Open Learning Exchange Nepal (OLE Nepal) in cooperation with Nepal's national Department of Education, launched a pilot-program to integrate OLPC XO laptops in regular school classes at two rural sites in April 2008.
OLE Nepal is one of the first organisations to do this kind of laptop deployment, and so it comes across a lot of problems for which there's just no guiding light that you can just follow. How do you make suitable educational software? How does the hardware hold out under these conditions? How do you prepare teachers that themselves often don't have any experience with computers?
So it would be very interesting to see from an independent source how the predictions and the decisions based on them have worked out in reality. A few months after the launch, Mr. Uttam Sharma, a doctoral student at at the Department of Applied Economics at the University of Minnesota, conducted an evaluation of the project so far.
First some facts, to get outsiders up to speed with the scope and goals of the project and the evaluation. At the moment 135 laptops have been distributed amongst grades two and six of the public primary schools of Bishwamitra and Bashuki, both on the outskirts of the Kathmandu valley.
The Danish IT Society generously provided the laptops for the schools. OLE Nepal provided teacher training, technical support, Internet connectivity, and in-house developed open-source educational software which follows the Nepali curriculum. Uttam Sharma collected his research material for the evaluation from mid June 2008 to mid August 2008. Amongst others he surveyed teachers, students, family members from 20 different households and relevant OLE Nepal personnel.
Excited about OLPC, 6 months in
To prepare the teachers that would use the laptops in the classroom, OLE Nepal set up a two-legged course to train them; which varied widely in scope. As mentioned before, a number of teachers had never even touched a laptop previously, so the training also included basic skills like using the mouse.
Other sessions were focused on for example integrating the laptops with the standard course material. All through those days the teachers were very enthusiastic and pushed the sessions well past their allotted time and into the night.
Just before the start of the pilot, the teachers from both schools, 21 in total, spent 10 days together to get acquainted with the laptop. When the teachers gave their thoughts about the training, a few points stood out on both the postive and the negative side. The things they liked: The way the program encouraged learning from your colleagues rather than simply from the instructor. The sessions about how to make lesson plans. The training supplied them with an opportunity to learn from new colleagues.
On the minus side, the teachers were in agreement that "a lot of time was spent on going over theoretical parts, which negatively affected the time spent on practical aspect", when using the laptop.
The other leg concentrated on in-school training. So a training which focussed more on how to use the XO in the classroom. Reading the evaluation, one gets the impression that the teachers appreciated not only the computer-related topics, but also very much teaching tips in general. When the teachers were asked what they found usefull, they mentioned encouraging the habit of incorporating lesson plans in their teaching, effective time management and how to present teaching material in the classroom.
Teaching in Practice
When evaluating the usefulness of the computers in the classroom the teachers seemed to be quite positive. "All (relevant) teachers feel that their lectures are now more organized and that it is easier to teach students new concepts." Also the "laptop based instructions have made it easier to give students more practice exercises. In addition, they also feel that the use of laptop based instructions has made the classes more interactive."
It's not all good though. "Many teachers feel that the noise level has substantially increased. This has to do with the fact that students have difficulty controlling their exitement." They have also trouble managing time, because they are not used to integrating laptop use in classrooms. And the teachers feel that the amount of effort that they have to put into their classes has icreased. The extra time spent ranges from making lesson plans to more practical issues, like the management of the laptops in the classroom.
As for the digital content supplied by OLE Nepal, the teachers were happy with the grade 2 content, but found that the grade 6 activities were often either to easy or to difficult. Also the amount of supplied content was to little to satisfy demand. And they would like to see the material being sequenced according to the chapters in the book.
From the students point of view, "more than 95% of grade 6 students find learning using laptops easier. As to why it is easier, the most common response was that you can do the exercise or activities as many times as you want. The second most common one was that these activities were very enjoyable to learn from. The students also appreciated the fact that you could do these activities at ones own pace."
Ease of Use and Laptop Care
Almost all teachers (16 out of 17) find the laptop and its layout easy to use. They think it is very intuitive and also feel that the students should not have much difficulty.
The biggest problem teachers, students and family members cited, was a hardware problem: The so-called 'jumpy cursor problem', which lets the cursor jump at random points on the screen when moved, so the computer is difficult to control. The "four-finger-salute", hitting a special combination of keys to reset the touchpad, helped somewhat but not consistently. More than a third of the teachers noted problems regarding sound: "Since earphones are not available, difficulties arise for students and teachers when using sound related activities. They largely ask students to listen to the sound from the teacher's laptop."
As for care-taking, the students and their family seem to be very keen on protecting the laptop: "At home the students keep the laptop away from fire and water. Those who have a closet at home keep the laptop there when not in use. The students securely keep the laptop in the bag. The family members are aware that the laptop should be stored securely." As a result, very few laptops are brought in for repair, and none so far have been reported stolen or lost.
