« Wattson, Come Here, I Need You | Main | Grameen Covers the Monopoly Board »

On the Horizon (02/17/06)

olpcsimp.jpgWhat's the best way to bring digital tools to young people in the developing world? One Laptop Per Child? One Cellphone Per Child? One Simputer Per Child? The race to bridge the digital divide is heating up.

The $100 Laptop in Progress: The last month has seen two big developments in the One Laptop Per Child project, also known as the $100 Laptop project (see previous discussions of OLPC here and here, along with Ethan's excellent overview). The first is that the OLPC project has officially teamed up with the United Nations Development Programme to deliver the low-cost computing device to the poorer parts of the world.

OLPC will first implement the program in seven diverse and very large countries. In each of those cases, the government will buy the machines to be given cost-free to students in well specified but large pilot projects. In the case of LDCs and poor countries, the UNDP will work closely with OLPC and other UN agencies on the ground to assist national governments to deploy the laptops to targeted public schools with a variety of internal and external funding sources.

The involvement of the UNDP gives a stamp of approval to the project from the mainstream global aid community, which guarantees a large audience for the device but risks tying up the process in abundant bureaucracy.

On the technical end of the project, Red Hat released the software development kit, or SDK, for the OLPC device a few days ago. The kit includes development tools and an emulator, and requires that you're running on Linux (preferably the Red Hat version). There's a Flash demo of what the emulator looks like, which may be of interest even to people who aren't ready to start coding.

The Microsoft Mobile Phone: Readers may recall that when the $100 Laptop idea was first proposed, my initial take was to suggest that a more suitable networked information device for the developing world might be built up from the mobile phone, instead of stripped down from a traditional computer. Given the near-ubiquity of mobile phones -- and given the growing attention being paid to the idea from groups like the global GSM organization -- I still think that they would provide a better fit, but I was a bit surprised to learn that the major tech player picking up that theme was none other than Microsoft.

Bill Gates, Microsoft's co-founder and chairman, demonstrated a mockup of his proposed cellular PC at the Consumer Electronics Show in Las Vegas earlier this month, and he mentioned it as a cheaper alternative to traditional PC's and laptops during a public discussion here at the annual meeting of the World Economic Forum.
Craig J. Mundie, Microsoft's vice president and chief technology officer, said in an interview here that the company was still developing the idea, but that both he and Mr. Gates believed that cellphones were a better way than laptops to bring computing to the masses in developing nations. "Everyone is going to have a cellphone," Mr. Mundie said, noting that in places where TV's are already common, turning a phone into a computer could simply require adding a cheap adaptor and keyboard.

I don't often find myself in agreement with Bill Gates on technology issues... and I still don't. Although the Microsoft proposal uses a phone as the core of the system, it tries to turn it into a PC, with a big screen and keyboard, rather than focusing on making the mobile unit itself a more useful information tool. Moreover, there's the closed-source/open-source difference: the One Laptop Per Child project is fully dedicated to making the code for the system entirely free, so that users can become developers (a theme we've been pushing from early on: Redistributing the Future). Although Microsoft made some noises about considering an open source version of the Windows Mobile software, they don't have a good track record on the subject.

The Simputer, Revisited: The dark horse in this race is actually one of the first ones to hit the track -- the Simputer, an Indian-designed handheld computing device. Initially proposed in 1998, they finally hit the market in 2002, but have failed to make a real splash. There are numerous explanations for why that might be, from too-high a price (US$240 to $480 at launch) to internal politics of various international organizations. But the Simputer is still around, and still available. WorldChanging alum Taran Rampersad got his hands on one last year, and has a generally positive take on the device.

What would it take for the Simputer to be a more viable alternative? Aside from a lower cost -- which, frankly, is as much an issue of scale of production as cost of components -- the one feature that I'd sorely miss if I had one is built-in wireless connectivity. WiFi, cell phone network, I don't care. I just would want to be able to use it to get online without being tied to a phone or ethernet cable, or connecting to a cellphone. That's the feature that both the OLPC and the Microsoft phone would have that the Simputer doesn't.

