Bill Buxton's Notes

I have this computer in my collection as a reminder of the delicate relationship between object and purpose, and how no matter how well one does on the former, it will likely have no impact on making a wanting concept any better. I include it in this exhibition as a cautionary tale of how the object may help sell that concept, regardless how ill-conceived. This product is here to provide a reminder to keep a critical mind and questioning discourse, regardless of how seductive the technology, or how well intentioned the concept. The concept may be broken, nevertheless.

From the perspective of hardware and software, one has to be impressed by much of what the One Laptop Per Child (OLPC) project was able to accomplish. In general, the team delivered a computer that could be produced at a remarkably low price – even if about double that which was targeted. Specifically, the display, for example, is innovative, and stands out due to its ability to work both in the bright sun (reflective) as well as in poorly lit spaces (emissive) – something that goes beyond pretty much anything else that is available on today’s slate computers or e-readers. In short, some excellent work went into this machine, and that is all the more impressive, given the nature of the organization.

Ultimately, however, the machine was a means to an end, not the end itself. Rather, the actual mission of the OLPC project was:

… to empower the world's poorest children through education.

As described by in their materials, the computer was intended to play a key role in this:

With access to this type of tool, children are engaged in their own education, and learn, share, and create together. They become connected to each other, to the world and to a brighter future.

Hence, making a suitable computer suitable to that purpose and the conditions where it would be used, at a price point that would enable broad distribution, was a key part of the project.

The Underlying Belief System of the OLPC Project

Since they are key to the thinking behind the OLPC project, I believe if fair to frame my discussion around the following four questions:

  1. Will giving computers to kids in the developing world improve their education?
  2. Will having a thus better educated youth help bring a society out of poverty?
  3. Can that educational improvement be accomplished by giving the computers to the kids, with no special training for teachers?
  4. Should this be attempted on a global scale without any advance field trials or pilot studies?

From the perspective of the OLPC project, the answer to every one of these question is an unequivocal "yes". In fact, as we shall see, any suggestion to the contrary is typically answered by condescension and/or mockery. The answers appear to be viewed as self-evident and not worth even questioning.

Others might call that hubris.

What staggers me is how the project has gotten this far as it has without the basic assumptions being more broadly questioned and such questions being seriously answered. How did seemingly otherwise people commit to the project, through their labour or financial investment, given the apparently naïve and utopian approach that it took? Does the desire to do good cloud judgment that much? Are we that dazzled by a cool technology or big hairy audacious goal? Or by a charismatic personality?

To explain my concern, and what this artifact represents to me, let me just touch on the four assumptions on which the project was founded.

Will giving computers to kids in the developing world improve education?

The literature on this question is, at best, mixed. What is clear is that one cannot make any assumption that such improvements will occur, regardless of whether one is talking about the developing world or suburban USA. For example, in January 2011, The World Bank published the following study:

Can Computers Help Students Learn? From Evidence to Policy, January, 2011, Number 4, The World Bank.
  • A public-private partnership in Colombia, called Computers for Education, was created in 2002 to increase the availability of computers in public schools for use in education.
  • Since starting, the program has installed more than 73,000 computers in over 6,300 public schools in more than 1,000 municipalities. By 2008, over 2 million students and 83,000 teachers had taken part.
  • This document reports on a two-year study to determine the impact of the program on student performance.
  • Students in schools that received the computers and teacher training did not do measurably better on tests than students in the control group. Nor was there a positive effect on other measures of learning.
  • Researchers did not find any difference in test scores when they looked at specific components of math and language studies, such as algebra and geometry, and grammar and paraphrase ability in Spanish.
  • But report also notes that results of such studies are mixed:

    Studies on the relationship between using computers in the classroom and improved test scores in developing countries give mixed results: A review of Israel’s Tomorrow-98 program in the mid-1990s, which put computers in schools across the country, did not find any impact on math and Hebrew language scores . But in India, a study of a computer-assisted learning program showed a significant positive impact on math scores . One thing researchers agree on, more work is needed in this field.

