Slashdot Mirror
Ask Slashdot: Setting Up a Computer Lab In a Developing Country
First time accepted submitter levanjm writes "Hi all, I am looking for some advice. I am a mathematician at a small liberal arts school who has dabbled in Linux for a number of years. I have had the chance to teach a few courses and summer camps about Linux to college and high school students. Recently I made a trip to Guatemala and visited a school in Labor de Falla. While there I was talking with people associated with the school about how great it would be to be able to set up a computer lab for the kids. To make a long story short, I approached my school about finding a way to make this happen and to get my students involved in volunteering. I have received notification that my school has given me an in house grant to try to get this project rolling. They have also donated six computers to get things started. While I have been making plans in case the funding came through, I wanted to open this up to as many eyes as possible because I am sure there are plenty of concerns I have not considered. What are your thoughts on how to best implement the lab setting? I am a firm believer in the Open Source philosophy so proprietary software is not on my radar. The PC's donated are a little old (4 or so years old), but would run Edubuntu without any issues. I originally thought about how awesome a Raspberry Pi lab would be to set up. I am also wondering if there are any Kickstarter type of foundations that might be used to help solicit donations to purchase additional equipment and help cover costs of getting the equipment to the school. It would be amazing to get enough funding to give computers to the teachers in addition to a lab. I am sure there are other issues I have not even considered yet, so any thoughts you have to share would be wonderful."
What is the purpose of the lab? Since it is a high school I am going to shoot low and assume “keyboarding skills” and basic net access rather than programing or CI.
What type of infrastructure does the school have? i.e. how good is the electricity and security?
What type of support does the lab have? What skills do the teachers have.
Answers these questions and I think you will have a much better idea of what you need.
You might also want to look at a multi-seat setup. ie 1 reasonably spec'd computer, with several monitor+keyboard+mouse sets.
Is electricity consumption an issue? A class full of pentium 4 computers is going to cost quite a bit in power. maybe enough to be worth paying for newer hardware instead.
Without knowing what those 4y/o machines are like, let me just suggest doing the math on shipping them down there and their ongoing power consumption, as well as checking into parts availability. It is possible that something like a briefcase full of PIs or similar be a better idea from an operating and implementation expense standpoint.
I am a firm believer in the Open Source philosophy so proprietary software is not on my radar."
I stopped reading right there. Setting up a computer lab is a good question for Ask Slashdot. Setting up a philosophical/religious indoctrination center is not.
Yes, heaven forbid they learn how the system works and learn how to modify it and build on it. That would be too active and involved for your liking, right? They should just be users, passive consumers of a black-box product that they can't inspect, can't modify, can't build on.
You see, that's a practical reason for wanting Open Source. You are the only one making this into a philosophical/religious matter. Apparently it offends you that people want to buy a car that doesn't have the hood welded shut, that people might want to use a computer system that they can understand and build on. Tell you what. If you don't like Open Source then don't use it. Simple? Those of us who want to learn will continue using it. Then we can both be happy.
So sorry it bothers you that other people want to do good things in a way that's not quite how you would do them. I bet they wear a brand of clothes different from yours too, those insensitive clods. Next time you want to move to a third-world country and provide for people who often have next to nothing, we'll then pay attention to how you think it should be done. Will you charge them hundreds of dollars for copies of Windows and Office and associated software so they can actually use their systems? That'll be only about several months to a years' wages you know. Per copy.
Making use of those computers is the difficult part.
Where will the computers be stored? At the very least, they need to be in a secure room in the school, free from leaky roofs (especially during heavy rains), free of bugs (ants love computers!), and you'll need bars on the windows and a reinforced lockable door; Is there security at the school? What about electricity? Is a generator needed? Voltage stabilizers? Or will they be using solar panels and an inverter system? Who will be appointed to manage the computer room? Do they need training? Will there be internet access? How will that be paid for? You may be able to make a special deal with a local provider (large corps love to look good by giving back).
Again, what will the computers be used for? Do you need specific software? If there is no specified curriculum, the kids will just be browsing porn and playing games (teachers too!) Or they'll be using them for private reasons.
The key word here is sustainability. Its very easy to give a poor school some hand-me-down hardware, its much more difficult and challenging to turn it into something useful, sustainable, and create a place where children will actually learn something about information technology.
Being fully for FOSS is as ridiculous as being fully for proprietary software. The balance lies in the mid.
There is an organisation in Australia called Computer Bank that has been collecting and re-purposing computers for years. There are a number of international organisations that do similar things, some are listed here - http://www.computerbank.org.au/links.
You could also seek sponsorship from a hardware vendor such as Dell - http://www.dell.com/Learn/us/en/aucorp1/corp-comm/corporate-sponsorships, or even Google.
