« Bruce on Bag | Main | Biology News »

The $100 Dilemma

osbourne.jpgAt last week's World Economic Forum in Davos, MIT technology celebrity Nicholas Negroponte announced a project to build $100 laptops for the developing world, for use as textbook replacements, information/communication tools, and the like. The idea is fairly ambitious: the proposed laptop would have a 14-inch color display, and will run Linux on AMD chips.

[Negroponte] described the device as a stripped down laptop, which would run a Linux-based operating system, "We have to get the display down to below $20, to do this we need to rear project the image rather than using an ordinary flat panel. [...] "The second trick is to get rid of the fat , if you can skinny it down you can gain speed and the ability to use smaller processors and slower memory."

The device will probably be exported as a kit of parts to be assembled locally to keep costs down.

Interestingly, the reports don't mention what Negroponte plans as the network interface (modem, ethernet, or wireless), or whether such a connection would even be included in the basic design. Connectivity of some sort seems an obvious need, but the drive to cut costs might end up making that a relatively expensive option.

A perhaps greater issue is that Negroponte seems trapped in an American-user mindset of how people will want to access information and communication resources. While laptops are wonderful tools (I'm writing this on one right now), the computer device with the largest global penetration isn't usually referred to as a computer at all -- it's the mobile phone. The number of mobile phone users worldwide is roughly 1.5 billion, and is expected to top 2 billion by 2006; current Internet use, conversely, is just under a billion, and is projected to hit 1.2 billion in 2006. Manufacturing processes for mobile phones are pretty efficient these days, and the operating systems and applications they come with are simple but relatively powerful. How would the "$100 Computer" look if it was built as a step up from the mobile phone instead of as a stripped-down laptop?

It would look quite a bit like a PDA, and that points us to a pathway for making a $100 computer now, without subsidies, as long as we can get over the idea of our $100 computer looking like a Vaio or a Powerbook. A rugged, lightweight, relatively well-supported sub-$100 computer already exists: PalmOne's Zire 21 retails for $99, meaning manufacturing costs are well below that. I cite it not because I think the Zire is particularly wonderful, but because it's proof that it's already possible to build inexpensive computing devices. The Zire has its limits -- it doesn't have built-in networking or expansion, runs a proprietary OS, and has only a smallish black & white display -- but it's perfectly serviceable for reading documents, and the touch-screen input is arguably more appropriate for non-Latin character languages than is a cheap keyboard. A version with color and expansion slot runs $150. A similar device (running Linux perhaps) with a built-in modem could almost certainly be made for around $100. Negroponte could be much closer to his vision if he broadened his design outlook.

Of course, this isn't the first time people have sought to create a low-cost computer as a means of broadening access; unfortunately, none of them come even close to the $100 price point. We've mentioned the Simputer in the past, and in terms of its specs, it's similar to the Palm devices above. It's a relatively well-known example of a computer designed for non-industrialized world markets. Developed in India, the Simputer looks like a largish PDA, but runs Linux, has USB ports and runs a variety of useful applications. It's now available, but has a few drawbacks: units currently for sale do not have modems or any other built-in connectivity hardware; applications use a non-standard scripting language, reducing the pool of potential developers; and it's surprisingly pricey, given what it lacks -- it starts at around $300 for a black & white version.

The Simputer was recently joined by the "Personal Internet Communicator," or PIC, developed as part of microchip company AMD's "50x15" project, which seeks to bring Internet access to half the world's population by 2015. AMD describes the PIC as "a low-cost, consumer-friendly, managed device that will put technology into the hands of first-time technology users in high-growth markets around the world, such as India, Mexico, Brazil, Russia and China." Although AMD is cagey about the price of the PIC, it's also expected to run about $300. Like Negroponte, AMD seems wedded to the idea of a computer looking like what would be found on the desk of an Average American, and that's where the PICs limits arise: while it's a small, rugged box, the PIC requires an added keyboard, mouse and monitor to function (thereby increasing the cost); it requires AC power, instead of running off of rechargeable batteries; and it only runs a built-in -- and non-upgradable -- version of WindowsCE.

