31/07/06

PCs for the poor: as good as their hype?

A prototype of the OLPC's US$100 laptop Copyright: laptop.org

Send to a friend

The details you provide on this page will not be used to send unsolicited email, and will not be sold to a 3rd party. See privacy policy.


Technologists are at odds over how to bridge the digital divide. What one group calls the ultimate solution, another dismisses as “the scam of the century”, reports Waleed al-Shobakky.


At the 2005 World Economic Forum in Switzerland a soft-spoken academic made an announcement that sent seismic waves across the computer industry. Nicholas Negroponte, then director of the Massachusetts Institute of Technology’s Media Lab, spoke of making laptops available at US$100 for schoolchildren in developing nations.


The price was not the only big news. Negroponte named companies that had agreed to collaborate on what would become the One Laptop Per Child (OLPC) project.


Notably, the list did not include Microsoft and Intel, the world’s largest software and microchip manufacturers, respectively. Instead, the laptop would use a processor from Advanced Micro Devices and an operating system based on Linux, whose code is freely available for anyone to modify and distribute.


The technology race


A cascade of announcements followed. In November 2005, Negroponte demonstrated the first prototype at the UN’s World Summit on the Information Society in Tunisia.


Microsoft soon responded. At the January 2006 Consumer Electronics Show in Las Vegas, United States, Bill Gates unveiled his company’s answer: a prototype cellphone that could be turned into a computer by connecting it to a TV and keyboard.


Microsoft touted its Cellular PC as a cheaper and more practical alternative, since it relied on components already in use. Craig Mundie, Microsoft’s chief research and strategy officer told The New York Times, “everyone is going to have a cellphone … in places where TVs are already common, turning a phone into a computer could simply require adding a cheap adaptor and keyboard.”


In February, leading Taiwanese microchip manufacturer Via showed off prototypes it has developed for its PC-1 initiative, which aims to enable a billion people to connect to the Internet using cheap and energy-efficient notebook and desktop computers.


Then in May, Intel’s chief executive officer Paul Otellini launched his company’s foray into the low-cost computing arena with EduWise, a US$330-400 ‘education notebook PC’ aimed at developing nations.


Intel spokesperson Mike Green says EduWise has an advantage over the OLPC laptop in that it can accommodate more standard software and tools. Its hardware will be almost identical to that of regular computers. Moreover, it will use Microsoft’s Windows operating system, which runs about 90 per cent of the world’s personal computers and is compatible with more software applications than most other operating systems.


Negroponte’s US$100 laptop, with its trimmed-down hardware, has drawn criticism from Otellini. “We do not think you [can] cross the digital divide with old technology,” he told The New York Times in May. “[EduWise] doesn’t need exotic technology and it runs real applications.”


These are not the first attempts to provide affordable computing for the poor. Back in 2003, Advanced Micro Devices introduced its Personal Internet Communicator, a relatively cheap box that could be used to access the Internet when it was connected to a screen and a keyboard. But it couldn’t be used as a word processor, and it cost US$400 — both major weaknesses. The company ended up pulling the plug and partnering with Negroponte instead.


Starting simple


Prior to all these initiatives was the Simputer — a simple, inexpensive, multilingual people’s computer, probably the first serious effort to develop an affordable computer.




The Amida Simputer
Credit: (PicoPeta) 

The first prototype, launched in mid-2001, was a handheld device close in size to today’s PDAs (personal digital assistants). It was the product of a collaboration between Indian entrepreneur Vinay Deshpande and scientists at the Indian Institute of Science, including Swami Manohar, who came up with the device’s name.


Simputer’s pioneers believe that the current excitement over low-cost computing initiatives is largely due to their project. But they accept that, for several reasons, their device was not widely adopted. “Perhaps we were somewhat naïve back in 2001 when we announced the Simputer without having ready-made applications for it,” says Deshpande.


Moreover, he says, because it looked like a PDA, people expected it to function like one, and were frustrated that it did not have the regular applications.


Manohar says that the Simputer concept needed redefining. What he learned from the initial Simputer launches is that “the under-privileged want to have the same technologies the privileged have, not some cheap stuff that do-gooders provide for them”.


The first Simputer was, for power and cost reasons, a monochrome handheld device. “When we launched it everybody was asking why it did not have colours in the era of colour television,” he says. “That is when we redefined the [Simputer] to be a world-class product that can serve everyone,” says Manohar.


Deshpande and Manohar are now the chief executive officers of Encore and PicoPeta, the two companies that have licensed the Simputer platform.


They say that the evolutionary routes that Simputer took after its initial launch were in response to market realities.


PicoPeta advanced the original Simputer platform into a handheld with features comparable to those in regular PDAs. Meanwhile, Encore retained a version of the Simputer similar to the original one and used it as a basis for more machines, including Mobilis, a tablet-PC-like handheld with a seven-inch display and support for applications such as word processing and spreadsheets.




The Encore Mobilis
Credit: (Encore Software)

Media blessings and curses


After Simputer failed to live up to expectations, Negroponte’s US$100 laptop rapidly became an international media darling, portrayed by most as the final word on how to bridge the digital divide.


And it is probably thanks, at least in part, to this favourable coverage that six countries — Argentina, Brazil, China, Egypt, Nigeria and Thailand — have each pledged to buy one million units, even before putting their hands on the final product or knowing its exact specifications.


India had also shown interest but this month pulled out. Education secretary Sudeep Banerjee said the laptops could be “detrimental to the growth of creative and analytical abilities of the child” and that the money would be better spent on more classrooms and teachers.


The OLPC project is no stranger to criticism. The huge visibility of the Negroponte’s machine brought it under fire from IT and economic development experts who believe its potential — and that of any other low-cost computing device — may have been overestimated.