Since it needs to be shipped there, perhaps approach transport companies. Also, some technology recycling companies will supply equipment cheaply or free for the right cause.
Some organisations require that a not-for-profit organisation be established before funds are allocated. You'll have to check your local laws.
I am a firm believer in the Open Source philosophy so proprietary software is not on my radar."
I stopped reading right there. Setting up a computer lab is a good question for Ask Slashdot. Setting up a philosophical/religious indoctrination center is not.
You must be new here.
"Yes, heaven forbid they learn how the system works and learn how to modify it and build on it. That would be too active and involved for your liking, right? They should just be users, passive consumers of a black-box product that they can't inspect, can't modify, can't build on. "
That has zero to do with Open Source philosophy. You have to learn the difference between "Open Source philosophy" (RMS), "Open Source", open systems and access to source. It is nice that there are people that have enough money or a guaranteed income from another source that they can give away their work, "Open Source philosophy" is not about that, its about FORCING people to give away their work by creating a platform that will allows nothing else.
Any dickbag license is BAD even if it claimed to support "freedom".
Your putting religion ahead of usability, and that's a mistake. The purpose of a lab is to educate your students, not indoctrinate them in your in your religious beliefs. What can your students use in their country with the skills they would develop and make a career out of? Can you make a career out of a Raspberry Pi, or is it more of a really cool toy?
If your local country values Windows for employment more than that is what you should use, because that is what will help /them/ build a future. Microsoft has educational versions of their products available for next to nothing worldwide, so cost isn't an issue for Windows and Office.
If your local country is all about Ubuntu than you use that because that is what is valued. The only way to know that is to talk your local business leaders and find out what /they/ value. Do they value someone that knows how to run a Windows computer and use Excel or do they all use LibreOffice? Leave your personal religion out of this and give your students what they need for their future.
Getting locals onboard is the most important thing. Dropping a bunch of PCs and running away is the easy part. You need someone local to take care of them, and someone (else ?) to teach on/with them.
Depending on the goal, PCs are probably a bad idea: transport alone costs you more than buying a bunch of $40 Android USB keys, let alone power and security issues. Android keys are OK for Internet stuff, even light Office work. Some can even take Ubuntu, if you want to force your philosophy at the cost of practicality. You'll need HDMI screens and keyboard+mouse.
In many cases, tablets will actually turn out cheaper, taking the screen into account. OLPC for edu cred, or any sufficiently solid chinese one.
In any case, you should ask the users. Depending on their setting, their goals, their expectations and constraints, whatever you get told on Slahsdot can be way off the mark and utterly wasteful of time and money.
The Cloud - because you don't care if your apps and data are up in the air.
The one potential issue with the rPi is VGA, or lack thereof.
For some insane reason, even to this day, when they barely sell analog CRTs anymore, bottom-of-range LCDs still come with only a VGA port. I don't know why throwing an extra ADC stage into the mix is economically viable; but them's the breaks.
An rPi will happily enough drive an HDMI display, with a basic HDMI cable, or a DVI display, with a passive adapter cable that costs barely more than a straight HDMI cable; but if you want VGA, prepare to pay more than the rPi cost for an active(which means powered, which probably means a nasty little wall-wart) converter box. It also has composite; but composite video brutalizes text pretty badly, making it ill suited for most computer applications.
If you are buying everything new the premium to get a monitor with some sort of digital-in isn't all that big(probably smaller than a VGA converter, and a lot more elegant) and the classier brand of off-lease donor gear should also have DVI support; but don't expect VGA to work, because it won't.
http://edubuntu.org/documentation/ltsp-live
In fact, there is the simplest setup possible. they even have Live CD images to get your setup online and running in less than a day.
Do not look at laser with remaining good eye.
More than 10 years ago, in South Africa, there was a noble and concerted effort by the open source comunity to expose computing to children in under-developed schools. Ubuntu, with its strong connection to South Africa through Mark Shuttleworth, was at the forefront of this effort. I was one of those in the vanguard and was convinced that the open source approach would give all participants a better grounding and understanding of computing. Then Microsoft made its Windows and Office software licenses available at no cost to all schools. It took literally weeks for almost everyone to switch to Microsoft. Ironically, it was only well funded private schools whose students were interested in software engineering careers that retained open source computer labs. Looking back we now understand why. Undeniably, Windows is the de-facto desktop standard in the business world. What little prior exposure eiher teachers or students might have had, it was almost certainly with Windows. Windows skills were generally more useful than Linux skills in a non-technical job market; for those whose future employment prospects are limited the choice s clear.
I lived in the Marshall Islands for over 7 years and was tasked with a project not unlike yours. Here are some of the issues I confronted.
1. Climate control. Spend part of your budget on a good window a/c unit. The tropical moisture will wreak havoc with your lab.
