The One Laptop per Child (OLPC) program is one of the most ambitious educational reform initiatives the world has ever seen. The program has developed a radically new lowcost laptop computer and aggressively promoted its plans to put the computer in the hands of hundreds of millions of children around the world, including in the most impoverished nations.
Though fewer than 2 million of OLPC's XO computers have been distributed as of this writing, the initiative has caught the attention of world leaders, influenced developments in the global computer industry and sparked controversy and debate about the best way to improve the lot of the world's poor.
With six years having passed since Nicholas Negroponte first unveiled the idea, this paper appraises the program's progress and impact and, in so doing, takes a fresh look at OLPC's assumptions. The paper reviews the theoretical underpinnings of OLPC, analyzes the program's development and summarizes the current state of OLPC deployments around the world.
The analysis reveals that provision of individual laptops is a utopian vision for the children in the poorest countries, whose educational and social futures could be more effectively improved if the same investments were instead made on more sustainable and proven interventions. Middle- and high-income countries may have a stronger rationale for providing individual laptops to children, but will still want to eschew OLPC's technocentric vision.
In summary, OLPC represents the latest in a long line of technologically utopian development schemes that have unsuccessfully attempted to solve complex social problems with overly simplistic solutions.
An excerpt from Can One Laptop Per Child Save the World's Poor? By Mark Warschauer and Morgan Ames, as published in the Journal of International Affairs, Fall/Winter 2010, Vol. 64, No. 1. and copyright The Trustees of Columbia University in the City of New York
Recent Comments
I think that Mark Warschauer's paper as well as the one from Emmanuel Yujuico ( http://eprints.lse.ac.uk/29551/1/Cautions_from_one_laptop_per_child_(LSERO).pdf ) provide a pretty good picture of the complexities of "Education" that OLPC pretty much ignored.
It is hard to tell if OLPC as an organization had the ability to implement the suggested approaches and even more so, if such an implementation would not have stumbled on other problems with similar or even inferior outcomes.
It is always easier to identify exiting problems than find solutions to whatever _future_ problems may appear. There is no "safe bet" in marketing, politics, education and other highly complex and dynamic areas. If it was, everybody would know...
There is no doubt that OLPC did(/does) have an impact. It did not solve the world education problems and is very unlikely it will (sic).
But what I really would like to know is _not_ if OLPC is the "silver bullet" (I have way past the bet-time-stories age :-), is if there is ANY _evidence/data_ that it solved ANY problem in ANY specific setting (eg: "kids with XO laptops in the Peruvian Andes have 35.26% less teeth cavities per year. pValue=0.0038"!).
After 3 years is time for some _data_ to start emerging even if OLPC has to pay for the research...
You would hope that after 3 years we would get some kind of objective metrics from OLPC deployments - the same we'd expect from any other intervention. But don't hold your breath. Remember, Negroponte doesn't believe in testing and has catagoricly refused to even acknowledge the need for objective metrics with XO deployments.
I don't buy that.
I can see that OLPC may not have the money to fund research or that has ethical issues in funding research for its-own "product", but they would sure love some positive data.
NN is a trained scientist and an academic in a prestigious Institution for 40 years. Not only he believes in data but "data" must be encoded in his DNA by now...
I would look elsewhere for the reasons behind the statements.
Now you're going to make me use data to counter your comment. Here's a good data point for you:
"Negroponte: My philosophy is, if you have to measure it, it probably wasn't worth doing"
http://twitter.com/BostonReview/status/10464854703865856
By the way, why do you think Negroponte is a "trained scientist"? He has a Master's degree in architecture from MIT.
Now, could you find the actual site/link that NN states that? Because the BR tweet points to nothing and a BR site search for the comment also yields nothing relevant...
If we are left at that your "data" is as good as any tweet/post/comment/claim etc "out there".
Regarding the academic status of NN... I would just suggest Google
Could you find the actual site/link that NN states that? Because the BR tweet points to nothing and a BR site-search for the actual-statement also yields nothing relevant...
If we are left at that, your "data" is as good as any tweet/post/comment/claim etc, by someone "out there".
Maybe you didn't but I did :-)
Here @4:13
http://www.youtube.com/watch?feature=player_embedded&v=_-pWcn8ug8o
While you're indeed right when you say "It is always easier to identify exiting problems than find solutions to whatever _future_ problems may appear. There is no "safe bet" in marketing, politics, education and other highly complex and dynamic areas. If it was, everybody would know..." the key IMHO is to set up your project in such a way that it can recognize difficulties (via monitoring and evaluationen processes for example) and subsequently adapt to them.
One example for me here is Plan Ceibal which has also run into at times very significant problems yet is seemingly able to at least begin addressing them over time.
Similarly I've also suggested that the rollout model which we see with OLPC deployments such as in Peru and Uruguay (one-year pilot, followed by a massive rollout) isn't necessarily the best way to do things. I think a more sane "staged rollout" model would have a better chance of success (whatever that may be) yet I of course do realize that within existing political realities this isn't very likely to happen.
OLPC and others helped alert the world to the need to provide equitable education for all and has developed technology and hardware/software to implement these good intentions. The bad news is that the ad hoc implementation has not been as successful as hoped. My wife and 6 Western educators just completed a successful week long residential Teacher Capacity Building Workshop for 30 Moshi, Tanzania rural teachers to address pedagogy. ICT4 was not a topic. We had applied to the OLPC Contributor Program, but withdrew the request because it did not align with the goals of the workshop. I had commented at http://1pwn.ixmoe.com/websites/olpcnews.com/sales_talk/countries/thoughts_on_one_laptop_per_chi.html.
From a personal perspective I think that shared computers provides a more collaborative/positive learning experience than one per child. This is probably a minority opinion at OLPC. As long as I am challenging tenets, to learn to swim I am happy that I wasn't thrown in to the deep end as "Constructionists" might have preferred. To me "Education = Teachers" and teachers should be the OLPC marketing focus.
The worldwide Classmate and Acer European School Net (http://www.netbooks.eun.org/web/acer/welcome) implementations seem to have been more coherently integrated with government, community, teacher and volunteer structured approaches to ensure that pedagogy, hardware, educational content and logistics for training, electricity, Internet, ICT4 and maintenance are addressed. OLPC should examine these approaches and use their good ideas.
The OLPC/Marvell XO-3 introduction could provide better hardware/software options (eBooks) and provide a turn around point for OLPC. "One size fits all" is not true for teachers and primary and secondary students.
Major companies like Microsoft and Google are making major efforts to provide better educational content, mature free software, cloud computing and thin client/computer labs. Small NGOs have had success implementing school computer projects with local industry (http://www.inveneo.org/download/Kicheko_SuccessStory_081209.pdf). These approaches could be interesting for OLPC.