OLPC Has A Network Problem


I have been fairly skeptical of the One Laptop Per Child project, not because it’s not a worthy cause, but because it doesn’t factor in the harsh realities of the daily lives of those who Nicholas Negroponte & Co. plan to uplift.

olpc.gifWhere food, water and shelter are largely unmet needs, it is my belief that a laptop is not a road to salvation. Looks like I might be wrong. Apparently kids love it, as per Brazilian Culture Minister Gilberto Gil’s speech at the Emerging Technologies (EmTech) Conference at MIT, where he talked about the magnificence of the laptop project and its deep impact on children.

Nevertheless, his country isn’t ready to order the devices just yet. Why? Because they don’t have the network infrastructure.

We can’t just distribute computers. We have to build a backbone. Just making the technology accessible is not enough. Technology leads to language, to spiritual dimensions. It’s the whole process that matters. It’s not just one item, computers are not enough.

Like putting the cart before the horse. In fact, the lack of networks is a problem that extends beyond Brazil; India and other emerging economies are trying to build network infrastructures as well.

The good news is that OLPC is going to prompt network makers to think creatively about it, and come up with ways to build networks very, very cheaply. John Roese, chief technology office of Nortel (NT), wrote on his blog about OLPC and the concept of hyperconnectivity.

While OLPC is not a Nortel product, it is a tool to stimulate the R&D teams to consider new communication models of hyper connectivity, new programming models and new collaboration methods. It also represents a new type of client, as well as new economic and networking models that are possibly a reflection of the future nature of broadband networking.

Well, let’s hope Roese is right. Leaving my skepticism aside for a moment, I wonder what the impact of a successful OLPC program might have on the network. Thoughts?



I see lots of technologists around here, thats great. I live in Brazil and I can confirm you thats a big layer of the population here is exactly in the situation mentiones by some readers: they have water, they have basic food, they have basic shelters, they have electricity (maybe stolen), they have basic health services. What they dont have is education, and the OLPC is absolutely key in providing that.

However, the constraint for developing telecom infrastructure (backhauls and so on) here in Brasil is not technology, is not lack of investments nor lack of companies eager to invest. Whats missing is competition. Here in Sao Paulo we have only to broadband providers, in many medium – size Brazilian city there is only one provider, in most of the remote area of the countries there is none. And this is consecuence of lack of liberalization / lack of transparent rules of the game from Lula´s governement.

Gilberto Gil is a great artist and a visionary, but doesn´t have a clue on liberal market rules / open competition and so on. It´s not technology, it´s not economy, it´s competition, stupid!

Amish Gandhi

Om, OLPC can still do well without a backhaul network. The key enabler here is the mesh. Following need to work well in the absence of a backhaul:

  1. The mesh setup and operation should be seamless. See http://www.olpcnews.com/hardware/wireless/olpc_mesh_networking.html
  2. Content sharing and messaging across laptops should work well
  3. Content should be distributed well across the laptops eg in a dictionary fashion, round robin.
  4. Even if one laptop has internet connectivity via sattelite or some other means, then based on the promises of the mesh, all other nodes should get connectivity. So if it can be solved for one member of the mesh, then all machines are on the network. Speed and bandwidth would be an obvious issue.

A good analogy is probably any corporate network which has a knowledge base, instant messaging and email. The question is how many nodes do you need in the mesh to make it self sustained and of course how does the info get there.

Alan Pater

“… what about basic educational needs like teachers, books, and a safe place to study?”

It will take 50 years to build up the infrastructure and train the teachers and print the books for 2 billion children. It will also cost a lot more then the $3/month/child for the OLPC program. OLPC enables kids to learn now, rather then waiting for schools to be built.


Om, you start your comments with the tired lament that OLPC isn’t the road to salvation for those who have unmet food, water, shelter needs. But this is a strawman; the OLPC people never said it was. They are not aiming the device at people who don’t have food, water, and/or shelter. They’re aiming it at people who DO have these things, but DON’T have access to better tools for education and world communication.

It’s getting very old to hear this complaint about the OLPC, which has been asked and answered countless times.


Rob O.

I’ve gotta call “BS” on the argument that “…the philosphical issue with OLPC is that children wolrdwide have the right and the need to have a basic education, and technology plays a critical role in today’s educational system.”

I’d challenge you to quantify how the exposure to technology has truly bolstered educational efforts here in the U.S. I don’t see it. In fact, it seems like technology actually be dumbing down American children.

If you want to sidesteps the “let’s help them with basic needs first” issue then fine, ignore the fact that some of these kids don’t have potable water or electricity – but what about basic educational needs like teachers, books, and a safe place to study?


“No potable water” does not mean “no need for technology”. Possibly a 10 year old with a OLPC machine will come up with the solution to her/his community water problem in the next 5-10 years.

To the other point (why spend money on giving away laptops when they need money for potable water), there are other effort in that vein and they need you support too.

Scott Allen

The network doesn’t necessarily need to be attached to the world wide web to be quite useful. Providing one standard server with a terabyte or two of educational content should suffice. I doubt kids in my local public school system have access to much more with all of the websites that are blocked. The kids wont be able to IM globally, be on facebook, download porn, or run 419 scams, but web pages with meaningful content can be set up locally.


Even a low bandwidth wireless mesh could really help with communications in Burma right now. Of course, authoritarian regimes are unlikely to purchase and distribute devices that could be used to form such a communications system, but having even that kind of basic infrastructure in support of education might help innoculate developing countries against the growth of authoritarian regimes (yeah, I’m probably being naive).


I think OLPC is a good step towards bridging the digital divide. It may not be a all-inclusive solution to all poor people around the world but it is certainly a good start to elevate life of “some” people. It will then trigger trickle down education to elevate life of those other people around these first “some” people.

