FAIR has dismissed the OLPC concept of a US$100 laptop (in actual fact the laptop costs more like US$200) as non-economic and the technology as unable to meet the needs of students over the age of 12. The aid organization believes a more economical and practical solution would be to build computer labs in schools, preferably using recycled second hand computers from first world countries.
"In OLPC's agreement with Libya, for example, one OLPC with Internet and support costs US$208 per schoolchild. A normal school with 500-1,000 students must thus invest US$100,000 to US$200,000 to join the OLPC programme. This price represents a normal 10 year budget for a school in the world's 50 least developed countries (LDCs). In addition there are the costs of Internet subscription, training, operation, infrastructure and responsible handling of EE waste. A PC-lab of new Pentium 4 computers in each school would cost a tenth of the OLPC-programme and is today the preferred solution in model countries like Norway, Sweden and the USA," stated FAIR in its original public release.
Dr Negroponte, however, in an email to FAIR's president Knut Foseide, refuted the economics of the above example and in turn claimed that the economics of computer labs and recycled computers do not hold up.
"It is not our policy to do anything but support efforts that provide access to children, including yours even though the economics of recycled computers or computer labs just do not hold or scale. If you or your people had looked more carefully you would have found how wrong your remarks were. For example, the amount you quote in Libya includes everything, including access, and has start-up costs that do not
continue. Even at $200, amortize $188 of that over five years. The long term total cost of ownership and connectivity of an XO laptop is $32 per year per child (still too high). Since we are not in the laptop business, as such, if you or anybody has a better solution, we'll adopt it," Negroponte wrote.
Dr Negroponte also refuted claims by FAIR that OLPC laptop, called the XO, is underpowered and therefore of little use to school children over 12. In his email he also made the puzzling assertion that the developed world does not need children learning IT.
"For reasons OLPC cannot understand, your organization has launched a scathing attack, seemingly in support of a very IT-centric view of
school, the developing world and the needs of children. Not only is Microsoft developing Windows for the XO, it already runs Open Office,
though no child should be doing so. The developing world does not need children learning IT. They need children to learn learning itself," Dr Negroponte wrote.
Dr Negroponte concluded his email by demanding a retraction of FAIR's "story" (presumably from its website) and an apology or face an "unecessary battle" .
"It seems to me that there are two choices at this time, to correct a situation of your own making. One is to remove your story and issue an
apology of some sort. The other is enter into an unnecessary battle, where win or lose, the lose-lose people are those in Eritrea and other
places that will learn sooner rather than later that computer labs are like tennis courts. Yes, you should have them. Yet, we do not advocate
one tennis court per child," Dr Negroponte wrote.
In response, FAIR issued the following public letter:
"In reference to your e-mail from 2007-01-19, we first wish to underline that FAIRs main goal is the bridging of the global digital divide, in accordance with the spoken goals of your organization OLPC. It is not in FAIRs interest to frustrate any efforts working for ICT development, however we feel obligated to express our concerns to the public when projects appear to cause negative effects within our target group.
"Further we wish to emphasize that FAIR does not hold an IT-centric view. FAIR is guided by the needs present in poor countries around the world and address them appropriately with the tools available. The OLPC drive for free content and tailored software solutions are ideas long shared by FAIR.
"If OLPC as you say run OpenOffice now, we congratulate your project on an important step. However, many more user applications are needed and our main question is if all the necessary applications will be able to run simultaneously on the limited capacity of the OLPC.
"Regarding your comment "The developing world does not need children learning IT” we regard this true as we define children as younger than 12 years of age. The age group 12-18, defined as children by the OLPC website, do however need to learn IT skills in poor countries, as they do in rich countries. The age group 6-12 years do not need an expensive tool like the XO to learn learning. That is also why there only exist a very limited market for this age group today, even in the world's richest counties.
"Your calculations of “OLPC's total cost of ownership and connectivity”, are erroneous. Your numbers simply do not add up. As we stated in our press release from 2007-17-01, a regular school of 1000 pupils need to invest more than USD 200.000,- to join the OLPC programme.
"Your statement that “recycling computers does not hold and scale”, is not in any respect valid. Automated recycling of computers yields a unit cost per PC at a fraction of the cost of OLPC and does scale. Further the PC-lab model demands a tenth of the volume compared to the current OLPC-model. Thus labs of recycled computers may be provided to poor countries at approximately 1,5 % of the cost of OLPC, and it is possible to recycle more than 5 million Pentium III/ IV computers each year from the EU and US alone.
"As you request a better solution for poor countries than OLPC's current project, we use the opportunity to recommend three alternatives, which are all based on one PC-lab per school instead of one laptop per child:
"Solution 1: A PC-lab of new Pentium 4 (or equivalent) computers. This costs USD 20.000,- per school (10% of OLPC), and is the preferred solution in rich countries today.
"Solution 2: A PC-lab with the OLPC as clients, with a Thin-Client server and Thin-Client boot media on the OLPC. This costs USD 10.000,- per school (5% of OLPC). This is by far the best choice for the OLPC project today. Not only does it reduce the investment cost by a factor of 20, in fact making the OLPC model viable, but it also reduces the technological risks of the project considerably, as including a thin client in OLPC will enable it to run any needed application via the server from day one. Immediately it stands clear to us that this is such a good alternative for OLPC to pursue, that we strongly recommend you to re-evaluate your model. Our suggestion is simply that you first distribute One pc-Lab Per School in all developing countries, and thereafter look to the possibility of expanding the concept into One Laptop Per Child.
"Solution 3: A PC-lab with recycled Pentium 3 and Pentium 4 computers. This costs USD 1.500,- per school (1,5% of OLPC).
"The costs in all three alternatives include 40 computers per school, network equipment and servers. With respect to recycled equipment, re-installation costs and shipping are included. Also alternative 3) has the added benefit of being the most environmentally friendly, as instead of increasing electronic waste, it reduces electronic waste on a global scale by adding years of use on equipment otherwise being destroyed and replaced.
"Finally we would like to point out that you have not argued our points of criticism given in the press release you refer to in your email. Since our initial criticism was public, and since we have the opinion that a public debate on this issue will be positive for all involved parties, we will continue to inform the public of our opinions. We therefore inform you that this letter and other correspondence on this matter will be made public by FAIR.