In Summary
In it's conclusion the evaluation is positive: "The head teachers in both the schools consider the pilot program as very effective and see great promise in reducing the disparity between private and public schools." and "With this student-centered approach, students are interacting with each other more often. It has made students more curious and they are eager to learn new things. It has also helped in developing co-operative spirit as students are willing to help each other learn new technology."
Note: For more detailed review of Nepal's OLPC evaluation, please see Rabi Karmacharya's article here.
About the Author: Ties Stuij, originally from the Netherlands, volunteers for OLE Nepal in Kathmandu as a software developer. He wrote earlier for OLPC News about the E-Paath open-source learning activities he works on. Rabi Karmacharya, Dr. Saurav Dev Bhatta, and Bryan Berry contributed to this article.
Thank you, Ties!
In a sense, I'm interested in the reports from Nepal more than any other deployments. Please keep us posted.
Een droombaan Ties!
Ties, we did a TCO "study" on OLPCnews.
http://www.olpcnews.com/sales_talk/price/xo_laptop_total_cost_of_ownership.html
Obviously, this study was done without any practical experience of the use of XOs in the classroom.
Could you comment on the model we used?
For instance, we expected one week per year teacher instruction sufficient. However, your report indicates two weeks per year might be more appropriate.
Especially our estimations of expected replacements, 4 year mean survival time of the XO, and cost of electricity were made without any actual experience.
Did we forget important costs?
All suggestions and comments are welcome.
Winter
Excellent and insightful writing! I'd love to see more such impressions from other OLPC pilots and deployments...
Dank je Winter, t is idd nie slecht hier!
About commenting on your model. I'm not involved in the money side of things, so I don't dare to comment on precise costs. Since this project has only been in the air for about half a year, we also can't say anything about the mean survival time. I'd say the two most glaring omission in your calculations would be that there's no mention of some kind of support-organisation that takes care of infrastructure, hardware and software. Also there's no mention of content creation and maintainance (be it interactive software or for example digitized books). In my opinion content would be the whole reason for these devices, but there are different opinions on this.
And then I noticed you mentioned networking/internet, gave it the variable name X, but forgot to use it. You didn't get a compiler warning? Around these parts setting up a network, keeping it running, and paying for it is a job not to sneeze at. Unfortunately I don't actually know any of these networking figures.
Winter,
I do handle some of the cost estimates for deployments and I concur w/ Ties that the content creation and maintenance are significant costs that can't be avoided. Well you can avoid them, but at the cost of not connecting the XO to the existing education system in any logical way.
We have found that support costs aren't that high. I spend about 5 hours/month fixing XO's from our 2 pilot schools, that's 150 XO's. The work I spend fixing them could easily be done by one of our 18-year old Nepali interns.
The Internet costs can vary wildly depending on the locale. If you are using wireless point-to-point radios the real cost is in power backup and surge protection for the radios. The radios we use cost about $400 USD each (I think).
Bandwidth costs can vary wildly as well. The maintenance is really hard. We are looking at partnering w/ private companies where we pay for the network installation and they operate it. We then pay them for the bandwidth we use and they use the remainder to set up Internet cafes that they manage.
I think 80-90% of XO's will last 4 years or more. But certain parts such as the camera and microphone seem to break much sooner. Out of 150 XO's I have had 6 cameras and 4 microphones go bad. At the same time, only a couple of motherboards have gone bad.
@Ties and Bryan Berry:
"I do handle some of the cost estimates for deployments and I concur w/ Ties that the content creation and maintenance are significant costs that can't be avoided. Well you can avoid them, but at the cost of not connecting the XO to the existing education system in any logical way."
We were working with a model of a massive, >500,000 laptops, roll out (eg, Indonesia or India). When you amortise content creation over so many laptops, you end up with 35 cents per laptop per year for each $1M in the budget. If you use local labor, you can get an awful lot of content for $1M.
For very large roll-outs you can mostly ignore the fixed costs, like content. However, I agree that for small deployments fixed costs can easily dwarf the variable costs.
@Ties and Bryan Berry:
"The Internet costs can vary wildly depending on the locale. If you are using wireless point-to-point radios the real cost is in power backup and surge protection for the radios. The radios we use cost about $400 USD each (I think)."
The reason we ignored internet access, the X in the budget, is that we had absolutely no idea at all what to make of it. Especially not in a large roll-out. If I want to hire a satellite link and put a dish next to the school the costs will be horrible. If the government can convince a local phone provider to supply, or even sponsor, free-band internet-over-GSM connectivity in down hours, the costs could be low indeed.
