Shared Access Computing is the Most Economical and Scalable Model
As Wayan appropriately points out in his introduction, a computer is merely a learning tool, albeit an increasingly important tool, in enabling higher quality education. And as Walter Bender pointed out in the insightful WSJ debate Will Low-Cost Laptops Help Kids in Developing Countries? with the CEO of NComputing, Stephen Dukker, “computing is not a cure; it is an agent that will enable children to engage in learning.”
So the debate we’ve been asked to participate in is to posit which computing model is better suited in the developing world to proliferate computers to enhance learning and education.
Intel's Classmate PC
Back in 2006, when I was co-General Manager of the computer division at Intel that was developing the Classmate PC, Intel was heavily promoting notebooks (which had higher average selling prices and higher margins than desktop CPU’s).
It may surprise some given my involvement with the Classmate PC, and Intel’s overall strategy, that I was not a proponent of 1:1 computing in the developing world. My passion for significantly increasing the access to computers for those in the under-served markets ultimately brought me to the role I have now at NComputing.
Access to fully functional, ultra-low cost, highly energy efficient connected computing is a critical component of enhancing and enabling the learning experience. My belief continues to be that shared access continues to be the best starting point for developing countries that are introducing computers to their schools for the first time.
First and foremost, if mature markets have not adopted 1:1 computing in any great degree beyond higher education, how can we realistically expect emerging markets with more limited budgets to adopt 1:1 computing?
The math is simple. Is it better to have 1.8M students share access to 50,000 computers for the first time vs. wait until the government can afford to proliferate notebooks to the same 1.8M students.
In the 1:1 model, who get’s these computers first? This particular example is from the state of Andra Pradesh in India:
The government saved $20M by deploying the shared model in acquisition, maintenance and electricity costs. They were able to
- deploy more computers and
- purchase generators to keep the computers running during power outages.
The $100 target price of the OLPC laptop was originally only the purchase price, regardless of being able to achieve it or not. There are other significant costs occurred during the life of a single computer, including maintenance and electricity. Secondly, where is the point of diminishing return where the farthest extreme is having a computer at a student’s fingertips 24/7?
As a longtime professional in the IT industry, I would be lost without my notebook by side. Blackberry’s, iPhone’s, etc. have reduced that dependence. But what about the kindergartner or sixth grader. I would agree that having increased access to shared computer model (more than one hour a day) would be better, but surely these students don’t need a computer with them all the time?
You could argue by digitizing textbooks you reduce their backpack load, but I have not heard of an outbreak of K-12 student back problems.
The portability aspect is another challenge, especially in developing economies. Kids drop and lose things in general. They have not developed their judgment skills to a point where they can be responsible for a notebook. I finally broke down and got my son a mobile phone – he lost it within six months, and if you looked at its shell, it is considerably marred.
I am not entirely against 1:1 computing, and in the subsequent debate we will discuss hybrid models that could work, but when it comes to primary and secondary schools, I do feel strongly that economic realities strongly support shared usage. I try to illustrate this in the chart below:
This is not a hybrid model. This is an evolutionary model. As students’ age/mature/progress, the need for a computer all the time becomes more critical. In addition, everyone has different needs, abilities, talents and skills. Some will gravitate towards the computer as if it is an extension of their body. Others will find it mildly useful but will prefer paper, pencil, books, etc.
This is where “try” vs. “buy” comes in. I would argue that 99% of people in the developed and developing world over the last 30ish years since the PC was introduced “try” before they “buy.” Whether it is a parents PC, a school lab, a cyber café, telecentre, or work place, they will be exposed first then build the interest and knowledge.
This is why, at Intel, before the Classmate PC “creosote bush” squashed all other projects (Rural Community PC, Amazon Kindle… yes, we were partnering with Amazon and e-Ink on a text book replacement product, and more), we had a significant push towards “shared access.”
In conclusion, I laud the efforts of Intel and OLPC who have significantly increased awareness of the importance of computing in education. The question and debate remains, though, as to how computing is deployed. The most economical and scalable solution is shared access computing.
I agree with the premise that it is better to have more students share access to computers for the first time vs. wait until the government can afford to proliferate notebooks at 1:1 rations, but I wonder if we can have a better learning experience than what is currently happening in many schools – "computers" taught as a subject rather than used as a tool to learn things (math to dance) with.
Wayan, i agree that computers should be used as a learning tool, not just for computer skills (of course first course is computers 1:1), but as a vehicle to enhance learning and content delivery. Luckily, there are many examples of this in the developing world. We are working with one such company in Chile and Ecuador that has a good education software package. In addition, Rosetta Stone is being used more and more to teach language skills through the computer.
I believe shared computing is more cost-effecrtive. One question: Are these computers used 16 hours a day or more. If not, why not?
@Mark: I wonder if the economics/math in the 1:1 scenario would change if the computers could be built (at least partially) locally and serviced locally also? Food for thought.
