One Laptop Per Child marches on

THE ONE LAPTOP PER CHILD FOUNDATION has had what might be described as a difficult journey since its birth in January 2005. It has been plagued by bad press, poor relations with the computer industry and an outspoken chairman who has been described as "erratic".

Nicholas Negroponte, the driving force behind OLPC, has a great talent for bringing change, however. When he first announced that the foundation would produce seven million $100 PCs which would be distributed to some of the world's poorest children, some pundits said he was crazy. A million tiny, bright green, underpowered laptops later and netbooks are the only growing PC market segment, according to industry reports.

When we suggested a few days ago that the OLPC project ultimately had been a failure, there was such a furious response from both supporters and detractors of the foundation that we felt it only right to take a closer look at what OLPC along with its diminutive and much-maligned XO laptop has achieved. And what it hopes to achieve in the near future.

There can be no doubt that OLPC has so far failed to reach its original goals. First, the $100 price point was wildly optimistic. Current estimates put the cost of each XO laptop at around $188. That's the cost of building each machine and doesn't include marketing, distribution, taxes or retailer mark-up. With Eee PCs currently retailing at around $300 it doesn't take a mathematical genius to work out that the build costs are probably quite similar. Box builder Quanta is also believed to have offered dramatically reduced rates for the first million units produced, so the chance of a price hike in the very near future cannot be ruled out.

And then there's that seven million number hanging over the foundation. There's no doubt that getting a million laptops into the hands of poor kids in over thirty countries is quite an achievement, but if you told your employer that you were going to work for seven hours tomorrow, and you only turned up for one hour, that could be seen as a failure on your part, as one INQ commenter pointed out.

Sean Daly from Sugar Labs, the company that develops the XO's kid-friendly front-end software, maintains that OLPC is still achieving some great results despite press reports to the contrary. "OLPC is concentrating on getting laptops to kids and not its image in the tech press," he told us.

"As a result, many pundits repeat inaccurate and incomplete information. It is true that Professor Negroponte's stated goals seem grandiose in retrospect. 'Only' one million laptops are in classrooms in 40 countries, rather than the seven or 10 million projected.

"They cost more than $100 each and many pundits claim the project 'failed' by not yet meeting that goal. However, when countries such as Uruguay and Peru provide a laptop to every single student in the country, they are making an investment in their next generation of students. Western countries are not making that commitment."

But XO laptops are still rolling off Quanta's Taiwanese production line, and Negroponte has claimed that another million machines are in the pipe. "The million laptops, our little green ones that are in the hands of children, are currently in 19 languages and 31 countries," he said. "Another million are on their way. Not bad. But even better, these countries include Afghanistan, Haiti, Ethiopia, as well as places like the West Bank and next month Gaza."

And the hardware is also moving on, with a refresh expected any time now. The XO 1.5 will have a whole new motherboard powered by a VIA C7-M chipset that replaces the current AMD Geode. Details are scant at time of writing but the new chipset is reported to be significantly faster than the first iteration. The new mainboard along with a larger SSD and more RAM will be shoehorned into the same bright green clatter-proof casing as the original.

There are currently 300 beta 2 XO 1.5 machines for developers and testers doing the rounds, according to Daly. "Quanta is doing this short run to be sure they can ramp up to large-scale production in a couple of months or so," he told us. "The software is not completely done yet and there may be hardware kinks we don't know about. It's important to realise that OLPC has not announced a release date for the machine. They want to get this one out the door, then resume work on the XO-2 dual-display no-keyboard model announced previously."

The first batch of new machines will be used to fulfil existing orders at the same price as the older model, but OLPC has hesitated to communicate hard numbers when it comes to distribution, since Negroponte's bold predictions of millions of laptops failed to come to fruition.