Leapfrog IT: An Incomplete Overview: Finally, C|Net provides a useful, if incomplete, listing of the various projects underway to bring inexpensive computing hardware to the developing world. The Simputer is mentioned in passing (described as having "flopped"), as is Brazil's effort to distribute a cheap Linux PC. The piece provides some details about the $100 Laptop and the Microsoft cell phone, and adds in some devices that range from "unlikely to be used" to "highly unlikely to be used," such as so-called "thin clients" that combine the mobility of a desktop PC with the stand-alone capabilities of a monitor (as in, essentially none of either of them). But what's missing is in many ways more interesting than what's included.

Nowhere to be seen, for example, is the Asiatotal "iT," which offers a free computer supported by ubiquitous advertising. Neither is the Mobilis, a low-cost tablet computer built for the Indian market. The EELS, a hand-held wireless computer/electronic textbook, is now being tested in Kenya; that it is not currently planned for the broader market is likely a big reason why C|Net doesn't mention it. More surprising is the absence of the Nokia 770 tablet, which costs about the same as a Simputer and is equally as Linux-hackable, but provides WiFi and bluetooth.

Still, it's a good reminder that there's a lot of action in the "low-cost computer for the developing world" arena -- and the OLPC is by no means assured of a dominant spot.

Comments (11)

The key here is to produce value instead of lowering the cost. That's the real challenge. Making something cheap is easy. Building something that can pass Koffi Annan's cursory inspection without a handle breaking off leans toward value.

And the mobile phone is leading the way. With a Linux GSM phone added to a Simputer architecture, a lot can be done. Why Simputer? Because it's open hardware, and can be modified without patents, copyrights and so on.

Build something that can pass through 3 evolutions of Moore's Law while remaining up to date with what is available after that time passes is the key.

Another failure is the lack of content or curriculum to be used in education. Lots of holes in all of these stabs in the dark, mainly because not one of them seems to learn from the other. A colleague in South America told me that someone from MIT Multimedia lab said that they couldn't get a Simputer to use as a baseline for the $100 million laptop.

So, I got one. Why couldn't MIT? I suppose that they didn't try.

The Simputer is being used in all manner of things right now in India - but you don't hear much about it. It's in production, it's relatively cheap and it broke new ground which it needs to continue doing if it is going to survive. But mainstream media - including mainstream blogs (which have become as bad as mainstream media in many cases) don't cover things in depth. The silence of successful technologies is an attempt at a 'Kiss of Death'; it should not be so. All of them could teach each other a few things, and I include the Simputer folks in on this. I have one, I like it, but it's not perfect.

There are several really serious issues with these hardwares IMHO.

First they are still environmentally unfriendly. They will still breakdown and need to be disposed of & chances are that in rural villages they will just be lying around leaching toxic chemicals. So how is that helping? For more info on this kind of problem, ask the Silicon Valley Toxics coalition.

The second issue is that some of these hardwares are being designed in such a way that decisions are being made relating to how children/schools/communities will use them without ever asking the end user what they need these technologies to do or how they would design them themselves. Imposing technology models on communities makes them less participatory and less user-centred. How can one computer be relevant to the learning needs of all children in all rural villages from China to the Congo to Nicaragua? Impossible. Try to think of it this way, if your computer had been designed with you in mind, you would not be constantly shaking a fist at your screen and threatening to throw it out the window, would you?

CleverGirl makes another valid point that has been kicked, beaten, stabbed and shot (and which I didn't touch): Environment. A lot of old and donated machines end up in the landfills of developing nations where there are no laws, or no enforced laws, on what happens when these machines die.