“Since the field is so new, nobody really knows an answer to the question ‘What is the role that technology will play in developing communities?’,” says Mary Bernardine Dias, director of TechBridgeWorld, a programme initiated at Carnegie Mellon University, United States, that aims to put advanced technologies to the service of poor communities.


Dias believes that the excitement over the latest initiatives does more harm than good. “Hype builds expectations, and in many ways fixes the expected outcome of a project before it is launched,” she says. “We all need to be flexible when we launch these projects and [should] expect the unexpected.”


“I have not seen any of the current projects addressing all of the lessons learned through the past endeavours,” says Dias. “Technologies such as the Simputer that were launched as computing for the poor did not live up to their hype.”


Najat Rochdi, coordinator of Information and Communication Technologies for Development in the Arab Region, a UN Development Programme initiative, believes one lesson learned is that new solutions are not necessarily better.


She cites the ‘PC for Every Household’ programme adopted in Egypt since 2002. It was a collaboration between the Egyptian government, local banks, local and multinational computer retailers and software manufacturers to provide computers for low-income families at affordable monthly payments.


“I consider this programme a success,” says Rochdi, noting that it widened the base of computer ownership without compromising on quality.


Another criticism is that the latest initiatives focus too much on low cost instead of the technology’s appropriateness. “Focusing on price alone is a distraction from the essential goal, which should be technologies that meet the actual needs of people in local communities,” says Rochdi.


Several initiatives in the past decade introduced computers to schools and public offices with negligible gains in productivity, meaning “you end up paying much more than you saved on the technology,” she adds.


She is also concerned about the little attention paid to local languages.


Rochdi believes that affordable computers are necessary, but “content is more important. It takes a generation of children to build content that really helps empower and advance a new kind of citizenship [for generations to come]”.


A final reservation against the latest entry-level computing initiatives is that they have anchored the word ‘computer’ in people’s minds to be synonymous with ‘technology’.


Rochdi points to the utility of using radio, TV and personal digital audio (MP3) players to share information and educate people. She says policymakers and the public are fixated with the belief that computers and computer-related inventions, like the Internet, are the only drivers of development.


Education first


Naturally, the proponents of the latest inexpensive computing devices rebuff these criticisms. The OPLC project’s director for the Middle East and Africa, Khaled Hassounah says, “Knowledge generally and education specifically have always been key factors in empowering people. Computers could act as great agents for learning, enabling children to interact with concepts in ways that their environment and traditional educational tools did not allow them before.”


On the issue of cost versus appropriateness, Hassounah says OLPC is about “education first and technology second”, stressing that the aim is not to build a cheap laptop, but one suitable for children to learn with in an environment that traditional computers were not designed for.


“If any manufacturer starts offering a laptop at continually declining prices and with features that solve rural and school-environment use challenges, we will switch to their laptops,” he says.


Intel also underlines the educational quality of its products. “The objective is to provide affordable, collaborative learning environments for teachers and young students,” says spokesperson Mike Green.


The upgrade cycle


While the debate over the relevance of low-cost computing for the poor goes on, the producers of the different models remain at odds over the relative merits of their machines.


The strongest criticism of the OLPC comes from Manohar of PicoPeta who says: “marketing is far ahead of the product. I will even classify this as a scam”.


Manohar used his blog site to outline what he considers the project’s blind spots, describing Negroponte as a “marketing juggernaut” and the countries that have promised to buy the one-million laptop batches as “gullible”.


Hassounah counters that, saying, “it is always easier to attack the messenger than to actually listen to the message”. He says Manohar’s criticisms are “wrong or inaccurate at best”.


But on one point the two agree. The Simputer inventors tried to address a dominant phenomenon in the PC industry — Wintel architecture. Wintel is short for Windows-Intel, a reference to machines or systems that use Microsoft’s operating system along with Intel’s processors.


“With the exception of the OLPC, all the latest initiatives are based on the Wintel architecture, which is not designed to work in developing economies,” says Deshpande.


Microsoft’s software, he explains, needs frequent upgrades. The newer versions may not run on old processors, so you have to upgrade the computer’s hardware all the time. Eventually, he says, you will be forced to replace your PC even if it is still good enough because there is no technical support or spare parts available for older models.


The upgrade cycle is not the Simputer inventors’ only criticism of Wintel architecture.




Vinay Deshpande, co-founder
of the Simputer
Credit: (Waleed al-Shobakky)

“It is a power guzzler,” says Deshpande, pointing out that in most of the developing world, where power is either unavailable or unreliable, the Wintel-based machines will not operate effectively. “Today, a four GHz Wintel-based machine consumes anywhere between 200 to 400 watts, whereas a Simputer needs only eight watts or less.”


He argues that the high power consumption and frequent upgrades mean that Wintel-based systems are too expensive for developing nations, even if the computers are provided free through foreign aid. “If you really want to bring costs down you need to take a fresh approach,” he says.


In April, eWEEK website reported that Negroponte voiced a similar opinion. “I have come to a conclusion that every new release of software is distinctly worse than the other … there’s a natural tendency to add stuff.”


However, Deshpande acknowledges that the microchip industry is now addressing the earlier shortcomings by minimising the processor’s power consumption and integrating many functions — like data processing and connectivity — into one chip. Which means Simputer’s advantages in terms of cost and power could soon disappear.


“We are not dogmatic,” Deshpande says. “One to two years down the line, the world will be different, and so will we. We will stick to the concept, not the hardware specifications.”


As the debate continues to rage, and companies continue to launch new ‘ultimate solutions’, Simputer’s inventors draw satisfaction from seeing what they have started. “This is a vindication of the Simputer philosophy; that is, you need low-cost information technology for the masses,” says Deshpande.