Before moving on, a search of the literature will show that these results are consistent with those that were available in the literature at the time that the project was started. The point that I am making is not that the OLPC project could not be made to work; rather, that it was simply wrong to assume that it would without spending far more time designing the process to bring that about, than designing a computer. Risk is fine, and something that can be mitigated. But diving in, assuming that it would just work is not calculated risk, it is gambling - with other people’s lives, education and money.

By having a thus better educated population, will it help bring that society out of poverty?

I am largely going to punt on this question. The fact is, I would be hard pressed to argue against education.

But let us grant that improving education in the developing world is a good thing. The appropriate question is then: is the approach of the OLPC project a reasonable or responsible way to disburse the limited resources that are available to address the problem?

At the very least, I would suggest that this is a topic worthy of debate. Just assuming that giving computers is the right solution is akin to the, "If you build it they will come" approach seen in the movie, Field of Dreams.

But that was a movie. These are real lives that are at stake here – lives of those who cannot afford to see the movie, much less have precious resources spent on projects that are not well thought through.

Can that improvement be accomplished by just giving the computers to the kids without training teachers?

Remarkably, the OLPC Project’s answer is an explicit, "Yes".

In a TED talk filmed in December 2007, the founder of the OLPC initiative, Nicholas Negroponte states :

"When people tell me, you know, who’s going to teach the teachers to teach the kids, I say to myself, "What planet do you come from?" Okay, there’s not a person in this room [the TED Conference], I don’t care how techy you are, there’s not a person in this room that doesn’t give their laptop or cell phone to a kid to help them debug it. Okay, we all need help, even those of us who are very seasoned."

Of course, the failure of this logic is that it misses that point that those unseasoned kids are part of "us", as in "we all need help". So, where do the kids go for help? To other kids? What if they don’t know? Often they won’t. What then?

No answer is offered. Rather, those who dare raise the serious and legitimate concern of teacher education are mockingly dismissed as coming from another planet!

Well, perhaps they are. But in that case, there should at least be some debate as to who lives on what planet. Is it the people raising the question or the one dismissing the concern, that lives in the real world of responsible thought and action?

Can this all be accomplished without any advance field trials? Should one just immediately commit to international deployment of the program?

As recently as September 2009, Negroponte took part in a panel discussion where he spoke on this matter. He states:

I'd like you to imagine that I told you "I have a technology that is going to change the quality of life." And then I tell you "Really the right thing to do is to set up a pilot project to test my technology. And then the second thing to do is, once the pilot has been running for some period of time, is to go and measure very carefully the benefits of that technology."

And then I am to tell you that what we are going to is very scientifically evaluate this technology, with control groups - giving it to some, giving it to others. And this all is very reasonable until I tell you the technology is electricity. And you say "Wait, you don't have to do that!"

But you don't have to do that with laptops and learning either. And the fact that somebody in the room would say the impact is unclear is to me amazing - unbelievably amazing. There's not a person in this room who hasn't bought a laptop for their child, if they could afford it. And you don't know somebody who hasn't done it, if they can afford it.

So there's only one question on the table and that's, "How to afford it?" That's the only question. There is no other question - it's just the economics. And so, when One Laptop Per Child started, I didn't have the picture quite as clear as that, but we did focus on trying to get the price down. We did focus on those things.

Unfortunately, Negroponte demonstrates his lack of understanding of both the history of electricity and education in this example. His historical mistake is this: yes, it was pretty obvious that electricity could bring many benefits to society. On the other hand, completely sure that he was right, and that DC was the way to go, Edison leapt prematurely into a very expensive mistake.

By analogy, yes, it is pretty obvious that education could bring significant benefits to the developing world. But in order to avoid making the same kind of expensive mistake that Edison did, perhaps one might want to do one’s best to make sure that the OLPC concept was not the DC of education.

A little more research, and a little less confidence in one’s own opinion might have saved those who invested in both Edison’s and the OLPC project’s rush to deployment a great deal.

The underlying question is this: in what way is it responsible for the wealthy western world to advocate an untested and expensive (in every sense) technological solution on the most poor nations in the world?

If history has taught us anything, it has taught us that just because our intentions are good, the same is not necessarily true for our consequent actions.

