Olpc challenges
== elegance for children The concept of elegance is easily taught by example: FRACTRAN is elegant; EULER is elegant. They are concise examples that appeal to adults and to children. Their use is C kernel programming should so delight the mind of the child that they will cut away the dross of overlaid programming and cut to the core of the CPU through a CPU independent language in the small C interpeter. This direct access to the AMD CPU should delight and enhance the child like mind and its desire for elegance. The primes are just that example of number theory that will be the subject for an enhanced mind for years ahead.
(One) Free Open Hardware Laptop (per Child)
OLCP is planning to use only free open source software to give every child the oportunity to investigate its computers software to make the computer be the childs own computer --- I am 100% sure this is a great idea!
As far as I can see, you are NOT planing to make this computer Open Hardware. Of course, you cannot publish AMDs CPU architecture and the interns of all the other properitary components you use. But would it be possible to publish as much as you CAN about the hardware design: which CAD tools did you use, the project files of the CAD tools, ...? If you already did that, I am very sorry - but I wasn't able to find information about it.
It might be a little bit off topic and very idealistic: But wouldn't it be a very great idea to sooner or later transform the project step by step into an Open Hardware project?
Why? -> Maybe then, SOME of the countries would be able to produce the laptops on there own, giving them the ability
1.) to have a little bit more economical growth in their country by producing the laptops (and not buying them from rich countries)
and
2.) to adapt the laptops to their specific needs - just the way they can proceed with the computers software.
There can, of course, still be some properitary components in it, which have to be bought by the companies but which MIGHT be exchanged later by open hardware components (e.g. by some of the open cores...)
Product Lifetime Expectency?
How long does the team anticipate the hardware will remain usable under daily use? The more an item is used, the quicker it will break. Items operated by children, especially ones being encouraged to experiment, are going to cease to adequately function even more quickly.
How much training it being given to educators? Will they be able to adequately implement these items in their curriculum? Will they be able to easily reimage a machine -- possibly through the use of an SD flash card or a system image obtained over the Internet -- if a student accidently takes experimentation a bit far?
How much thought has gone into server structure, so students have a place to collaborate and work together on projects? Teachers will want to record grades and give secure tests through the machines. Students will want to instant message their buddy during a lecture, and look up answers to tests.
Some students will want to tamper with the machines and intentionally break them; what should happen to the student?