The End of The One Laptop Per Child Experiment--When Innovation Fails.

Nicholas Negroponte’s One Laptop Per Child organization admitted defeat in its effort to sell millions of open-source computers in Asia, Africa and Latin America by joining with Microsoft to load Windows XP onto its green and white laptops. The decision marks the end of the effort to spread Constructionist learning pedagogy—learning by doing—to tens of millions of poor children in villages around the world.

However you view Piaget’s learning philosophy (it works for me), the retreat from a global effort to bypass teachers, parents and the educational establishment in India, China and elsewhere to promote self-learning by children holds very important lessons for anyone interested in promoting innovation and change. The original goal of OLPC was to use open source software to connect children directly to one another and the web so they could learn from one another and directly from many sources of information. Ivan Krstic, the key software architect at OLPC, explains this on his blog. That’s the heart of

Seymour Papert's Constructionist theory of education. Open.

The problem from the very beginning was that this is a Western educational concept encased in a beautiful little childrens' laptop designed by Westerners (Boston-based Continuum and fuseproject's Yves Behar) for non-Western children and non-Western cultures and educational institutions. The education ministeries in India, China and elsewhere saw OLPC as a challenge to their authority and their abilities. After all, the rise of China and India and the lifting of half a billion people out of poverty in the shortest period of time in history is based on their existing educational institutions. They argues that with US companies chasing Chinese and Indian school graduates, why change their systems to conform to some Western ideal of learning?

And with the US moving to testing children with No Child Left Behind (with its emphasis on rote learning and teaching to the test), why experiment with the children of Asia, Africa and Latin America? So went the argument.

OLPC set a target of selling 100-150 million laptops by the end of 2008 and so far it has placed one order of 100,000 computers with Uruguay, a well-off country with middle class, not poor, children in most of its schools. Constructionism may well be better suited to these kids.

The OLPC has consistently been lauded in the US in terms of its design and its technology, not its underlying education pedagogy. But it is the pedagogy that has always been the crux of the experiment and it is the pedagogy, in the end, that proved unacceptable to governments around the world because they felt it insulted and challenged them.

The lesson here is that however brilliant the innovation, it needs to be appropriate to the context and the culture. It needs to fit in and not be imposed. And it needs coalitions, teams, to support it. In fact, in the case of education, which is extremely politically sensitive in every country, OLPC should have developed both the design of the computer and the pedagogy with the Indian and Chinese teachers and administrators, not for them.

With Microsoft's XP software now being loaded, the One Laptop Per Child XO laptop becomes just another inexpensive (the price will be around $200, double the original estimate) machine competing with Intel's Classmate and others. It's one virtue is that it will run all the educational software being produced by educators in Asia, Latin America and Africa.

Before it's here, it's on the Bloomberg Terminal.