The Indian government wants the current Internet Governance Forum to be beefed up so that it will actually produce decisions. Those who do not represent governments would prefer not to, because the word-by-word negotiations that would result from such a change would not be productive and because such a procedure would necessarily privilege governments. They would prefer decision to be taken elsewhere, informed by IGF discussions. There was no enthusiasm from anyone for the idea of an International Internet Union as favored by Hamadoun Toure and Vladimir Putin.

Partly due to bad architecture (too big a room; too high a stage; big old barriers between audience and panel), it was not the most lively of sessions. I tried hard to allow for questions from the audience and the promised participants from afar, but got only one person asking questions. It could have been livelier, but I think we did justice to the key issues.

For those who like such things, video and transcripts are available here.

While the actual data on the number of Internet users in Sri Lanka shows a significant increase in the last few years according to ITU data, it is useful to understand the methodologies behind the numbers.  The ITU, which is the basis of Amis’ article, recommends the use of demand side data (i.e. data obtained from representative sample surveys of the population) to measure the number of internet users.  But user surveys are costly and difficult to organize.

When survey data is not available, the current method for  estimating the number of Internet users is to take the number of internet subscriptions (as reported by the country’s operators, or by the country’s telecom regulator) and multiply that by a number/ multiplier.  This is because each internet subscription is assumed to be used by more than one person: so the multiplier accounts for people who use Public Internet Access Points (PIAPs) or people who use the Internet at their work places, schools or other public locations.

As at 2010 there were 298,100 fixed Internet subscriptions and 2.503 million Internet users in Sri Lanka according to the ITU, with an Internet penetration rate of 12%. But according to the 2009 Household ICT Literacy Survey conducted by the Department of Census and Statistics, Sri Lanka had an “Internet using household population (aged 5 – 69)” of 13.1%.  It is estimated that there were 17.7 million people aged 5-69 in Sri Lanka according to UN Population Division, Department of Economic and Social Affairs.  In that case there were 2.3 million internet users in Sri Lanka in 2009.

The Telecommunication Regulatory Commission Sri Lanka (TRCSL) website gives 280,000 as the number of “Internet and email subscribers – fixed” as at Dec 2010, a rather mysterious and undefined indicator that does not seem to have international comparators.   Though ITU recommends use of demand side survey data, they do not seem to be taking the data of the 2009 ICT literacy survey into account, nor do they seem to be using the TRCSL data, so where ITU obtains its figures of 298,100 or 2.503 million is unclear.

LIRNEasia conducted a research to understand the relationship between the multiplier used to calculate the number of Internet users from the number of subscriptions and GNI per capita income ranking.  We found that a reverse correlation exists – that is, higher the GNI ranking, lower the multiplier. In other words in countries with a high GNI ranking, fewer  people share their Internet subscription and therefore the multiplier is expected to be closer to one, while in countries with a low GNI ranking more people use public Internet access points and share any other given connection, which would give a higher multiplier. We recommended to the ITU the use of this correlation to estimate the number of Internet users, instead of the current arbitrary method/multiplier. Using our method, with ITUs 298,100 Internet subscriptions as the base we get 2.17 million users.

Indi had estimated 2 million users in Sri Lanka in a previous post. While this is closer to the calculation of number of Internet users from our methodology, here too there maybe some issues with the data used to arrive at the estimation as the different sources of data such as ITU and TRCSL give different figures. Considering these different estimates, it seems very probable that the number of Internet users in Sri Lanka should be between 2 – 2.5 million, but due to the inaccuracy of the available data in Sri Lanka it is not possible to narrow it down further.  The only way to be certain is through regular demand side surveys and by ensuring that all official figures are consistent with each other.

Scientific Advisory Council member Johannes Bauer had developed a method to allow customers figure out the mobile packages best suited their use patterns. In 2010-11 we worked with him to convert his method to an online tool that could be used to compare broadband prices. This is now available to customers in Bhutan at the BICMA site.

