Rohan Samarajiva, Author at LIRNEasia — Page 38 of 182


Network shutdowns in the news again

Posted on March 22, 2016  /  0 Comments

This time it’s on account of elections in Congo. For an overview of the phenomenon, see CDT’s map and timeline. We first wrote about this problem in 2006. Then it was worked up into Gyanendra’s Law and its various exceptions. Image source
BSNL is on life support courtesy of the Indian taxpayer, whatever fictitious paper profits it conjures up. The authors of a recent op ed in Hindu Businessline see through the fiction. The government firms had their own share of sops such as (i) year-ahead early start in the assignment of both 3G and Broadband Wireless Access (BWA) spectrum in 2009 and (ii) reimbursement of about Rs. 10,000 crore of licence and spectrum charges during 2001-06, as part of commitment to BSNL corporatisation. However, the 6.
The nationwide unplanned outage last Sunday has obviously got everyone upset. While some are trying to make political capital out of it, the responsible thing to do is to understand the causes and act to avoid a recurrence. This appeared to be the objective of the reporter who interviewed me last week for this article. There is one place where the report does not exactly reflect what I said. I was asked whether I was happy with CEB’s /PUCSL’s role.
LIRNEasia was early in getting into systematic reviews. We have faced considerable challenges when communicating the findings. We are working on pulling together what we learned in the process for a special issue of Information Technology and International Development here at a nice, quiet hotel in Galle.
We are aware that the UN has identified tourism data as priority area in terms of exploring the potential of big data to contribute to the work of national statistical organizations (NSOs). However, this was not something we took on, given our programmatic priorities which are urban development, improved socio-economic monitoring and epidemiology. When we were asked to share ideas on tourism data and a few other areas by a major business group, we did apply our minds to the problem. Here is the slideset. What are the key ideas?
In our teaching-focused comparative work on electricity, we found there was a fundamental difference in the way the problem of costs was approached in Sri Lanka and on the sub-continent. In Sri Lanka, the focus was on the costs of NOT having power. In India, the focus was still on the costs of inputs, per se. That is, they cared about the costs of switching on another power source to meet peak demand. On that basis, they got along with load shedding and low prices, around half that charged from Sri Lankan subscribers.
We generally know how to measure performance in the telecom sector: increased connectivity in voice and data; lower prices; improved quality of service experience; and greater choice. Similar in electricity. In each of these cases we can also identify the factors that led to improvements in performance. Recently I was thinking about the healthcare sector. This sector has commonly accepted, internationally comparable indicators such as the infant mortality and maternal mortality rates.
I had been invited to moderate a panel discussion on consumer rights in electricity, in the context of a recently issued charter of consumer rights and obligations. This was set to be a ho-hum affair, until the country experienced its third nationwide blackout within the last six months. This resulted in the shutdown of the 900 MW coal-powered plant, which means that the system will be in distress for 4-5 days until they get it fired up again. Since 2002, Sri Lankans have got used to uninterrupted power which they pay a lot. There is a lot of anger.
Even before we officially launch the gender study, it is being used in Myanmar. A recent report on gender and connectivity by GSMA and Sri Lanka based think tank LIRNEasia said that mobile phones have come to symbolise status in Myanmar. “Despite the cost barrier, people are often not interested in keypad phones or less expensive smartphones, which are also perceived as low-quality,” said the report. Mr Meza said that outside Myanmar’s cities, the phone would find its audience. “In Yangon people will go out of their way to get … the latest iPhone, Samsung, HTC [device], but the country has 680,000 square kilometres so life doesn’t end in Yangon.
Appears to have been landed at both ends, la Seyne-sur-Mer, in France and Tuas in Singapore. The contracts were let to different parties, Alcatel Lucent and NEC, for the France-Sri Lanka and Sri Lanka-Singapore segments respectively. It is not evident from the web that the landings have been made in Sri Lanka yet. SingTel has announced it has completed the landing of the SEA-ME-WE 5 subsea cable at Tuas in Singapore. The 20,000km Southeast Asia – Middle East – Western Europe 5 cable is expected to be complete by the end of the year.
The NYT Sunday Review carries a fascinating piece on how US and European wildlife officials are using the full panoply of ICTs and big data analytics to manage eco-systems and human-animal conflict. I’ve always felt that Beniger’s discussion of control was central to any realistic understanding of what is happening with big data and ICTs. What happens with animals today may happen with humans tomorrow. Starting in the early 2000s, the recovery program employed ancient and contemporary technology: Net-guns, fired from helicopters, were used to capture bighorn outfitted with collars that carried both GPS and VHF radio transmitters; professional hunters, meanwhile, tracked and darted every mountain lion in the area to outfit them with collars that carried VHF radio transmitters. Biologists at computer monitors began to watch bighorn movements.

Network resilience gains momentum

Posted on March 8, 2016  /  0 Comments

In the policy world, one does not want to be alone. I have even dressed up new policy ideas as variations on existing ones, in order to get them accepted. When Abu Saeed Khan persuaded me that international backhaul was an important issue in Islamabad in May 2010, he was quite alone. When I made the first presentation on the subject to the expert group at ESCAP in November 2010, LIRNEasia was a lone voice in the wilderness. Abu then took the lead role.
I was invited to speak at the opening session of the ESCAP training workshop organized by the ICT and DRR Division, March 8-9, 2016. This was to introduce the report we had prepared for ESCAP on Building e-resilience, which is about to be released. The slides I used in my presentation are here. Because there was enough time (unusually), I went into some depth on one recommendation per inter-governmental organizations; governments and telecom service operators. Insurance appeared under two headings and took much of the time devoted to discussion.
That’s title of a report Sriganesh Lokanathan and I completed for the New Venture Fund. Here is an extract from the executive summary. Much of the discussion of the socio-economic implications of behavioral data has focused on the inclusion of more citizens and more aspects of their lives within the sphere of control enabled by pervasive data collection. Effective public policy rests on good information about problems and the efficacy of the deployed solutions. Governments obtained such information through National Statistical Organizations (NSOs) in the 19th and 20th Centuries.
I’ve been writing about Indonesia’s tsunami buoys for a while. This was when I was trying to dissuade people in Sri Lanka of the need for our own tsunami detection system. We are not located in an earthquake zone and thus only vulnerable to teletsunamis that come from the Sunda Trench. Indonesia and Thailand need to sensors because of their proximity to the Trench. But little did I know that the Indonesian system suffered from the project syndrome: money for installation but nothing for maintenance.
We have always said that big data is about control, in the soft form first described by James Beniger. Information and control are closely connected. Beniger (1986, pp. 7-8) states that the twin activities of information processing and reciprocal communication (or feedback) are inseparable from the concept of control. Control is defined in the broadest sense as “purposive influence toward a predetermined goal.