And according to sources close to the project, this isn't the only bold claim erroneously made by the well-meaning but sometimes misguided chairman. "He has a tendency to make pronouncements without checking with others what is possible," an insider told us. "A few months back he advised Microsoft to release Windows 7 for the ARM processor, announced for the next-gen XO-2 dual-screen laptop. Microsoft has no full Windows version for ARM and has not planned one. He hadn't spoken to Microsoft before offering this advice. Microsoft replied with a statement calling into question Professor Negroponte's working methods."

Despite all of this, there is still a chance that newer versions of the OLPC hardware will be capable of running Microsoft's much-anticipated Windows 7 operating system, as Daly explained.

"I'm told the motherboard of the XO-1.5 more closely resembles a 'classic' netbook, and the specs are higher, so it is possible Windows 7 will run on the XO-1.5 with little or no modification. Then again, drivers for the XO's exotic hardware - including the webcam and the keyboard with lots of special keys - may be necessary. I'm not aware of anyone from Microsoft working on it. They have been throwing their weight behind Intel's World Ahead initiative with Classmate PCs, just relaunched in Western Europe by Archos [with Chinese-built hardware]. In my view, the real question is, can Windows 7 can be installed from a USB stick? There are rumours to this effect with Microsoft wanting to provide an upgrade or GNU/Linux replacement path on netbooks."

But Microsoft's relationship with OLPC has not always been an easy alliance, as Daly points out.

"A lot of noise was made about Microsoft 'taking over' the project. In fact, although Microsoft did try to replace the OpenFirmware with a BIOS more friendly to Windows, Professor Negroponte insisted that countries have the choice between Windows and Sugar, the child-friendly learning interface which runs over GNU/Linux [in OLPC's case, Fedora].

"I am told that, although nearly all countries ask about Windows when evaluating OLPC laptops for children 5 to 10 years old, Sugar wins - it is better adapted to kids, has built-in collaboration, automatic backup, and is free, unlike Windows. Smaller children, especially those seeing a computer for the first time, are confused by the classic Windows interface. As an indicator, many Intel Classmate projects add a proprietary locked-down kids interface over Windows, such as EasyBits Magic Desktop. This adds to the cost. So in the end, to date all countries have preferred Sugar; the less than one per cent running Windows XP were pilot projects or were donated by Microsoft. Of course, that could change if Windows 7 runs well on the new XO 1.5, but I suspect the added cost, plus the additional cost of a kids' environment, will discourage countries from making that choice."

Perhaps OLPC's most visible legacy is that the introduction of a cheap, easy-to-use laptop has kick-started a whole new generation of personal computers. It could be argued that, without the XO, netbooks would never have seen the light of day, let alone become the dominant force in a confusing and fragmented PC market. But did big computer manufacturers steal all of OLPC's altruistic ideas for corporate gains? "The XO was never meant for the mass market," Daly says.

"It was meant for government distribution to schools. But OLPC broke new ground by throwing out the rules and starting from scratch. It can be argued whether these were the best strategies or not.

"However, I do consider that Intel's World Ahead program specifically targeted OLPC's positioning. Intel has found success [around 400,000 Classmate PCs have been distributed] by working on local production with the largest developing countries, the same countries that would have given OLPC the opportunity to scale to millions of machines. Professor Negroponte has indicated that the XO 2 to come may adopt this model of local production.

"Asus took the OLPC idea - small form factor, flash memory SSD, light GNU/Linux system, SD-Card slot, ergonomic desktop - and launched the netbook craze in the mass market. At Sugar Labs, we welcome inexpensive netbooks as an opportunity - every machine we have tested so far boots Sugar on a [USB] stick."

At the end of the day, it's not for us to judge whether OLPC has been a success. It is down to the more than one million underprivileged kids all over the world who have been give an opportunity to expand their education using the tools that the rest of us take for granted.

"It's amazing is to see what kids do when they see each other on a mesh WiFi network in the classroom, and can share the same document or activity in Sugar," enthuses Daly. "This is perhaps the most special part of why OLPC works in the classroom, yet sadly, very few journalists or pundits have been able to experience it." µ

L'Inqs
OLPC
Sugar Labs