The point is, that if a government asks around for a tender on supplying internet access to tens of thousands of schools, they will get a different answer than when you ask it for an individual school or village. We also have things like "Internet for the Developing World: Offline Internet Access at Modem-speed Dialup Connections"
http://www.cag.lcs.mit.edu/~umar/publications/ICTD2007-fin.pdf
The OLPC could piggyback on other, existing, initiatives that could be adapted or extended to the XO:
http://news.bbc.co.uk/2/hi/technology/7106998.stm
(although this seems rather expensive)
In short, as Bryan already wrote "Bandwidth costs can vary wildly as well. The maintenance is really hard." We did not put a price tag on the internet access because we had no clue at all what a realistic price would be. I suspect that a lot of the benefits of the XO can be realized with slow, off-line, internet access. Maybe by getting a special deal using cell-phone networks in night hours. But in no way am I knowledgeable in that area.
Winter
@Winter:
We were working with a model of a massive, >500,000 laptops, roll out (eg, Indonesia or India). When you amortise content creation over so many laptops, you end up with 35 cents per laptop per year for each $1M in the budget. If you use local labor, you can get an awful lot of content for $1M.
I concur w/ that. You can get a lot of content for $1 M. If you spend more than 1 M I suspect you will waste a lot of money on an overly large an inefficient team.
"I suspect that a lot of the benefits of the XO can be realized with slow, off-line, internet access"
From my experience, the XO can have amazing impact w/ out Internect access. But it is very nice to have, particularly for remote maintenance and incrementally providing more local content to the schools.
great comments Winter
@Bryan (&Ties):
"great comments Winter"
That is the beauty of the internet. I would be completely useless as a field worker or technician, I have no organizational talents, and are a lousy developer. Still, I might occasionally be able to help out with a little comment on things I might know enough of to be useful.
Keep up the really great work.
Winter
@winter
you wrote: "The reason we ignored internet access, the X in the budget, is that we had absolutely no idea at all what to make of it."
Sure, but it does look weird to just ignore that cost, and proclaim a total cost of 8 cents. I'd perhaps do a low- and high-end estimation, just to make things clear.
@Ties:
"Sure, but it does look weird to just ignore that cost, and proclaim a total cost of 8 cents. I'd perhaps do a low- and high-end estimation, just to make things clear."
Please do! The point is that I am so clueless about this for, say Asia, that I do not even know what low- and high-end estimations to use.
Satellite might develop rapidly, but not necessarily be cheap:
http://www.scidev.net/en/news/satellites-to-bring-speedy-internet-to-developing-.html
I simply cannot get any reasonable estimates from where I sit. Here are two sites with links to many (most) initiatives:
Hardware
http://www.infodev.org/en/Publication.107.html
Connectivity (often sneaker based)
http://www.infodev.org/en/Publication.109.html
Maybe there is some information hidden there (eg, in a language I cannot read).
Winter
Ties,
I obviously should have used Vital Wave's estimates for the high end prices:
http://download.microsoft.com/download/2/0/a/20ac945c-34d0-4a60-8245-f80e80fe954f/Vital_Wave_Consulting_Affordable_Computing_TCO11June08.pdf
Page 12
Rural: $3600 Urban: $2100 (5 year)
Where the rural case is for dial-up and the urban one for broadband DSI.
This would come down to $720 - $420 a year.
This is for the 16 computers in a computer lab. If we throttle down the bandwidth per student to get the same bandwidth for 60 pupils, we end up with $12 - $7 a year per student, or 1.2-0.7 cents per hour. A significant portion of the total costs.
So this means the rural access point servers must cache most of the pages and might have to work with an urban proxy that uses heavy compression for communication. I think schools will have to stimulate students to work with pictures blocked (which would also help with some "filtering" problems).
So this is a high-end low-bandwidth (worst case) prices: 1.2 cents per hour internet costs.
Winter
I wouldn't be surprised if you can pave the roads for trucks to carry the tapes and hire people to manually download from a list of URLs before you match the cost and reliability of satellite service.
Communications satellites are good for:
1. Unidirectional high-bandwidth broadcast (TV and radio).
2. Internet for places so remote (ships at sea, Antarctica) nothing else can reach them.
3. Emergency services and military, to remain connected when all infrastructure is destroyed, broken or impossible to build.
In addition to being not anywhere close to the scale of school's uplink, all mentioned applications (except for the first one) are expected to be extremely expensive, and satellite companies don't have much competition, so their prices are far outside the reasonable range.
Anything with terrestrial infrastructure -- cellular network, wireless relay stations, T1/E1 lines, or combination of those things, would end up cheaper in the long run, and their latency will be few times lower.
@Teapot:
Did you actually read the report in the link? A lot of standard telephone connections go through satellites.
Personally, I think some way to weave TCP/IP between cell-phone calls will be the cheapest option. And some requests can go the way newsgroups and email did, in batch and non-interactive.
This is not the WWW we know, but it is already of a different order from what they ahve now, ie, nothing. Caching can also help, at least in education.
Winter