Less well-known is the Alphasmart line of portable devices. Developed for American educational markets, but with obvious potential for global use, the Alphasmart devices look like smallish computer keyboards, but with wide multi-line displays built into the top. The lowest-end units, which retail for about $250 (less for schools) are little more than word processors, but Alphasmart also makes a portable unit which runs the Palm OS and has built-in WiFi, along with SD expansion slots and USB. This model runs over $400 for general buyers, but -- like the Simputer and PIC -- around $300 for schools. The downsides are clear: the screen is made more for word processing than web surfing (okay for email, less good for reading documents); and it runs Palm OS instead of Linux.

Reading various articles about Negroponte's project could easily leave one believing that a $100 computer was impossible, or only possible with heavy subsidies, or only possible in the next decade. Such conclusions are only arguably true if the type of computer one is trying to make has to match preconceived notions of computer design arising from the American (or Western) environment. Once you toss aside the requirement that it look like a Thinkpad, the $100 computer moves from wild fantasy to present-day reality.

Of these different examples, the Simputer comes closest to what I imagine will be a successful design for a $100 world computer. Handheld and running on Linux, its major drawbacks are the lack of built-in connectivity and the price, and I would expect that a bigger manufacturer could likely produce the Simputer for well below current costs. Of these examples, the Simputer is also the one designed with a particular non-Western market in mind. As Negroponte negotiates with the Chinese government to put his $100 laptop in homes around the Middle Kingdom, he should keep that lesson in mind.


Listed below are links to weblogs that reference The $100 Dilemma:

» El portátil de $100: Maná-vaporware from carobotero-co
... y Jamais Cascio de Worldchanging preguntó ¿por qué el instrumento debe ser un portátil? En la discusión que siguió al post de Jamais ... [Read More]

» $100 Laptop from Technology Meme
Nicholas Negroponte's vision for $100 laptop for every child in poor countries has stirred quiet a debate with Bill Gates/Microsoft, Craig Barrett/Intel expressing apprehensions and suggesting alternatives. Others too have chimed in with articulate arg... [Read More]

Comments (21)

Thanks Jamais, glad to see someone is paying attention and talking sense!

I have to wonder if maybe the $100 would be better spent for microcredit for an equal number of people. Then 10 years down the road their $100 laptops would be ready for them. Maybe they could even build them...

These $100 computers are meant for educational systems partly to replace expensive and heavy textbooks so they will be serving a somewhat different purpose than commercial electronics. I don't think that there needs to be a choice made between educational tools for kids and micro lending because different development groups have different focuses. Some micro lenders might even combine donations of $100 computers (maybe after a fraction of the loan has been repaid) with the loans so their clients would get both start-up capital and an organizational/communication tool.

I've been thinking about this a lot lately as well, in regards to education. One thing you have to remember is that this is aimed at education, and to my mind, it only pays off if you can replace paper textbooks, maps, and similar resources with the computer display. Likewise you need to be able to write and edit a lengthy paper on a computer you're distributing to high school students. So I see a small pda display as being a major handicap. I don't think you should go smaller than a Newton, and in fact the old Apple eMate might be the perfect form factor. Also, remember that you can get a laptop today from Wal-Mart for under $500.

Tom, I had considered putting a line or two in the piece arguing that the Newton was the right model for this. I think you're right that the 160x120 screen of the Palm is probably too small, but a 320x240 screen -- as found on the Simputer, as well as the old Newton MP2K -- would work relatively well.

The Palm Tungsten T3 and T5 both have a 320x240 screen. There are numerous other handhelds with this resolution. The T5 has nonvolatile memory which I think is critical to making a truly useful portable tool and that is one reason on demand binding of current texts may still be the most appropriate way to deliver some materials. I have had the chance to use both the T3 and T5 recently and I certainly feel them more than adequate to learn and communicate with. $400.

I've also seen some really nice low power Sony eBook prototypes, they only use power to change the display, it stays static without any energy. A little bit larger format would be an excellent book replacement.