In the developed nations, it's economically viable for companies to donate old systems through tax breaks, which means that in an odd way, the governments of developed nations are providing an incentive to export harmful waste with a half-life of Moore's Law. Considering hardware manufacture has resulted in things that die sooner, well... The U.S. may have 'Silicon Valley', but the developing world is quickly creating 'Silicon Landfill'.

BUT... part of becoming a developed nation means involving technology.

Personally, I'd like to see more work toward biocomputers - but that opens up the dilemmas of patenting life, as well as all sorts of ethical dilemmas that I'm fairly certain by the state of the planet that humanity simply isn't ready yet.

I agree with you but I think lower costs with the same product value is what the consumers are looking for.


This project suffers from a fundamental flaw, in that it is aimed only at children. I think it really should be a laptop per a person. All people should be encouraged to learn and improve themselves, it should not be a critera based on age. Also, there would be great benefit from vast numbers using and developing on the platform, not to mention a lower of cost through a larger market.

MIT's "children only" approach appears to be designed to create a segregated class of consumers, who are getting a toy that will not be used when they get older.

It would be much better to make this an information/communication machine that could be used by all.

About a month ago, I went to a meeting at the MIT Media Lab for their one laptop per child project. They had several different dummies on the table and one supposedly working prototype in Taiwan with the hardware team talking to laptop manufacturers.

These things have a nice form factor, like a paperback book. The computer is in the screen, keyboards and other peripherals can be plugged in and out. The power supply appears to be four C batteries with a built-in handcrank. Each machine is a node in a mesh network in its own locality. This thing is gonna get built.

Media Lab has a wiki on the OLPC Project

I don't doubt that gmoke is right -- "this thing is going to get built". The real question (one without an answer) is : Yes, but will it be used? And how will it be used? We must all wait and see.

But anon has an important point about the MIT target -- children only. Technology adoption among children is fast & efficacy high. Anon's point was proven a few years ago in a paper by Mark Warshauer on First Monday. He wrote about several tech projects including India's Hole in the Wall. He points out in that paper that some parents were less than pleased that there children were neglecting their chores in order to play with the Hole in the Wall screen and that they did not see the value of giving their children access to it.

Children are part of a larger community of adults and no matter how commited one might be to child rights, the only way to successfully get something done with children is to involve the adults in their lives. If that is done, I believe there will be less adult resistence and children may then be better enabled to participate.

Thanks to Taran for thoughts on the environmental argument! Beaten around, yes... but cannot be forgotten. Including the issues relating to the metals and chemicals used in building a computer. NYTimes did a huge article a few years ago on the toxic slave-labour conditions through which coltan, a metal essential to laptops, cellphones, PDAs is being mined in Africa.
http://en.wikipedia.org/wiki/Coltan It is disappointing that these problems and debates are not being raised in the "digital divide" sector.

Jonathan Cooper:

In my opinion the OLPC will end up being the next version of the GI bill. The GI bill allowed a segment of the population who had the capability but not the resources to attend college.
At the time it was first introduced there were a number of concerns from colleges to the effect of "these idiots don't belong here". Also, there were no provisions (that I know of) for jobs for the new graduates.
None the less the effect of having a massive number of college educated workers (a large fraction of whom were in science and engineering) transformed the US and the world.
I think the same thing will happen with the OLPC.
Because of the way the laptop is set up, the more people who have one the better and more complete the network will be. That is the key. If it were just a laptop with no network connection it would not be of any great use.
There are currently some resources for education out there and more all the time.
Also note that what was once a system flaw, how to make sure that only children get them, becomes a feature. If some adults steal the laptops and use them for their own purposes what are they going to do with them? learn?

Because of the open nature of the software and the expandability of the hardware I see these laptops as "bricks". They are useful on their own but with open source tools and a bit of cheap hardware a few of these things could be combined to make a server for the community.

I also think this is why Microsoft doesn't like it and want there own version. The last thing they want is a large number of bright people using computers who have no intention of using MS software.

I in no way think this will solve all the problems but I think the OLPC will go a very, very long way (much more than is seen in the "developed" world) in allowing the people who get these resources to solve their own problems much better and faster than we can.