Now LIRNEasia Research Fellow Christoph Stork and CPRsouth member Ibrahim Kholilul Rahman have come up with a new method to compare mobile voice prices:

Trying to apply the price basket following OECD framework to Indonesia is, unfortunately, not suitable, as the discounts and specials are mostly on a daily basis.

Therefore, a new daily basket is required to compare prices in Indonesia. Given the complexities, an alternative framework to assess prices has been constructed in this simulation. All prepaid products are being priced for a basket constituting six calls each for every hour of the day and night (24 hours) – three calls of the length 35 seconds, 75 seconds and 200 seconds to the same network (on-net) and to another network (off-net). The daily basket also includes one off-net and one on-net SMS per hour.

It is fully worth consideration, especially if it is offered as an online tool by the regulatory agency. The problem of ensuring that the fast-changing discounts are continually updated in the database remains. If that problem cannot be solved (involved commitment of resources by regulatory agency), we’ll just have declare that mobile is an FMCG (fast moving consumer good) and engage in forbearance ourselves.

Vietnam’s score and ranking on the Skills subindex remained unchanged (value of 5.72 and rank of 108), behind Sri Lanka, which was not exactly in line with its superior ranking in the Service Location Index. But leave that aside. The score in the Use subindex improved from 0.93 to 1.57, but the rank advanced by only one (77th to 76th). The 10-place jump was caused solely by the improvement of the Access subindex score from 3.11 to 4.39, resulting in the Access rank rising from 88th to 72nd.

The report explains Vietnam’s rise as being caused by “an increase in the IDI use sub-index, with 13 mobile-broadband subscriptions per 100 inhabitants, up from no subscriptions in 2008.” This is not, however supported by the evidence: the Use sub-index rank increases by only one. The Spider chart on p. 23 provides a contrary explanation: the increase was driven by the Access subindex, and within that by a doubling in the value from mobile-cellular subscriptions. Elsewhere in the text it is stated that Vietnam’s “already high mobile-phone penetration in 2008 (87 per cent) climbed to 175 per cent by the end of 2010.”

What is actually meant by 175 mobile-cellular subscriptions per 100 people? Why is it better than, say, 75 per 100 people? The only possible explanations for mobile SIMs over, say 70 per 100 people in any country are:
• There is significant multiple SIM use;
• There is no control of the reporting of SIMs, with inactive SIMs being reported along with active SIMs;
• Large numbers of foreigners are holding SIMs; or
• Machines are being assigned SIMs.

In any country, a certain number of citizens/residents do not own mobile phones, because they do not want to, or because they are too young or too old or sick. That is the reason why any number above around 70 implies one or another of the above explanations.

LIRNEasia has surveyed multiple SIM use among those at the Bottom of the Pyramid and conducted qualitative research to find out why people hold multiple SIMs. Our research shows that multiple SIM use is decreasing in South Asia, principally because of the greater controls of SIM registration and reporting by regulators. Qualitative research conducted in 2009 showed that the poor carried multiple SIMs because they were
• Elements of strategic behavior intended to keep costs down by using affinity calling plans offered by different operators;
• Responses to significant differentials in on-net and off-net call prices;
• Needed to address differences in signal coverage in different localities frequented by users; or
• Used to segregate different aspects of one’s life, such as work-related calls and personal calls or relationships with spouse and paramour.

Based on this research, I have objected to the government imposing rigid limits on multiple SIM ownership. However, that does not imply that there is anything intrinsically good in public-policy terms of people carrying around multiple SIMs. Affinity calling plans reflect actual calling behaviors. They are offered by operators with the intention of making it more difficult for customers to defect, or to reduce “churn.”

Differential off-net pricing is prima facie evidence of failures of interconnection pricing. There is nothing to be proud about patchy signal coverage, causing people to develop workarounds in terms of multiple SIM use. There is nothing wrong with segregating aspects of one’s life through the use of multiple SIMs, but should not cause a country’s ICT development to be seen as superior to anothers where that practice is less common.

