Talk about being an optimist.

Nicholas Negroponte, founder and chairman of MIT's Media Lab, is heading up a project with the goal to distribute specially designed laptops, each with a price tag of less than $100, to more than 150 million schoolchildren within the next three years. That's an amazing statistic, considering that PC makers will manufacture only about 50 million laptops this year.

In fact, getting manufacturing partners to produce so many laptops presents the program's biggest hurdle. Negroponte says producing 150 million of anything is a significant challenge. Though daunting, he says, the problem is being solved by "mere resolve."

Another problem that must be addressed by the not-for-profit group, One Laptop Per Child: a possible gray market in which laptops distributed free to children are instead illicitly sold by parents or others, especially those in third-world countries. One way to thwart this is allowing PC makers to manufacture commercial versions of the laptop at a relatively inexpensive price, say $200 each, with part of the profit going back to One Laptop Per Child.

Negroponte contends the laptop is a tool to help children teach themselves, a more cost-effective alternative than hiring additional teachers. "A lot of learning comes form explorations, interactions, curiosity," Negroponte said Wednesday at an MIT emerging technology conference sponsored by Technology Review, an MIT magazine. "That's how we learned how to walk, how to talk. It's the kind of learning kids do very well. This [laptop] is a tool to make that more continuous, seamless versus … today, when we say at age 6, 'Stop learning that way, and learn by being told by books and teachers.'"

A prototype of the laptop will debut in November at the World Summit on the Information Society in Tunis, Tunisia. The first 5 million to 15 million laptops could be distributed by the end of next year or early 2007, with another 100 million to 150 million to be handed out a year later. The first countries to distribute the laptops will be Brazil, China, Egypt, South Africa, and Thailand.

How can One Laptop Per Child produce such a cheap laptop? Half the cost of a commercial laptop consists of sales, marketing, and distribution costs, as well as profit, items that don't weigh down a laptop designed for a not-for-profit program. One Laptop Per Child will market the laptops directly to national ministries of education, which can distribute them like textbooks.

Commercial laptops require operating systems that can handle byte-intensive applications that Negroponte characterizes as fat and unreliable and which slow down laptop performance. Two-thirds of commercial laptops' software manages the other thirds, which mostly does the same functions in a variety of different ways, he said.

The first-generation machine likely will employ a dual-mode LCD display found in inexpensive DVD players but that can also be used in black and white, in bright sunlight, and at four times the normal resolution--all at a cost of less $30.

Despite cost-cutting, these machines will be robust. Negroponte envisions a Linux-based, full-color, full-screen laptop that will use innovative sources of power--including wind-up--and will be able to do most everything except store huge amounts of data. These rugged laptops will be Wi-Fi- and cell-phone-enabled, and have at least four USB ports. Its current specifications are: 500 MHz, 1 Gbyte, 1 megapixel.

Laptops distributed to schoolchildren in a specific area will create a peer-to-peer mesh network, a process developed at the MIT Media Lab.

Negroponte sees the $100 laptop program as a change agent in the way to educate children worldwide. "All we can do is seed the change," he said, "and like Wikipedia grew, and like Linux grew, do [the same for] open-source education."