(Full disclosure on me…I am in marketing at Intel and do share Mark's passion for seeing technology improve people's lives.)
I understand how these computers are being used to improve general education. I don't understand how these computers are being used to improve the living standard of BOP personnel unless one believes that a small amount of computer education will automatically improve their living standard (which I do not).
How are these computers being used for needed vocational training? In particular, how do these computers improve the lives of subsistence farmers? Why are they are better investment than irrigation systems, fertilizer, or pesticide for subsistence farmers? How many jobs have the computers created for BOP people living in urban slums? How do BOP people in urban slums use computers to make a living? Why are computers a better investment than microfinancing or use of a Grameen phone for BOP people iving in urban slums?
@Paul As I mentioned at the beginning of this post, a computing device, no matter what kind, is merely a tool. How it is utilized is what matters more on the impact to living standards. An ITU study in 2003 implied that a 1% increase in "infodensity," which I will loosely term ICT access and availability for simplicity, results in an .9% increase in per capita GDP. (You can find the full definition in the report here: http://www.itu.int/ITU-D/ict/publications/dd/mate…
That being said, you need the device at a minimum regardless of how it is used. And the more economical the device, the quicker and more broadly it get's deployed.
Vocational training is an important point. I don't know the facts on this, but I would think if you were to compare two individuals in a developing country, both with comparable skills plus a base-level education, except for the fact that ONE of them is computer trained, while the other is not, I would argue that the one with the computer skills is much more employable than the other (please forgive my generalization to keep my point brief).
Regarding your other examples, all of which are good, I would argue that there is a hierarchy of needs, of which at some point, a computer matters.
Mark,
I am trying to improve the lives of hundreds of thousands of BOP personnel in Cordoba Colombia. I know how to improve the lives of people in the rural areas (see description of Mas Dinero project at http://home.comcast.net/~prigter/site/ ) but I don't know how to improve the lives of people in urban areas such as Monteria and Sincelejo. ECHO (see http://www.echotech.org/mambo/index.php ) provides excellent vocational training information in Spanish and English for rural people but I have never seen comparable information for urban BOP personnel. Perhaps your readers know of some ideas. Assume that these people speak Spanish and have gone to school from 3-6 years so their reading skills are limited. They can probably obtain up to $100 of investment capital. Assume unemployment is at least 30%. Food security currently is provided by eating plantains, white yams, and yuca. Their annual income is $300-600/year
Wayan
You also may want to focus on the following. I have been told there have been millions of micro finance loans made. How can computer professionals improve the record keeping of the microfinance industry so that better BOP loans are made in the future? What are the best microfinance record keeping programs currently available? In particular, I would like to know that given existing computer records, what have been the most succesful BOP investments for a particular climatic region, population density, and income of the general population?
Paul,
Actually, I'll not be focusing on microfinance loans, their accounting systems or how the industry impacts the BOP. Will all due respect, this is not the Microfinance Technology Debate – its the Educational Technology Debate and therefore we'll be sticking to technology impacts on the education systems of the developing world.
If you'd like to know more about the most successful BOP investments for a particular climatic region, population density, and income of the general population, I would suggest you try NextBillion. Good folks over there who would gladly entertain your inquiries.
Wayan
According to my training at Computer Sciences Corporation, there are six things you can do to change educational systems technology. You can change the following
1) Systems used
2) Data provided
3) Applications available
4) Physical location of the systems
5) Organizations and people that use the systems
6) Logistics
I believe there are major weaknesses in the technology being offered to the educational systems in the developing world in the areas of applications available and data provided (I realize that you and Mark are primarily interested in the systems used). However, I believe the major requirement for educational technology is to teach BOP personnel ways to get out of poverty using the computer as a tool. One cannot focus on only the systems aspect (I am sure you know all this).
I am in contact with the people in NextBillion and appreciate your ideas and opinions
The trend is for mobile phones to have the power of laptop and desktop computers. Most children have mobile phones and they out-number PCs and laptops at least 3:1. If I had VGA out and a USB keyboard driver for my G1 I could dump my netbook and probably do everything I need for running our business from my phone. Since this trend will continue, it seems likely that the school might provide keyboards and screens but the learners will carry their own computers everywhere they go. It will probably take a few years for school administration to get their heads around it all but in the end if they don't, schools are going to look increasingly anachronistic in a world where all but very specialised information is free from the internet to anyone with a cell phone.
If the computer industry could design a computer system that would help poor people earn $3000 rather than $300-$600, I believe a billion people would buy them for a $1000. However, the necessary requirements analysis on what is needed has not been done. I used to lead between 80 and 500 computer professionals as either the technical director or project manager. We would usually conduct a requirements analysis before we started any project rather than pick a technology and see what that technology could solve. I am not aware that a good requirments analysis has been conducted on how to improve educational technology for the world's poor. If you feel differently, please tell me where it is. Colombia is a leader in improving the technology of poor people as described in the book Gaviotas: A Village to Reinvent the World by Alan Weisman. Although I work in a different region, University, Government, and Church personnel in Monteria and I are making substantial progress.