So it seems that the logical thing to do is to not reward a country for having more SIMs than, say 100 per 100 people, if it is difficult to arrive at a lower number that will be appropriate for all countries.

If this was done, Vietnam would not have experienced a 16-place jump in the Access ranking. Therefore, it would not have experienced a 10-place advance in the overall IDI ranking.

If the ITU wants its ICT Development Index to be taken seriously, it will make the above correction, among others, in its methodology.

The IDI is a composite of three sub-indexes. South Asia has done well in access, with rankings increasing in all countries, except for the Maldives. India jumps seven places. The only countries that do well in the use sub-index are Sri Lanka (advance of 10 places) and Nepal (nine places). Pakistan and the Maldives fall back 13 places. Only Bhutan and the Maldives advance three and two places respectively in the skills sub-index. Sri Lanka and Bangladesh fall back one place, while all others hold their places.

What the above illustrates is that one cannot move ahead in the overall rankings simply by improving one dimension or one sub-index. For example, Bhutan advanced by four places in access, by three in use, and by three in skills. This translated into an overall advance of four places for Bhutan. Sri Lanka advanced by 10 places in use, but this appeared to be negated by a one-place retreat in the skills component and a lower-than-most advance of two places in access.

The Maldives increased by the most (0.51), yet dropped a rank position, indicating that the countries in the range where the Maldives is located is highly competitive. The least increase shown by Bangladesh (0.21) resulted in a slippage of one position in the rankings.

In his closing message, he said: “State-sponsored attacks are not in the interests of any country, long term… those governments that perpetrate them need to bring them under control.” He did not name names.

Some private-sector delegates like Wikipedia founder Jimmy Wales were less reticent.

“People do realize that there are some legitimate problems and that those problems need solutions,” he told Reuters in an interview. “The difficulty comes when you’ve got countries like China who maybe view freedom of speech as the problem that needs to be solved.”

Canadian academic Rafal Rohozinski, an expert on cyber warfare and chief executive of the SecDev Group, said the West was under pressure to regain control of the agenda on Internet governance in the face of a growing bloc of developing nations that want more influence.

“The G8, the Euro-Atlantic alliance if you like, needs to come up with an effective counter-narrative,” he told Reuters.

The picture, for them, is bad.

Fiji and PNG have both dropped four places, Fiji to 94th rank and PNG to 143rd. Both have increased their IDI scores, Fiji from 2.82 to 3.16 and PNG from 1.24 to 1.38, but as stated in the previous post one must run fast, even to maintain one’s place.

Vietnam, a high performer on all composite ICT rankings, has leaped forward from 91st place to 81st place, in a rare 10-place advance.

In South Asia, Bhutan advanced four places to 119th; Nepal by three places to 134th; and India and Sri Lanka advance by one place to 116th and 105th respectively. Pakistan and Bangladesh drop two places each to 123rd and 137th, respectively. Maldives, the leader among the South Asian countries, drops one place to settle at 67th place.

Thailand drops nine places to 89th, something that should cause concern. Philippines advances by 3 places to 92nd rank and Indonesia by one place to 101st.

The overall winner remains Korea (no change from 2008). Hong Kong SAR is at 6th place (same as in 2008). The gap with Singapore has widened, with Singapore falling back to 19th place from 15th in 2008.

The general lesson is that one must run fast even to maintain one’s rank. Singapore’s score improved from 6.71 to 7.08, but it still fell back four places.

The somewhat ironic* Broadband Commission has done something good. It has declined to define broadband either in terms of advertised (though rarely delivered) download speeds or in terms of specific technologies.

The Commission did not explicitly define the term “broadband” in terms of specific minimum transmission speeds because countries differ in their definitions. Recognizing that broadband is sometimes also defined in terms of a specific set of technologies, many members of the Commission found it appropriate to refer to broadband “as a network infrastructure capable of reliably delivering diverse convergent services through high-capacity access over a mix of technologies”. The Commission’s report therefore focuses on broadband as a cluster of concepts, such as an always-on service (not needing the user to make a new connection to a server each time), and high-capacity: able to carry lots of data per second, rather than at a particular speed.

