At the World Summit on the Information Society two weeks ago, MIT's Nicholas Negroponte unveiled the laptop he believes will digitize the developing world. The cute green computer sports a WiFi card, a 500 MHz processor, a 1 gigabyte flash drive, and a novel power source—a 6-inch hand crank that juts out from the side. It will run free, open-source software, most likely some derivation of Linux. All of this for the low, low price of just $100.
Negroponte promises that bringing cheap laptops to countries like Brazil, Thailand, and Egypt will help "children to 'learn learning' through independent interaction and exploration." That might be true, but this green machine won't be the computer to do it, no matter how much Kofi Annan and the international press fawn over it.The $100 laptop is a huckster's gambit—poorly thought out, overly ambitious, and too sexy to be true.
The $100 price point is the obvious grabber—Negroponte's computer wouldn't make it out of the back of the business section if it cost $499. But is it possible to build such a computer so cheaply? Negroponte says on his site that the screen alone costs $35. A 1 gig flash drive retails for around $70, a WiFi card for at least $25, the RAM perhaps $50, and the hand crank who knows how much. Add in labor, distribution, service, and maintenance costs, and you're over $100 by a couple hundred dollars. If you're willing to assume that MIT can somehow keep the cost at or near the century mark, there's still the question of who will support the computers (and who will pay for that support). When the computers are already in India, there's nowhere to outsource the help desk.
Negroponte claims he won't have a problem hitting the $100 mark because he's only selling to government buyers who purchase a minimum of 1 million computers. But it doesn't seem likely that any client will buy a single computer, much less a million, until they're convinced the computer will work and be as cheap as advertised. Besides, does the Thai Ministry of Education really have a couple hundred million dollars sitting around? According to the MIT Web site, "[m]anufacturing will begin when 5 to 10 million machines have been ordered and paid for in advance." Don't wait around for those conveyor belts to start cranking.
What Negroponte doesn't want Brazil, Thailand, and his other clients to know is that this isn't the first time a $100 computer has briefly seduced the press. Just like the little green laptop, each of these machines came with a gigantic catch.
In 2000, a now-defunct Texas company called Netpliance marketed a $99 all-in-one computer called the i-opener. The catch? You also had to buy their Internet service for around $20 per month. After a series of shipping delays, price increases, and customer complaints, Netpliance stopped selling the i-opener in January 2001. A few months later, the company settled a Federal Trade Commission complaint that they used deceptive advertising to hide the machine's true cost.
In 2001, a group of computer scientists in Bangalore, India, developed the Simputer. It was supposed to be a cheap (around $200), robust computer for India's rural poor. But according to the Associated Press, the brains behind the Simputer have sold only 4,000 of an expected 50,000 units in 2004 and 2005. In addition, only about 10 percent of Simputer buyers live in rural areas. Why? Probably because they have more important things to do than write e-mail.
There's no reason to think that Negroponte's computer will win wider acceptance in the Third World. The fact that each laptop comes with a built-in WiFi card won't be of much use if there isn't a WiFi access point nearby. How many access points do you think there are in rural Egypt?
So, what's a better way to do this? Negroponte should check out some of his competition in the low-end computer market. Dell will sell you a new laptop with a DVD player and a WiFi card for about $500. Linspire has been selling their Linux-based desktops at places like Wal-Mart for $300 for a couple of years now. AMD's Personal Internet Communicator, which retails for $300 (or $190 at Radio Shack after rebate), runs Windows CE and is compatible with existing Microsoft products like Word and Excel. (Both the Linspire and the AMD machines are sold without monitors.)
There's also an American company that's already making a profit on a $100 machine—the Palm Z22. Sure, the Palm PDA doesn't have WiFi, but you can use it to take notes and it has a color screen. The point is that there remains a trade-off for price and functionality. Negroponte can't have it both ways.
There is a much better way to build a computer that can serve the rural poor, if you're willing to pay for it. The San Francisco nonprofit Inveneo builds Linux boxes with solar panels, a bike generator for cloudy weather, and a directional WiFi antenna that can get you on the Internet via satellite from almost anywhere. Inveneo can build one of these setups for $1,800, but they're only making a small handful. Increase the production run to tens of thousands and the price would drop precipitously.
Inveneo is building these machines to be durable, upgradable, and to run on renewable energy. Sure, they're not laptops, but they give anyone access to cheap worldwide voice communication. That's something that even someone who is illiterate can take advantage of—without having to use a hand crank.