I was invited recently to give a presentation, and when I offered the One Laptop per Child (OLPC) project as the subject, the initial reaction of the conference organiser NetEvents was: “We want something more controversial for our audience – the project is too obviously a good thing.” My reaction was: “If only.”

Educational outreach for poor children is so clearly a worthy cause, but that does not guarantee all-out support.

On the contrary, there are revolutionary implications at many levels. I have even been labelled a telecommunications terrorist by one leading telecoms firm, so I will focus here on the more controversial aspects of the OLPC project.

There are two opposing currents in the electronics industry: one is that prices are constantly falling, the other is that it is necessary to keep adding more features to maintain the high price.

Today’s laptops are almost the same price they were 10 years ago, but the devices are now stuffed with extra features.

As providers pile on features, the system grows fatter. The OLPC laptop is not like that. It does not rate highly on processor power, but it is leaner and fitter, and that goes against the accepted business culture.

As a director at Motorola, I was pushing for years for the organisation to make a low-cost handset but the firm resisted.

When Motorola finally did announce it was releasing a low-cost unit, the company’s stock fell.

However, the financial markets got it wrong, because that basic handset now serves a vast global market.

Early on during the OLPC initiative, I met with the chief executive of a large display manufacturer and said that I needed a small display, which did not have to be very bright or have perfect colour uniformity – but it did have to be very inexpensive.

He told me that his firm made big, high-quality displays and that his strategy and my project were incompatible.

I told him that was a shame because I would need 100 million units per year – and he replied that he could possibly change his firm’s corporate strategy.

The numbers are staggering: while the worldwide production of laptops was 47 million in 2005, our project anticipates shipping between 50 and 150 million in 2008 at around $100 each.

We gained credibility as soon as Quanta agreed to make the devices – it makes over a third of the world’s laptops.

It is no longer just a case of Nicholas Negroponte’s OLPC project – it is now a case of executives at major corporations joining forces, including Quanta, AMD, New Corp, Citicorp and Google.

OLPC’s second revolution is to create its own network mesh. Shut down our laptop and the processor goes off, but not the routing function.

Provided a user had 15 per cent or more of the battery, they will be routing other individuals’ messages – and that is how the network works.

The telecommunications industry is still trying to control that last mile, but we are not prepared to wait.

There are many other crucial innovations, such as the one concerning electric power. We use small, human-powered generators, designed for even an undernourished six-year-old to use – giving at least 10 minutes of computing for one minute of effort.

Most laptops burn 30 to 40 watts. Our system averages two watts, with a dual-mode display that is reflective for bright conditions and backlit when needed.

And it is no good giving a six-year-old a delicate instrument that needs servicing. Time and money spent on good design – paring down the frills and focusing on the structure – saves a fortune in mass production and delivers a more rugged product.

It is the same with the software: what is the value in teaching Microsoft Excel to a 10-year-old in the Himalayas?

They should be making music, building things, writing their own programs – they should not be office workers.

To this end, we have embedded a lot of tools, including music, drawing and children’s programming languages.

We are also completing Wiki textbooks in six languages and our e-book reader is itself a Wiki.

There is no digital rights management (DRM) on the device, so anybody that publishes anything has to put it in the public domain. Such a DRM stance is extreme – but we are committed.

I am constantly told that you cannot hand out free computers to children because they will all be stolen or sold. So we make sure there is no market for our laptops – none will be sold commercially.

Only children and teachers can get hold of the devices. Stealing one before delivery will render it useless.

The laptops are delivered to each child personally and installed so that, if the machine is stolen, it soon disables and is again useless.

Most people still say that if you have money to invest, it should be put into the hands of the teachers, not children.

But consider the reality in the developing world: most rural schools run two morning and afternoon shifts, so a typical child’s in-class tuition lasts 2.5 hours a day.

Build a school computer lab and each child might have 20 minutes per week of use. Provide a laptop, and the pupil can play and learn all day.

You seldom hear me use the word teaching. We all learned to walk and talk by direct interaction with the world.

At about six years we are told to stop learning in a specific way and to undertake all further learning by being taught.

Years ago, I worked with Seymour Papert on a language called Logo and he noted that children writing programs about drawing a circle have to understand “circleness” a lot more deeply than if they just read about it or have it described on a blackboard.

And debugging a program involves operations that are the closest you can get to understanding the process of thinking.

We have overturned a lot of conventional wisdom so far, but there is one more assumption I would like to tackle: OLPC is not, I repeat not, a laptop project.

It is an education project. Even more basically, it is about eliminating poverty.

The belief is very simple: no matter what solutions you have for big problems, they all involve education. In some cases you need more, but in no case is it ever without education.

We are proud of the laptop. It is cheaper but better than almost any available.

If you come up with any other device that could offer as much educational opportunity to as many people as this does, we will take it.

Nicholas Negroponte is chairman, emeritus at MIT Media Laboratory and chairman of One Laptop per Child. This article is based on a speech at a NetEvents summit. To watch a video of the presentation visit: www.netevents.tv

Visit ComputingTV for a video on the OLPC project: www.computing.co.uk/tv

The OLPC laptop

*You can use the device like a conventional laptop, or a flat tablet PC. It has retractable
antennae ears that provide a wireless link.

*After turning on the laptop, users see themselves straightaway among other groups of children. Individuals are then presented with a choice of groups to interact with, like in a school playground.

*The user interface acts like a buddy list ­ it shows who is online, what they are doing and how an individual can participate. A user’s friends are allocated various colours and onscreen items identify which programs are running and how much memory the device has left. The software runs on the Linux operating system.

*Users form their own network mesh. The mesh is either connected via cell, WiMax or satellite, but it is usually a school that has a server. OLPC provides the server and it costs $100 with 330GB of storage.