I believe that replacing paper textbooks with electronic will benefit most in the long run, I just had in mind the Rough guide and how those without water might be able to use the $100 even more.

Of course I hope there is more than enough support for both of these ideas as they are very complimentary!

In with this headlong rush to pronounce developing world computing a 'good thing' we need to insert a note of caution. Unless a widely available computer is built with recyclability in mind, we're going to end up with an awful lot of hazardous waste at some point in the future. The EU are introducing stringent new rules to encourage the recycling of such goods but computers could well be foisted on the developing world without this due consideration.

I enjoy and learn from WorldChanging every time I stop by, but I feel that the site sometimes has an overly joyous tone about the top-down tech-heavy approach to democratization and globalization. It's fine for people living in mud huts without running water or toilets to have access to $100 laptops (hell, I want one too), but will it really accomplish anything? Will those computers help rebuild homes smashed by the tsunami? Will they help feed a family? Will they even help educate kids in countries where the education system is corrupt and haphazard? I spent the better part of two years living in shantytowns across the developing world. Many of my friends in the mud hut neighborhoods of Nairobi, for instance, earn around 3,000 Kenyan Shillings a month (about $39). They make horrific trade-offs every day and yet still bring amazing dignity and faith to their lives. Our part of the world has become obsessed with gadgets, and so we believe that that others need them too. But we are the ones who will benefit more from handing out cheap computers. In the developing world, needs are much more concrete. I'm not saying that Negroponte is wrong to lead the charge to create computers that cost a C-note. I'm just saying that we need to remember that people in the developing world are quite capable of establishing their own priorities -- and cheap computers may not be high on the list.


In rural Zambia a few years ago, we encountered a man fixing his bike, and stopped to help. He turned out to be the local "delivery man" for an incredibly large area, riding as much as 200km a day to bring food, medicine and parts to a distributed community. Of course, our first thought was, you need a motorcycle - it would be so much easier. But by the end of our discussion, we all concluded a motorcycle would be a bad thing. How would the gas be paid for? Where would the parts come from? What happens to the (even larger) community when the motorcyle breaks? And most importantly, the "delivery man" was adamant that he *liked* riding his bicylce, and the personal nature of quietly riding around the countryside, being able to talk to people as they passed him or vice versa.

This whole discussion is predicated on the notion of the "personal computer." While it's common in the west for everyone to have their own computer, my experiences around the world is that this is not how many cultures think, nor would it necessarily be useful.

Presumably one of the major goals is to connect rural poor to the world at large, but without power or telephone lines, what good is a laptop? Instead, perhaps we should think in terms of some kind of sub-$1000 "community digital device." It would include (of necessity) power generation of some kind, and a transciever, presumably wifi, with which it could be access the internet, and provide telecommunications. It would not need to be affordable to an individual, but rather by a community, who would share its use (a far more likely usage pattern in my experience).

Too often we're eager to ship our culture overseas, without questioning whether it's wanted or even useful, and I think it makes sense to look at some significantly different alternatives to the "$100 PC." I'm all for cheap PC's mind you, but I do think it's worth questioning whether this is in fact the best thing we can do for the billions who don't live like us.


neil verplank


After reading Robert's post I want to know if someone could please describe (summarize, really briefly) the benefits they would expect these computers would bring to the people who receive/buy them.


1. Micronutrients instead of microchips
2. Malaria prevention with bednets instead of laptops
3. Small scale water technology instead of weather info from the internet
4. Community managed sanitation instead of notebooks with color screens
5. Low cost drip irrigation technology

Really, it continues to amaze me how even well educated people set the wrong priorities.


Neil, you're spot on. My experience in sub-saharan Africa tells me there's a definite hierarchy in the usefulness of communication devices. Anyone who has spent some time in the south, knows this ranking:

1. radio
2. mobile phones
3. newspapers
4. truckers and women in the market place
5. computers and the internet, maybe when other priorities are met first

A couple of points in response.

