OLPC vs. Virtual computing

A post with the following title has recently landed on my RSS reader: “California firm undercuts $100 laptop scheme“. Apparently it was not a single instance of similar reporting.  The “Silicon Alley Insider” titled their article “Take that OLPC!“, “Forbes”, with a title “One Virtual PC Per Child” was more gentle, and there were others.  In most instances however, there was a subtle finger-pointing at OLPC as a failure.

The story is of NComputing, which produces terminals that distribute resources of a single PC across numerous terminals, which from the user’s side supposedly operate as a stand alone computers.  The company just signed a deal with the Indian government to sell 50K of its units, which are supposed to be used in 5K schools reaching 1.8 million children.  The cost of a “seat” at the terminal is US $70, which makes it rather attractive ($30 cheaper than yet to be achieved goal of OLPC).  This is probably why just about a year ago NComputing sold 180K of its devices to Macedonia.  And apparently the rivalry, particularly rhetorical one, between NComputing and OLPC is not new.  However, i think the critics are missing a point here.

Although looking at the numbers, it looks like NComputing’s solution is more attractive.  In fact most of the discussion surrounding OLPC has been around the costs and distribution (here is a really interesting and thorough post on the subject).  I tend to agree with a lot of criticism of the project, but not with the idea that is an obsolete idea.  I wrote before about the contribution of OLPC to developing the market for ultra small and cheap laptops aimed at education.  Yet it took me to listen to Matt Keller, Director for Europe, Middle East & Africa for OLPC, at the last ITU Telecom event, to realize the substantive philosophical argument this project is struggling to make.

In the articles I read about the NComputing story, OLPCs response to the deal was presented as an excuse.  Negroponte claimed that the two projects are incomparable as they are addressing different issues, but I am not sure that his point went through.  Indeed, the idea behind OLPC (and I sincerely wish they would do a better job communicating it), is for the child to make laptop an integral part of his or her learning.  Learning in this class does not include just sitting in front of a computer in class and following the teacher’s instructions, but to be able utilizing this resource anywhere this child goes.

For example, one of the particularly interesting and under-studied aspects of OLPC in my view is what happens when the children bring those devices home.  My intuition tells me that there are important repercussions on the ways knowledge and information are generated and distributed in OLPC families.  Although it is an empirical question, I think it is an important component in trying to compare the true value of OLPC vs. a centralized and non-mobile solution such as NComputing.

Using the so-much “liked” terminology I think on a conceptual level NComputing solution vs. cheap laptop solution (OLPC and potentially others) is the difference between 1.0 and 2.0 education.  In first version we indeed familiarize people with technology and socialize them into particular ways of using it focusing on how to perform tasks.  Having limited and segmented interaction with technology leaves little room for innovation and creativity, since the entire activity is taking place in the formal setting.  At the same time, owning the technology (which also has been found in research as correlated with “digital literacy”) allow more natural interaction with it, thus leaving room and opportunity to go beyond the institutional boundaries of the fixed model.

Abdul-Muyeed Chowdhury, the director of an organization working to build subsidized cyber cafes across Bangladesh,is cited in Forbes article as saying “If we could afford to buy one computer per child, we wouldn’t be a poor country.  In a country where people make $1 or $2 a day, it doesn’t make economic sense for everyone to have their own computer. It makes sense to share them.”  Indeed, the sharing model worked with the mobile phones (see GraminPhone as an example), but I think the remaining question is what’s the purpose of deploying that technology.  However, I think we should remember that the purpose of simply providing access (even providing access with a focus on creating business opportunities) is different from the purpose of educating.  It seems to me that the government officials who are making decisions as to whether spend $70 or $100 per education of a child in their country (or more precisely $200-$300 or  $400), should consider the long-term value and the indirect effects of these $30 (or even $100).

What do you think?