Of course I could be wrong.

Rich McComas:

I feel like much of the idea of this project is missed in these discussions. Some of the thought that went into the OLPC project, and in many ways this is the strongest part of the portfolio, has not seen the public light of day. The $100 laptop is more of a project in using distributed systems to reform heirarchal education patterns through emergent learning than a heartstrings tugging attempt to clothe the children of the world in software. It's a bottom up approach to education that has far reaching implifications for the targeted countries. The central design document of the $100 laptop concept in its larval state can be found here:


To summarise, most solutions to education are designed at a high level, tested in relatively small and fantasticly controled pilot studies, and then enforced from upon high. Due to the distance between the hierarchy that decides what proccess to apply and the majority of lower level groups charged with implementing the chosen directives, there is rarely a good fit. And because any real changes must involve the entirety of the heirarchy, change is slow to come. By providing the tools of learning, and in the western sense of the concept this means technology, and by deliberately not giving out any top-imposed direction, emergent systems custom tailored to the region of choice are a natural consequence. The document itself is an entertaining read and I highly reccomend it. Whether this model will work or not on the fantasticly large scale that the OLPC project has engaged remains to be seen, but the research done on the subject seems to support it.

As far as the issue of content is concerned, in the past few months I have been inundated with information about open source education efforts from the kindergarten level up through graduate school. The most noteable efforts include the dramatic expansion of both the Stanford on iTunes and MIT OpenCourseWare projects, as well as the continual growth of the Wikibooks archives. If the example of the Wikipedeia's dramatic growth from seed concept to titan is any measure, a somewhat complete free textbook repository centered around Wikibooks could be a reality by the time of the first major $100 laptop rollout in 2007.


While OLPC sounds like a panacea to bridging the digital divide, I'd rather see all this money being spent on say training teachers to teach math and science in such a way as to kindle a child's curiosity forever. Or on exposing underprivileged children to neat problems such as space travel and astronomy. Face it, for most kids in third world countries, education & critical thinking skills are far more important tools for life. And computers don't play a critical role for that. They might hamper more than help here. While computers are great tools to use, they almost always make thinking lesser easier. e.g. I'm more inclined to do simple math such as division or multiplication with the "Calculator" in a computer than using my mind. A long time ago, I used approximate math within my head to get ballpark answers for most calculations. For greater detail I remembered a few key logarithms (e.g. for 2,3,5,7) and used these along with anti-log tables to figure out exact answers. It saddens me that with computers and calculators, kids nowadays might never learn to exert their minds as much. Perhaps more worrisome is the thought that exposing kids to computers at such a young age, might predispose them to eye problems (No computer screen is ever as safe as reading from paper). Computers are only getting easier to use. Does it make a difference if a person learns to use a computer at age 5 versus at age 16 (perhaps after she/he learns some fundamental math and science to help her/him in a future career)? And perhaps after she/he learns the sheer joy of stumbling upon an unplanned topic while thumbing through an encyclopedia.

MIT is thinking hard about the teaching and learning that will go along with the $100 laptop. David Cavallo has been working on various pilot projects in Brazil and other countries for years now, including a project in sustainability poster-city Curitiba.

Another thing about the laptop is that everything is hackable, both the hardware and the software. That is one reason why MIT refused Steve Jobs' offer of the Mac OS for Linux (and probably why both Jobs and Gates are now promoting cell phones as the answer to the digital divide in the "undeveloped world.")

When everything is hackable some interesting things will happen. Some $100 laptops will probably become the guts of IEDs but others will become useful devices we probably can't even imagine today and the software will be the same - new viruses but also new applications, especially new language applications for previously underserved languages.


This page contains a single entry from the blog posted on February 17, 2006 1:59 PM.

The previous post in this blog was Wattson, Come Here, I Need You.

The next post in this blog is Grameen Covers the Monopoly Board.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34