2. Use local talent. Within a day of starting your project there invariably will be a local who starts hanging around asking questions and wanting to know more. I had 2 Marshallese students who found what I was doing very interesting so I started teaching them the basics of networking and basic computer repair. They caught on very fast and when I left they were able to take over the entire network we had built and keep it maintained. Local talent also will win you over with other key local decision makers. Locals want locals to sustain projects, not an endless parade of expats.
3. Determine your usage requirements. Do you have internet? How fast and reliable is the connection? Will you be teaching classes? What kind of classes? The computer labs we built started out with no internet access as there was none on the island. We were still able to teach word processing and spreadsheet classes(you may want to think word and excel here because more than likely that is what your local government uses). Eventually we were able to get limited internet access through a partnership with the university of Hawaii. The connection however was only 56k and was bounced off an old GOES satellite(GOES 7 to be specific). Because these GOES satellites were in a figure 8 orbit, the 3m dish required constant realignment. This meant that community access to the internet had to be at very specific times and eventually we went with a white-list of sites that would fit the needs of most of our users. This meant we could share 56k across 20 computers without worrying about porn and games. Slow but it worked and the community love it.
4. Have a good maintenance plan. Stuff just breaks more in the tropics and getting new parts can be difficult and time consuming. Keep a few extra network cards and hard drives around. Make sure the lab is cleaned and dusted regularly. Bug bomb as well. Cock roaches love computers. Try and keep your computer hardware as consistent as possible. This allows you to setup software images that make recovery much easier.
5. Have a detailed usage policy. If someone wants to save a copy of a letter to a relative in another country, how do they save it? Where do they save it? You will be very surprised how quickly the desktop and other directories start filling up with stuff from your users. Keeping a consistent and universal interface that is uncluttered for your users makes teaching SOOOOO much easier.
Finally, remember that you won't be there forever (although you have probably met a few expats who were like you but never left) and that someone will have to take over what you leave behind. Sustainability, in my experience, was the most difficult challenge of any project. I have seen brand new office buildings built by grants from other countries literally crumble into to disrepair within in 3 years because no one knew what to do after the building was completed.
Finding eager local talent to take over will ensure the hard work you are putting into this project will last long after you are gone.
I am very jealous of your opportunity. Good luck!
I know nothing about Guatemala except that my best friend in college lived there and he introduced me to masa which is delicious. Wouldn't surprise me if Guatemalan food is a really healthy alternative to ordinary Western cuisine, I wonder if they grow that non-sweet corn in the U.S.? (google guatemala masa). See below some of this may not be useful since it seems you are not so much in the boondocks.
I have a friend who did this in Cambodia. I remember he got Apple to donate computers (this is one reason why not using open source hardware may have a good point, it counts as CSR (Corporate Social Responsibility) for a manufacturer to do so). It was for an orphanage he created, and the idea was to educate the next generation of leaders. Also he started a newspaper, probably also had Macs I forget.This was over a decade ago. Point being, they had to hire two armed guards so things wouldn't be stolen and I believe one guard was killed. FYI.
Getting locals who will carry it on, and talking to (global) missionary group as other posters mentioned are good ideas. I believe Hope Worldwide was a group he worked with for this charity.
Using open source may be cheaper and may help jump start an industry even if you had a university (need to connect them to the Net possibly) and local people who are enthusiastic.
You may be able to get the World Bank to help you, I know they did a dollar matching program for building rural schools (villageleap) of which hundreds were built.
Power and telecommunications may be a big issue. I'm sorry I don't have data for you but you know it is not first world. Maybe there are no phones and power? I remember one original idea was to have a networked school be a hub for the community, don't know how it worked in the end but I do know one thing they did was have a wifi equipped motorbike travel among rural schools and pick up messages. Useful for medical care.. Also the geography etc. makes you wonder about can you get a line of site to an access point, can you get wind power, etc. Of course the top priority for a community might not be computer education. Maybe power to cleanse drinking water, or communications to notify a doctor they need to get a helicopter somewhere. Getting X-rays sent to a specialist hospital was one thing we did but you don't need that.
On the other hand if it is the Labor de Falla that is 17 nautical miles from Santiago, then it is just a suburb not in the boondocks over the horizon from wifi. Possibly you could even get support from some place like Microsoft or IBM, if you say you are going to start training locals in computer science from a young age. Apparently Google discovered a mother load of such talent in Viet Nam just the other day (on /. today). Maybe that is your goal.
Anyway, figure out what your goal is, and don't spend all your time on the technical side. The key to making these kind of projects happen is getting the parts together, putting your own time in to monitoring it daily with someone on the ground, and being extremely tenacious and single-minded about getting this goal achieved. But you need to listen to people there and if there is no enthusiasm or problems maybe you need to ask what they want. There probably are a lot of smart people within 1 hour of your Guatemala location and not clear that they even need you. So I would focus on fund raising, enabling it, setting a mission and making sure it happens.