* why ironic? If appointing as the co-chair of the Broadband Commission the world’s richest man who has, in collusion with the region’s governments, kept Latin America’s telecom prices way above developing-country averages and thus is responsible at least partly for preventing the citizens of Latin American countries from using broadband, is not ironic, what is?

Prime Minister Vladimir V. Putin of Russia met with Hamadoun Touré, secretary general of the I.T.U., this month in Geneva, where he said that “Russia was determined to contribute to the work of the union and to strengthen the collaboration with the organization,” according to an I.T.U. news release.

The official Russian government Web site carries a more detailed description of the discussions, saying Mr. Putin told Mr. Touré: “We are thankful to you for the ideas that you have proposed for discussion. One of them is establishing international control over the Internet using the monitoring and supervisory capabilities of the International Telecommunication Union.”

The I.T.U., which coordinates international use of the radio spectrum and allocates satellite orbits, among other things, also plans an international discussion on the future of the Internet during a meeting next year at its headquarters in Geneva. There, I.T.U. members are scheduled to discuss revising existing international telecommunications regulations, which were written in 1988, when the Internet was in its infancy.

Full report.

Sri Lanka had 574,000 broadband customers by end December 2010, including 294,000 mobile broadband users.

Sri Lanka Telecom (SLT), the country’s only wireline operator has been pushing ADSL (assymetrical digital subscriber line) aggressively since last year notching up 213,000 customers by end December.

Fitch said it was equal to 24 percent of the firm’s wireline base. Malaysia’s UT group is now the effective managing shareholders, being the second largest shareholder after the state in SLT.

And while there had been speculation before the E-G8 Forum that Mr. Sarkozy might call for the creation of a new international body to oversee the Internet, this idea was apparently rejected.

Eric E. Schmidt, the executive chairman of Google, said technology, rather than regulation, could take care of many of the challenges facing the Internet, including potential limits on capacity as more and more video traffic and other bandwidth-heavy content passes through telecommunications networks.

“Before we decide there is a regulatory solution, let’s ask if there’s a technological solution,” he said. “We will move faster than any of these governments, let alone all of them together.”

Full article.

There is no alternative, suggests the secretary general. “Once people have tasted the goodies of education and communication you can’t cut it off. If you cut it off you’re gone, and that’s what happened in Egypt,” says Touré. “This is a very good reason for us to encourage the rolling out of broadband infrastructure so people can have better access everywhere in the world. The citizen will be able to take care of their own destiny — if they are given this powerful weapon, a telephone, a smartphone.”
It requires some heads of state, though, to be more adult: some still surely want to clamp down on communications?

Think twice

“Yes, we have not to underestimate people. They are mature enough to take care of their own destiny. That’s what we saw in Tunisia and Egypt. I hope that anyone who is ready to cut off citizens from ICT will think twice from now on.”

Guess the odds are not good for Qaddafi (Gaddafi)?

The existing momentum of the mobile voice industry is such that we can expect considerable progress to be made in connecting most, if not all, of the world’s people through their own or neighbors’ and friends’ mobile handsets. Indicators 1 and 2 proposed by the ITU seek to document this progress at the country level. Indicator 2 in particular requires the use of demand-side survey data which is not uniformly available for all countries and for every year. The method proposed for calculating Internet users may also be used to synthesize demand- and supply-side data on mobile users and create incentives for national administrations to conduct demand-side studies.

Indicator 4 is a challenge. The only sources are demand-side household surveys. If they are conducted the ITU will be able to report results. But a backup to demand-side data does is not evident. Since it is unlikely that all countries will conduct demand-side surveys every year, it is doubtful that Indicator 4 can be reported.

The proposed solutions are interim solutions, appropriate for, and hastening, the transition from public-utility type forms of supplying ICT services to forms akin to those found in fast moving consumer goods industries that will require a complete shift from today’s supply-side dominated indicators to demand-side indicators.

