Bill Gates on the excellent OLPC (One Laptop Per Child) project, which aims to provide simple, reliable, low-cost computers to developing countries:
"If you are going to go have people share the computer, get a broadband connection and have somebody there who can help support the user, geez, get a decent computer where you can actually read the text and you're not sitting there cranking the thing while you're trying to type."
This has been described elsewhere as 'trolling', but I'd be more charitable. Bill is being sincere...he just doesn't get it. Bill lives in a world where computers exist in order to generate metric tons of corporate documentation, and to use MySpace, MSN Messenger, and iTunes (well, not iTunes, obviously...let's say Napster). I suspect Bill's version of the OLPC project would look like this:
The system would be a full Intel powered laptop which cost the user $100 to buy. This would be because the country's government would have been massively pressured into subsidising half of the rest of the cost, with Microsoft subsidising the rest in order to get a toehold in the market. Contrast to the real OLPC project, where the system is genuinely designed to actually realistically cost $100 or so (at first a little more, economies of scale will bring the price down), and where the expectation is that the government will buy the systems and the people who use them will pay nothing.
Bill's system would run on conventional batteries. Half of the systems would be used for two hours then be relegated to doorstop duty for the rest of eternity as the users would have no way to recharge the batteries.
Bill's system would be designed, as modern conventional PCs and operating systems usually are, on the assumption that a fast, always-on Internet connection would always be available. Internet Explorer would be right on the desktop. The system wouldn't be heavily tested because updates could always be shipped over Windows Update. Bill would probably enter into some partnership with selected governments to build out a massive, expensive communications infrastructure in order to hook up all these expensive laptops, which would predictably end in disaster.
The real OLPC is designed to be usable in absolutely any situation, particularly where no electrical power is available, as will be the case for many of its users. It does not presume the existence of any network connection at all and will be designed with the idea in mind that it will principally be networked together with other OLPC machines in a peer-to-peer style network.
It's particularly sad to see that Bill is so far gone he can't conceive of a technological product people can use without someone being instantly available to tell them how to use it. This is clearly a logistical nightmare for a project as ambitious as OLPC and would shoot the entire thing out of the sky if it were necessary. Thankfully, the OLPC project has its head thoroughly screwed on, and will explicitly be designed to perform simple, useful tasks easily and reliably without the need for constant technical support.
Overall, I'm immensely glad Bill didn't get his hands anywhere near this project, because the people who did seem to be amazingly practical and it seems they'll turn this project, which at first glance seemed a little ridiculous, into something genuinely useful and exciting. Bill wouldn't.