The first is that nobody -- not even Negroponte -- is claiming that getting low-cost computers into the hands of people in the developing/non-industrialized world should be either the top or the sole priority. But development is a complex system, and trying to fix problems one at a time (a "first we get them clean water, then we improve nutrition, then fix poverty..." approach) is not going to work. Providing tools for clean water and tools for economic development and tools for education and communication (and.. and.. and...) is a far more effective path to getting everyone up to Millennium Development Goal standards.

(And before anyone says it, this isn't really a case of limited financial resources going to computers instead of malaria nets. I'd be shocked, frankly, to find any organization choosing to dump their malaria net/vaccination/clean water programs to spend their money on getting laptops to the developing world. Technology companies spending money and time to develop and provide these tools will not reduce the resources available for other, more worthy, projects.)

In short, this isn't an either/or scenario, and to cast the argument in such terms oversimplifies both the scope of the problems and the range of solutions.

The second is that the spectrum of what constitutes a "developing nation" is quite broad; not all of them are "people living in mud huts without running water or toilets." Even within the nations furthest away from meeting the Millennium Development Goals, not every person suffers from the same level of deprivation. Don't imagine that everyone in rural India or Brazil or South Africa suffers from the same level of poverty as Sudanese refugee camps or "mud hut neighborhoods of Nairobi." The solutions useful for different parts of the developing world will vary -- malaria nets and sanitation may be the clear focus in some areas, while solar power and cheap information tech the need in others.

Robert asks (and I am sorry if I seem to be picking on your comment, Robert, but you phrased the issue in a particularly compelling way): "Will those computers help rebuild homes smashed by the tsunami?"

Homes, as in houses? No, probably not. Homes, as in communities? Possibly, as cheap information tech could be helpful in a variety of small-scale economic activities. As an example, Lorenzo argues for "water technology instead of weather info," and that's fine, but that weather info could be pretty damn useful for people who make their living fishing or other weather-sensitive work.

Sam, the benefits people could receive from having access to cheap information devices are fairly basic, and not too surprising: communication, access to information resources (textbooks, weather, commodity prices, community info, etc.), tools for managing finances (of particular value to people running small businesses, even microloan-enabled businesses), tools to tell one's stories and be creative...

Neil, the community/telecentro model is definitely working in some parts of the world, and may in fact be more appealing than personal devices in some cultures. That said, the mobile phone seemed to follow the same model initially, if I recall correctly -- a center where there was a single or small set of phones for community use, which eventually gave way to family ownership of phones (and then to individual ownership of phones).

Which leads me to Lorenzo's second post. I note that mobile phones are listed as the #2 most useful communication method in the hierarchy, and I see no reason to dispute that. But I ask -- a decade ago, when mobile phones were first trickling into sub-Saharan Africa, would they have immediately been placed at that level? Or would they have been seen as well below word-of-mouth methods, as you now place computers? Is it possible that one reason why information devices and the net aren't as useful is because they aren't yet terribly common? Is it possible that, as they become more common -- and especially if they are an evolution from mobile phones, as I suggest, not the Negroponte laptops -- they will rise in the hierarchy, too?

Great comments, people. Thank you.


Jamais, I think most of us will agree with your view that development is ultra-complex and that action is needed on several spatio-temporal levels.
And you're right to point out that the "priorities" approach is too simplistic. But sometimes one has to stress a hierarchical order, just to remind ourselves of what the realities are on the ground.
Most empirical research in a local setting (say the village level) naturally tends towards a complex hierarchical model. Really, if you go out and do field-work, you will be amazed to find that on the most local of levels, clear priorities exist.
Nobody is saying that the cheap computer fantasy (we've had so many of those) isn't worth exploring further, but there comes a time when you have to actually talk to the people on the ground.
Nobody is saying we don't need creatives (like yourself) constantly exploring new technofixes, but we need realists with field experience as well. There's no contradiction between them. It comes down creating a strategic symbiosis between them, on the basis of their different findings.

