I've been surprised by the amazing amount of chatter around a short paragraph from Nicholas Negroponte quoted in The Sugar daddy for future generations on the hardware development process for the XO-2 laptop:
Talk about a blurry image!
"One important thing about the XO-2 is that we're going to do it as an open source hardware programme. The XO-1 was really designed as if we were Apple. The XO-2 will be designed as if we were Google - we'll want people to copy it. We'll make the constituent parts available. We'll try and get it out there using the exact opposite approach that we did with the XO-1 [...] The next one will be different: everything from the dual display to the touch-sensitive, force-feedback, haptic keyboard will be available."
If its true, if One Laptop Per Child will really open up all its hardware intellectual property, especially the dual mode screen, then this is a radical change for them and real game-changer for everyone looking at information and communication technologies. Just listen to Braincast latch on to the concept of Open Source hardware as OLPC's next paradigm shift for ICT4D:
"This is pretty much the biggest news of 2009 - I am going to email the OLPC folks confirm, but according to this article - The new OLPC XO-2 will be an open source hardware project... Just to recap what that means... the source, the schematics, the PCB files, the firmware, the CAD files, everything will be available and commercial use is OK. If it's true, this is extremely exciting, I'd love to see the best company that can make these at the lowest price / highest quality flourish."
I am excited too, but I temper my joy with the nagging feeling that this is another Negropontism - spin and hype that has little relevance to what is or will actually happen. OLPC's history is full of grand pronouncements, from string power to a view source key that have not come to pass.
And I suspect we'll soon find the "Open Source hardware" idea is not as open as we would hope. Just listen to Chuck Kane, OLPC President explain the idea in more detail in OLPC Reboots with a New Focus and Big Plans:
The next version, XO 2.0, is some 18 months out. Rather than doing the whole design in-house, the new plan is to outsource as much of the work as possible. Ideally, "what we should do is build something akin to a reference design that other people can freely use," says Kane. "The idea would be to hand off 2.0 so that partners would manufacture the product and sell the product through their own channels."
So while I hate to dim people's enthusiasm, an XO-2 laptop reference design is not the same as open source XO-2 hardware. Nicholas Negroponte may have been looking for another way to justify local assembly rather than GNU hardware.
I would also love open hardware to happen, but I too share your pesimism.
It makes sense to me that olpc would want to open source the hardware design.
Start from the begining. The idea of olpc is to spread education in the developing world by giving the students laptop computers. But no suitable computers existed, so olpc had to develop one.
Now if the XO-2 hardware design is open sourced, then a number of companies will make them, and olpc can get out of the hardware business and devote all its funds and energy to the rest of the project, such as software and deployment.
I suspect there might be yet another reason for offering the XO-2 as a licenced reference design or even open sourced hardware. It could simply be because OLPC has trouble finding a production company for the XO-2 after the low numbers of XO-1's rolled-out and the desaster of G1G1v2. They need to find new manufacturers. If they don't find one willing to produce all XO-2 alone they try to push local manufacturing of reference designs. If even this is not attractive (too high investments for the expected turn-out) then OLPC might be forced to open-source the hardware and just hope.
The reference design or open source hardware ideas are exactly related o the low XO-1 numbers vs. Negroponte's projections. I think OLPC is well into "hope mode" already.