Later on in his presentation, Negroponte states

… our problems are swimming against very naïve views of education.

With this, I have to agree. It is just whose views on education are naïve, and how can such views emerge from MIT, no less, much less pass with so little critical scrutiny by the public, the press, participants, and funders?

In an interview with Paul Marks, published in the New Scientist in December 2008, we see the how the technocentric aspect of the project plays into the ostensible humancentric purpose of the project. Negroponte’s retort regarding some of the initial skepticism that the project provoked was this:

"When we first said we could build a laptop for $100 it was viewed as unrealistic and so 'anti-market' and so 'anti' the current laptops which at the time were around $1000 each," Negroponte said.

"It was viewed as pure bravado - but look what happened: the netbook market has developed in our wake." The project's demands for cheaper components such as keyboards, and processors nudged the industry into finding ways to cut costs, he says. "What started off as a revolution became a culture."

Surprise, yes, computers get smaller, faster and cheaper over the course of time, and yes, one can even grant that the OLPC project may have accelerated that inevitable move.

But in the context of the OLPC project, one might well ask, "So what?" Yes it is obviously necessary to have computers in the first place, before one can introduce them into the classroom, home, and donate them to children in the developing world.

But is not success in this regard just meeting a prerequisite? The meaningful metric of isn’t the raison d’être of the initiative, and only metric of success which one can claim vindication from the skeptics, the impact that those computers have on education?

Rather, the absence of such an educational basis, and the reliance on the technocentric one, speaks loudly in terms of what is at the front of the mind.

But perhaps the extent of the nature and extent of the utopian dream is captured in the last part of the interview:

Negroponte believes that empowering children and their parents with the educational resources offered by computers and the internet will lead to informed decisions that improve democracy.

Indeed, it has led to some gentle ribbing between himself and his brother: John Negroponte - currently deputy secretary of state in the outgoing Bush administration and the first ever director of national intelligence at the National Security Agency.

"I often joke with John that he can bring democracy his way - and I'll bring it mine," he says.

So, apparently providing inexpensive laptops to children in the developing world is not only going to raise educational standards, eradicate poverty, it is also going to bring democracy! All that, with no mention of the numerous poor non-democratic countries that have literacy levels equal to or higher than the USA (Cuba might be one reasonable example). The words naïve technological-utopianism come to mind, which lead me to the last article that I want to mention.

I began by admitting that I was conflicted in terms of this project. Yes, as both a designer and computer scientist, there is a lot about what the project accomplished impressive from the purely technological perspective. But what I find worrying is the apparent inability to distinguish between the technology and the non-technological.

This is shared by a paper published in late 2010 in the Journal of International Affairs, by Warschauer & Ames . What they write is this:

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.

As I said at the start, I have this computer in my collection as a constant reminder of the delicate relationship between technology and society, culture, ethics and values. The fact that this project got the support that it did, and has gone as far as it has, given the way it was approached, says that such reminder is necessary. And if anyone ever wonders why I am so vocal about the level of public discourse around technology, one need look no further than the OLPC project.

Bill Buxton
April 2011

3To be fair, questions and criticism were publicly raised. For example, see Bruce Nussbaum’s essay, Is Humanitarian Design the New Imperialism? In Fast Company: But from my perspective, such commentary was too rare, and reasonable, as opposed to dismissive, responses to it were even more rare.
5Angrist, J. and Lavy, V. (2002) New Evidence on Classroom Computers and Pupil Learning. Economic Journal, 112, pp. 735-765.
6Linden, L., Banerjee, A. and Duflo, E. (2003), Computer-Assisted Learning: Evidence from a Randomized Experiment, Poverty Action Lab Paper No. 5, October.
9Jason Dedrick, Kenneth L. Kraemer, and Prakul Sharma (2009). One Laptop Per Child: Vision vs. Reality, Communications of the ACM, 52(6), 66–73.
11Warschauer, Mark & Ames, Morgan (2010). Can One Laptop per Child Save the World’s Poor? Journal of International Affairs, 64(1), 33-51.

Device Details

Company: One Laptop Per Child (OLPC) | Year: 2007 | Original Price (USD): $199