Eduardo Silva

Antoine of MMM, what you suggest seems to be the mesh network of the XO. The laptops will be able to network without outside infrastructure. The only limitations as proximity (in a plain field it has been tested that the maximum range of two laptops communication (one was streaming a music to the other) worked flawlessly to 3 kms. Of course, in a urban, forest, and mountain environments the range will decrease. This is why unless the antennas are 100 meter high, it must be held up with low-cost and low-power wireless mesh repeaters, wireless access points, and future technologies (like Wimax, for example, although the current XO doesn’t handle that protocol in its hardware. Only the typical wireless AP network and this new mesh network, which is a standard still in draft stage)


Giving these kids computers is like teaching me the metric system, here in the US, back in the 1970s: It’s not going to help you when you have to go into the work force.

I spent almost a year living in East Africa performing humanitarian missions. These people need potable water, food, and medications to treat malaria, TB, HIV, etc.

When it comes to charity, look where Bill Gates puts his money.

Bruce Baikie

The issue on the network side is power. These regions have no or very unreliable power grids. This is the number one issue I have been trying to solve with my efforts at Green-WiFi, an affordable approach to solar powered mesh wifi networks for schools in developing regions.

Access to information, i.e. the internet, allows people to solve their own issues with water, food, heath, etc. First they have to have the network to access.

Affordable long distance links are key also. The work the Tier Group at UC-Berkeley on this is important. We have a 44km link from Berkeley (Intel R&D) to Menlo Park (SUN Microsystems) and 12km from Menlo Park to San Jose (Cisco) using software they developed for this.

While this does not solve the problems for every site and region, it is a big start to address the issues. I do believe mobile phone network will have a bigger role to play in the whole equation for this.

George Snell

Hi Om:
I work as a PR volunteer for OLPC. The concerns you raised about food, water and housing are ones the project hears a lot. OLPC, of course, believes basic necessities are crucial to developing countries.

However, you’re making the mistake of many people when you focus on the laptop as a “gadget” The project isn’t about hardware and software, its about education. You might not believe a laptop is the road to salvation, but certainly you must believe education is. OLPC proceeds from the basic assumption that education is the root of solving all problems.

You can find more information about OLPC’s mission here:


Thanks for the interest.

Alex Goldman

If you’re interested in technology in emerging economies, talk to this guy.


At Freedom to Connect in 2005, he noted that although the hardware is relatively expensive in emerging economies, labor can be extremely cheap, making, say, fiber, more economical than it might be elsewhere.

Although a national fiber infrastructure would be inappropriate for Brazil, a coastal fiber line supplemented by a powerful wireless backbone could work.

And the fiber would have to serve the favelas as well as the high rises.

Antoine of MMM/Brighthand

Why not ditch the idea of something backend at all, and allow each mobile device (phone, smartphone, laptop, etc.) be a node that is essentially a connector that creates a network. While I am pretty sure that I am oversimplifying things, what if two computers sitting near each other played like a magnet and attracted one another and then “became a network.” Others in the area could see that network, and with one device playing host, be accepted into it (with the idea of host probably being shared between the first two who connected so that if one leaves, the network doesn’t).

I do agree that just giving the laptops isn’t enough. You have to think of the process of learning, working, educating. Those who are giving these tools need to also understand the ethical lessons that need to be adapted per culture with this. It is opportunity yes, but just giving the cart and no training how to fix a broken wheel or what roads carts are allowed is just as bad as putting the cart before the horse.


what about leveraging gsm and using a s60 mobile? Or similar devices based on linux?


Concepts of universal access to education and medicine will prove (in some ways) to be a catalyst for greater backhaul growth in places like Brazil (was just there last week.) I don’t buy the OLPC vs. food and water dichotomy. We have to remember that the global poor are not one monolithic population, but are divided into smaller and more complex segments. There are global poor who have nothing more than a piece of cardboard to sleep on, for them, I admit, OLPC is useless. But there are those who eek out a subsistence-level existence (i.e., they have food and water, albeit in meager quantities), but who’s lot in life could be improved through education and greater access to tools of economic growth, like cell phones and OLPC. Funny how few people mock the cell phone industry for selling (successfully) to poor people globally. Our economic well-being could not be sustained or expanded without tools of technology. I don’t understand how expect other countries to expand their economies without similar tools.


Om — have to disagree with you philosophically on OLPC’s role vis a vis children’s basic needs. Yes those needs are important and hopefully governments worldwide take this seriously. (I include the US but unforunately as you know even this country has serious problems taking care of children livin in poverty.) But the philosphical issue with OLPC is that children wolrdwide have the right and the need to have a basic education, and technology plays a critical role in today’s educational system. Why should we deny this to kids just because we are still trying to figure out how to meet basic needs. If governments stop thinking of this as a “zero sum” game — that buying laptops means fewer resources to go around to pay for food, healthcase etc — and instead view OLPC as an important contributor towards a long-term poverty alleviation strategy — then I think we’ll witness a shift in mindset that will do wonders for kids worldwide.

But that’s the blue-sky idealist in me. :-)

amos gregory


Exactly! Om, we have just demonstrated that the business model for MuniMesh networks does not work in Developed nations, but maybe the real use of MuniMesh is in underdeveloped countries. I hope that you will take the opportunity to talk to the folks over at GreenWifi and Meraki to guage their thoughts on this.


Om Malik

yes, sean, but to feed the wifi you still need the backhaul networks which are missing in most places. it is going to be slow process.


The OLPC is already fostering innovation in the network infrastructure business. Just look at Green Wifi (http://www.green-wifi.org/index.html) which provides a solar-powered WiFi grid device that can be deployed in remote areas. OLPC seed-funded Green Wifi as it can address the network connectivity issue head on.

Comments are closed.