As Indonesia develops connectivity domestically, it has to begin thinking about international links. The emphasis to date has been on unifying the nation via the internet, and considerable progress has been made. The building of the internet backbone has been important in getting more of the remote corners of the archipelago online, while wireless broadband has helped expand high-speed access.

Yet domestic connections are of limited value. If the country is to have a robust technology sector, its internet will need to be better integrated with the global network. Otherwise it will be hard for Indonesia to extract full value from its network. Its system will be relatively closed and less useful than one seamlessly connected to the rest of the internet.

International Connections

The country is currently linked to only one intercontinental cable – the South-East Asia-Middle East-Western Europe 3. This optical fibre submarine cable runs 39,000 km from Europe, through the Middle East, across to South-east Asia and Korea via China and Japan. It lands at Medan and Jakarta. Indonesia has no direct connection to the Asia-America Gateway, a 20,000-km cable running from the US West Coast across the Pacific Ocean to South-east Asia, despite the fact that Telkom is an investor in the cable. The country links into that line indirectly, via the 73-km Batam-Singapore cable, which has six pairs of fibre, though four pairs have intentionally been left dark.

Indonesia does have good regional connections. Between 2003 and 2010, 10 cables were installed between the country and South-east Asia. But it lacks direct connectivity to the global network. This is worrying as, according to Terebit Consulting, most traffic is bound for the US and Europe; even much of the intra-ASEAN traffic is only making a hop toward a destination outside the region.

Choke Point

A 2013 report by the UN Economic and Social Commission for Asia and the Pacific (UNESCAP) highlighted the concern that most of Indonesia’s international internet capacity runs through Singapore, and this choke point invites less-than-competitive pricing. The numbers indicate that Indonesia is faring poorly in the wholesale markets.

By the end of 2012 volume purchases of bandwidth by Indonesia were estimated at over $60 per Mbps per month. In Malaysia, it was around $25 per month.

Indonesia is also suffering from a shortage of international connectivity. According to the UNESCAP study, the country had 362,000 Mbps of international bandwidth in 2013. Capacity is up, and set to grow further – it has increased 340 fold in a decade, and is expected to double by 2015 and again by 2017. However, on a per capita basis, Indonesia still ranks low, with very little international bandwidth per person. UNESCAP estimates that there is 1 Kbps of bandwidth for each Indonesian. That compares poorly with regional competitors. Thailand is at 6.6 Kbps;

Vietnam, 5.1; and the Philippines, 5.4. Singapore has 258 Kbps of international bandwidth per person.

New Lines

Efforts are being made to improve connectivity. In 2012 Telkom made a bid for Pacnet, which was formed from the 2008 merger of Pacific Internet and Asia Netcom, and is region’s largest owner of undersea cable, with 46,000 km worth.

The deal was cancelled soon after, however.

Two additional cables are in the works, but they do not seem to promise a solution to the capacity bottleneck. The Brunei Darussalam-Indonesia-Malaysia-Philippines East ASEAN Growth Area has been discussing a $150m cable to link the member states. The project was talked up a few years ago, but has received very little attention since then. It would also appear to not offer much in the way of international bandwidth to Indonesia.

A Singapore, Indonesia and Australia cable has been on the drawing board for a number of years; indeed, two groups have been competing to install the line. The projects are very compelling, as they would help lower Australia’s internet costs and improve Indonesia’s connectivity. However, the projects are stalled and it appears that little progress will be made in the short term. Discussions about cables are about more than just Indonesia’s needs. The country’s technological fate is part of larger discussions in the international bandwidth community.

An intercontinental cable brings participation from a diverse range of parties. In the Pacific market, entrenched interests have the financial might and expertise but are best served by limiting capacity, whereas entrepreneurs see unmet demand and an opportunity to serve markets and circumvent existing bottlenecks. So while the demand for new capacity exists in both Indonesia and Australia, the right set of stakeholders could not agree on the project.

Domestic Hosting

Larger and more recent issues also have the potential to affect the future of a submarine cable business, according to Sunil Tagare, developer of the 28,000-km Fibre optic Link Around the Globe (FLAG) project. He said that the trend toward in-country hosting requirements – in which Google, for example, would be required to have its servers for Indonesian customers located in Indonesia – threatens the sector. He writes that this trend is so significant that it could lower demand for capacity and bring the cable industry to a standstill.

The push goes back a number of years. Some relevant regulations were drafted in Indonesia in 2009, and heated up in 2011 when requiring BlackBerry to host in-country was discussed. The issue is now back, with a bill under consideration that would require all data centre companies to keep their data and recovery facilities onshore. The mood in the developing world is to force major internet companies to host domestically, as evidenced by Brazil’s push to have Google do just that. The countries are driven by security and commercial issues. Politicians in developing countries argue that keeping data within their borders would limit the exposure of their citizens to hacking. While experts say that the location of servers makes little difference, as the internet is interconnected and one part of it can be reached from any other, many electorates feel that having data at home would help protect information.

The other motivation is commercial. Countries such as Indonesia are concerned that most technological developments are currently taking place abroad, and that the resulting products are being sold within their borders via the internet. The idea is that by forcing firms to open operations domestically, more value will remain onshore. The push for in-country hosting is related to other semi-protectionist measures, such as those involving downstream processing of raw materials.

Servers

Just as the push for smelters has received international criticism, so has the push for domestic servers. The key to the cloud is critical mass. By building very large and efficient server farms in the best locations, the cost of data storage and processing can be cut dramatically. The networking effect can also be enhanced because of the low latency within the cloud. By forcing servers onshore, Indonesia would miss out on many of the cloud’s advantages. It would not get the same economies of scale, and would put its corporations and entrepreneurs further from resources they need. Hosting would also be taking place in a less-than-ideal location – a country which suffers frequent power outages and has high logistical costs. Indeed, forcing hosting onshore could actually slow technological development in Indonesia, rather than enhance it.