentives for managers of core infrastructure to deploy the new protocol.