Indicators Paper

As latest ITU data reveals, active mobile subscriptions continues to increase the world over. Just under two years ago, mobile subscriptions were reaching the six-billion mark. 2009 data from the ITU suggests we are well on our way to reaching seven billion connections. Developing countries, in particular, experienced a 19 percent increase in mobile subscriptions per 100 inhabitants between 2008 and 2009, compared with a modest 5 percent growth in developed countries according to the ITU.

Mobile subscriptions in the Asia-Pacific alone have now passed the two-billion mark; according to the ITU, mobile subscriptions per 100 rose by 22 percent from 46 in 2008 in 56 in 2009. However, total fixed telephone lines (which reached a peak in 2006) continue to decline both in developed and developing countries. In the Asia-Pacific, the number of fixed lines declined by 6 percent from 15 to 14 per 100 inhabitants between 2008 and 2009.

It should be noted, however, in the case of ITU data, that mobile subscriptions should not be equated with mobile ownership. Richard Heeks’ essay points out several reasons. One of them, evidenced by LIRNEasia’s Teleuse@BOP3 study, is that mobile owners may own multiple subscriptions or SIM cards, thus overstating “ownership” levels (Figure 1).

Figure 1: SIM card ownership (% of BOP mobile owners)

LIRNEasia’s third consecutive study on ICT ownership and use among the poor, Teleuse@BOP3 (2008) reveals trends in mobile and fixed growth, similar to the ITU findings. The study was among the bottom of the pyramid (BOP, or those belonging to socio-economic classification groups D and E) in Bangladesh (2008 only), Pakistan, India, Sri Lanka, the Philippines and Thailand. Figure 2 shows that over 75 percent of all respondents had used a phone in the last three months to make a phone call.

Figure 2: Use of a phone in the last three months to make a phone call (% of BOP respondents)

Figure 3 compares changes in the use of one’s own mobile for making phone calls between 2006 (when the previous study was conducted) and 2008. All countries with the exception of Thailand recorded increases in the use of personal mobiles as their primary mode of communication; India, in particular, recorded the highest growth, a massive 194 percent. Interestingly, though, Thailand actually recorded a decline in the use of personal mobiles.

Figure 3: Use of personal mobiles as a primary mode of communication (% of BOP teleusers)

Internet use

While LIRNEasia identifies drawbacks in the methodology used by the ITU in calculating Internet users,, such data, nevertheless, points to emerging trends. One significant finding is the rising number of mobile broadband subscriptions over fixed broadband, particularly in developing countries. While both types of subscriptions have grown in number, mobile broadband growth far exceeded that of fixed. Developed countries experienced a growth of 44 percent in mobile broadband subscriptions, compared with 9 percent in fixed, according to ITU data. Likewise, developing countries reported increases of 63 percent and 24 percent in mobile and fixed broadband subscriptions, respectively.

In the Asia-Pacific, mobile and fixed broadband subscriptions have grown to 5.4 and 4.6 per 100 inhabitants respectively, a 26 percent and 18 percent change between 2008 and 2009. Again, one should keep in mind that subscriptions do not necessarily equal ownership or usership; the problems in accurately assessing the number of Internet users in a country (besides through large-scale demand-side surveys which are nationally representative; Teleuse@BOP is only representative of SEC D and E in the respective countries) is something that LIRNEasia is looking at in research which will be released shortly.

LIRNEasia’s study confirms that that the BOP were much more likely to own a mobile phone than a household personal computer, and, likewise, more likely access the Internet over a mobile than a PC (Figure 4). However, Internet awareness and use among the BOP is still low, particularly in South Asia. Over 95 percent of respondents in the South Asian countries studied had never used the Internet before; over 55 percent in Bangladesh, Pakistan and India had never even heard of it. This contrasts with the Philippines and Thailand where only 9 and 20 percent of respondents reported never having heard of the Internet.

Figure 4: Ownership vs. use of ICTs (% of BOP teleusers)