Its called the smartphone 😉
We need to stop focusing primarily on yesterdays and to-days technologies and look at business models and social enterprise models that can sustain learning in the developing world. What we need is a free on-line K-12 curriculum so that anyone with an internet connection can access it. Then we need low cost devices. Smartphones perhaps connectable to bigger screens and keyboards could do that especially as they evolve further. Many more people in the developing world have cell phones than PCs. We then need to pair schools in the rich west with those in the developing world so kids can gain social enterprise qualifications in schools in the developed world for raising money to provide an internet connections for their peers in poorer countries. They can then communicate and those in the developing world then have access to learning from the on-line curriculum. Education is really the only hope for long term leveling up of global living standards. For education the first need is a curriculum and organised learning environments, the means of accessing this will get easier and easier as time goes on.
I think its interesting that you mention a "free on-line K-12 curriculum" when there is even less Internet access than access to un-connected computers, and the provision of Internet access can be even more expensive, on a 5 year basis, than the computing technology itself.
I think we'll see at least another decade where intermittent Internet access (at the best) is the norm for most schools in the developing world. And I'm being hopeful with that estimation.
Since cellphones outnumber PCs more than 3:1 and are rapidly becoming routinely connected to the internet, it is fairly easy to envisage a disruptive innovation that shifts the de facto standard for internet access to the cell phone. A recent UN report reveals that half the globe now pays to use one, with the fastest growth taking place in Africa and a 50 billion investment in broadband networks there. Ok it might take 10 or even 20 years but it's likely to be the future so given that, starting on a free on-line curriculum is a reasonable strategy because that too will take time to develop. Let's face it a lot of the developing world does not have electricity but that isn't a reason to say that will be the case forever. Its more getting sustainable social enterprise models than specific technologies because then you can finance sustainable access to the information needed for education.
Why not load the full K-12 curriculum on a computer server and distibute the server with an appropriate number of workstations to schools that need it. You would not need internet access and would save the cost of buying a full set of textbooks for each student as the students progressed from one grade to another. With properly designed software you could have the student interact with the textbook material from his/her workstation.
Updates to the K-12 curriculum could be distributed on a yearly basis or as needed. The curriculum material could be translated into a wide variey of languages by the computer software.
Because there are all sorts of issues getting schools to install stuff on servers and why should we not provide the flexibility for individuals not in schools to learn in the way they want to? Of course you could have an option to download the stuff but what about links to thousands of other sites on the internet? Generally we don't install Wikipedia on a local server or You Tube, My Space or Facebook. I'm not talking about just a big text book on-line, but a means of letting learners participate in designing and originating their own learning resources. That is a logical extension of web 2.0. The idea that you "update" a curriculum every year seems very limited given the way technology has changed. Language translations are better done by people even now and projects like Wikipedia and web based content management systems such as Drupal have all the tools for doing it.
You may want to check out projects like Moulin and eGranary for simple yet effective "content on USB" examples that do not require connectivity at all and can offer great content repositories to everyone quickly.
Ian,
School personnel would not be asked to install stuff on servers. The system I envision would be a server with 20 workstations that costs less than $2000 (including software) and requires no maintenance except for yearly updates ( a turnkey system). The system you are proposing seems to be more expensive and seems to require much more system maintenance by computer professionals (which can be expensive). The system you are proposing seems to require training that third world teachers don't have (It is expensive to train them). What are your budget requirements? I am trying to keep the costs to less than $100/child including maintenance costs.
Ian,
School personnel would not be asked to install stuff on servers. The system I envision would be a server with 20 workstations that costs less than $2000 (including software) and requires no maintenance except for yearly updates ( a turnkey system). The system you are proposing seems to be more expensive and seems to require much more system maintenance by computer professionals (which can be expensive). The system you are proposing seems to require training that third world teachers don't have (It is expensive to train them). What are your budget requirements? I am trying to keep the costs to less than $100/child including maintenance costs.
I've found many benefits of 1:1 since I've had it in my classroom, but also many technical and social challenges. But to date, the benefits have outweighed the disadvantages. The ad-hoc internet searching to find just-in-time information is super-valuable and the classroom discussions generated have been great. Then there's the jewel-in-the-crown – using the laptops as "clickers". We use a program called Student Response Network – http://studentresponsenetwork.com – which has revolutionized what we do. I can get instant responses from the whole class to a variety of questions and the brainstorming that happens with the free-text response has been fabulous. Highly recommend this for any 1:1 setting.
This is such an awesome useful useful resource that you’re offering and in addition you give it away for free. I like seeing web sites that perceive the value of offering a high quality useful resource for free. It?s the outdated what goes round comes spherical routine. Did you acquired numerous hyperlinks and I see numerous trackbacks?