To be honest, it's not a difficult exercise. Go to a mudhut village in Congo (you know, they do exists, those villages) and ask the people what they need most urgently if they had to choose between bednets or computers (because they know better than anyone that they have to choose; they know it from experience). You'll get a straight answer. Go to a local town with basic medical facilities but no computers, and they might have a different view.
It all depends on your where you are, and who you're talking to and what your development goals are.

But I would ask you not to give up on low level, locality-based empirical research amongst the poorest.

Go to a mudhut village in Congo (you know, they do exists, those villages) and ask the people what they need most urgently

I know they exist. I know that places with even greater deprivation and stress exist (refugee camps immediately come to mind). In no way was I trying to deny them when arguing that they're not the only manifestation of life in the developing world.

Lorenzo, rest assured that, even though we post about technology-based tools for development (and leapfrogging and the like), we will continue to write about other, more fundamental, development goals and priorities, as well. We've done that from the very beginning, and we will continue to do so.


Jamais, a quick word about providing internet to the bush (to put it bluntly). I can only give you some numbers about Congo.

30 million Congolese live in a rural setting.
-2% of them have electricity (you need it to run a laptop on, and your laptop can't provide it, not even if you consult a course on rural electrification on the internet)
-95% have a sub-existence level diet, based on entirely locally produced and gathered food (tele-food does make sense though - but FAO results on this front have been very disappointing, because the tele-infrastructure gets usurped by local elites)
-the medical services profile is not very encouraging: there is approximately 1 physician for every 100,000 people (India has about 30, Europe about 300) (tele-medicine has been very much a top-down affair and experiences in Africa have been extremely disappointing as well, since the learning curve for even the most basic of health related actions is too high for rural people; something could be said in favor of tele-medicine on a higher level: the local doctor who connects to Western collegues; but forget the village level)
-there's no transport infrastructure (no road, rail or air)(this requires top-down, state initiatives)

The list goes on.

I'm not saying that providing those villages with computers doesn't make sense, but if Negroponte finds someone who wants to deliver them, I hope he makes sure that the person in question is a doctor or a nurse, an low level agricultural scientist, or an electrician.


rest assured that, even though we post about technology-based tools for development (and leapfrogging and the like), we will continue to write about other, more fundamental, development goals and priorities, as well. We've done that from the very beginning, and we will continue to do so.

We all admire Worldchanging for doing just that: providing a very broad spectrum.

Yes, Jamais, different communities do have different needs and desires. A non-profit called Viva Rio has established high speed Internet centers in a bunch of favelas in Rio de Janeiro. These have been very successful, charging R$1 -- or about 40 cents -- for a half hour of browsing time. Similarly, some of my friends from the mud hut neighborhoods of Nairobi have Yahoo! accounts and, when they have the cash, go to the shilling-a-minute Internet centers downtown. The same holds true for Istanbul and Mumbai, the other cities in which I lived.

It's not that poor people have no interest in the Internet. But let's not confuse matters. There may well be a market for $100 computers in all these countries (and here at home, too.) If so, someone will sell them and make a profit. Hooray for capitalism. But let's not pretend that providing computers is a powerful economic or political statement.

You argue that computers can aid "communication, access to information resources (textbooks, weather, commodity prices, community info, etc.), tools for managing finances (of particular value to people running small businesses, even microloan-enabled businesses), tools to tell one's stories and be creative."

Maybe (though I find them horrible for reading and questionable for communicating and even bookkeeping.) Computers are a tool that people with the right skills and education can utilize if they want to. I'd simply argue that, when it comes to changing the world, on-the-ground organizing is much more fundamental than technology.


Robert, again, spot on.
Social organization is the way to go, when it comes to changing the world on the poor's ground.
And mobile phones, radio and radio trottoir are very handy on this front. Laptops may come in handy at a higher socio-geographical level though (when organizing structured "campaigns", social science software and internet info comes in ok).

One more detail about the locations we're dealing with: I have the vague impression that many techno-minded creatives simply forget how many people still live in rural environments (a bit more than half of all people on the planet).
And here, a "rural logic" is needed.

At first sight, it looks like internetconnections might bridge a gap and connect the rural to the global, but that's simply too reductionistic. The rural habitus is too different and based on other priorities and needs.

I can imagine that in a dense urban context like that of the favelas they make sense, but not in a village of 500 people in the matto grosso.


I didn't make this clear, but it seems to me there are two separate but important points to think about. One, as I've said, is the "will this help these people" issue. That is, might there be some other technology that would prove more effective.

But the other, and I think this is implicit in what you said Jamais, is what kind of impact will such a technology have? While the acceptance pattern is indeed likely to be "no one / community / family / individual", there is to me a larger question of whether even setting people on this path is in fact a good idea. We (westerners by and large, but we know who we are) have this basic assumption that "they" in fact need our help.

I know that there are a great many people who live, by my standards, very difficult lives. But all too frequently, the "rural poor" are only aware that they are rural, or poor, in juxtaposition to how we live. And frankly, I'm not so sure about the way we live. I've got healthcare, long life, and a lot of other things to boot, but my car poisons the planet, the people who live next to me are not part of my "community," and in many cases, I find the "rural poor" to be living qualitatively happier and more fullfilled lives than me, despite their relative lack of "things."

I know that may seem romantic, and in fact I know many of them would trade places with me in a heartbeat. But we have the luxury of choice, and I think with that comes responsibility. Every effort we make to help has consequences, and it seems very clear to me that setting the rest of the world on the same path we've taken is not a good strategy, based on the problems we so clearly have created.

And it's very true that of course there is a much greater spectrum than simply rich people and those in mud huts. Still, I wonder, and this was the subject of much debate with the people in Zambia I was with, whether we should be looking to the people on the lowest rung not as in need of aid, but as a source thereof.

I'm in no way trying to be critical of the aid that is provided around the world, and the fact that those of us who have are willing to help and share with those who don't is the only hope for our species. But I do believe we need to think pretty clearly about what indeed consitutes legitimate help. Are we "helping," or merely trying to give "them" what we have already got, when it's quite possible we shouldn't have it either. I think there needs to be a greater dialogue and exchange between what we classify as the "haves" and the "have nots" about where we all want to get to as a species.

I also know and have seen first hand the kind of suffering to which poverty leads. I am *not* advocating that we would all be better off poor and desparate. To sink to aphorisisms to make my point, just because children like sugar, should you necessarily give them candy?

And may I say, utterly off topic, thank you thank you thank you to all of you at worldchanging for making these conversations possible, and being invested in them. Worldchanging, you rock!

Good, thoughtful comments, everyone.

I'd agree that for the billion or so people who live in or close to absolute poverty, a $100 computer is just not (much) of a priority, though I also know that information poverty is a very real component of the fabric of poverty as a whole, and there may be places where $100 laptops are part of the fix.

But please also consider that there is a trap of assuming things will always be the way they are today. They won't. They won't, for the simple reason that they are not sustainable, and that which is not sustainable does not continue. (Unfortunately, this impermanence is just as true of many poor and rural areas as rich and industrialized -- a slashed and burnt rainforest is just as dead as a coral reef killed by climate change)

But they also won't stay the same, I believe, because I think we're going to address a lot of these problems. The global public debate is shifting. The MDGs may not be perfect, but they're a damn big step forward, and there are some reasons to believe we'll reach them. That billion people at the very bottom may simply not be at the same bottom in ten years: it's entirely within our power to make that happen. Cheap, even, compared to the budgets of developed world governments.

And once people have clean drinking water, decent housing, enough medical care and education for their kids, they may well be damn glad someone developed a $100 computer and other cheap high-tech tools.


The main point is not the hardware. If we can provide any kind of hardware as a shared resource (like the kiosk model) then it is kind of solved.

What is needed are services? What do these people do with a computer and connectivity?



This page contains a single entry from the blog posted on February 8, 2005 4:47 PM.

The previous post in this blog was Bruce on Bag.

The next post in this blog is Biology News.

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

Powered by
Movable Type 3.34