Just my 0.02, I clearly know nothing about the area. Best to be sure you accept there may be things you also don't know about it, and try to set smaller achievable goals for yourself. Maybe you can get a manufacturer to get you new equipment for free, that would be best. Imagine you are the student there. As for linux, yeah it would be nice but depending on the age group if they need to get a job in the city will it really help them? If you can make a success maybe you can then scale it up and make that part of your timeline for phase II.
I've been setting up and teaching computer skills part-time in northwestern Cambodia for about 8 years now.
Getting equipment to a remote location is an expensive and perilous task. Damage, theft, bribes, delays, fees, more bribes, and unforeseen problems will cause you more headaches than you can imagine. I buy my equipment locally from a seller I have built up a relationship with. Because I'm a repeat customer, he goes out of his way to make sure the computers keep running when I'm not there, which is most of the year.
Because Cambodia is one of the poorest countries in the world, the idea of spending $300 on a copy of Microsoft Office is unthinkable. That's enough to run a small household for a couple months. You can install open source alternatives if you like, but it might not be a necessity since the machines are chock full of apps.
Getting the lab running is frankly the easy part. Your lesson plan needs to take precedence. Teach skills that are commercially viable in the country, inspire them to learn more, and give a solid foundation of basic skills. I have former students that can directly tie their lessons to helping them find jobs later on. They then take these skills and teach others, which creates a virtuous cycle. Good luck!
Yes, second that, I'm from the UK, white[ish] preppy type education, still don't feed the trolls. On the subject, actually this is a big advantage for Linux in these settings anyway. I've set up a couple of drop-ins in the east end of London where there's a fair amount of random crime and I've used recycled computers and Linux. The computers have practically no resale value, so they are not worth stealing and they can be replaced pretty quickly and cheaply. Other projects with 'brand new' have, in fact, had trouble apart from being trashed by all the viruses associated with random downloading because they are Windows based projects. Ours have the problem that 'we can't teach Word' for example, this may not be a problem in your setting.
I wouldn't use the Rasberry or only have that as a small hardware hacking part of the mix, for reasons stated in other posts. I agree with LSTP idea, if power is going to be a problem, though it's harder than individual systems.
Finally we've always found sustaining the teaching to be more difficult than setting up and sustaining the infrastructure. The next project I start will include a heavyweight, teach the teachers element too.
On y va, qui mal y pense!
I would concentrate some effort on making sure that no matter what happens, they have a clear path to "putting it back together".
Grub boot option to "reprint" the individual machines: clonezilla backups. If these machines are going to be used and then reprinted,
you might want to look into a grub partition that can clonezilla reprint the machine at boot time. Are they going to be able to save
their work ? Print? They should be able to print so they can prove to someone they did the work.
Better they should learn now embedded systems like the RPi work than any of the computers used in the "Developed" world?
Teaching them Linux scratches a philosophical itch - using software without license fees solves a practical financial problem. The desktops being donated most likely have WinXP or (gasp) Vista OEM licenses/CoAs on them, meaning they already have "proprietary" software licenses without cost, and large swaths of open source software runs on Windows, Mac, and Linux/*nix.
The original poster (of the question, not the parent of this comment) supports the philosophy of open source, which comes across as foisting his own beliefs on the folks in the "developing country" of Guatamala. Rather than teach these high school students about using the software and tools the Guatamalans use in the cities of Guatamala, he'd rather run a little social experiment and see how they do with learning an OS used on about 1% of desktops and hardware most folks hook up to their TVs for a laugh instead of using very reasonable (likely Core 2 Duo) desktop systems and the bundled OS (WinXP, in use by many/most Guatamalan businesses I assume).
Is his goal to help them enter the workforce or see if he can foster a village full of little Linus Torvalrds & Richard Stallmans?
Ken
Obviously there are teachers, there are great forums with a nice community and hey... figuring out problems is one of the most best things that can happen to your learning-curve.
You see – I don’t give that as a given. You need some combination of intellectually curious students and teachers. Assume the students are better informed then the teachers, which happens a lot with new technology. Does the teacher encourage the students to explore? Or do they lock down the lab?
In short, without the right kind of teacher environment the lab could start collecting dust.
During my senior year some boys played around with the computers in the lab and made them do creative / mischievous things – nothing serious, just being immature. After that, the lab was only open when a teacher could be there reviewing approved projects. Since she was not technically sophisticated or intellectually curious she drastically curtailed what could be done. Kind of hard to experiment around in an environment like that. The value of the lab fell dynamically.
I set up a computer lab in Uganda 3-4 years ago.
We bought second hand computers locally. They came with 256 MB of RAM and we upgraded them to 512. It was good enough to run Gimp and Firefox. That's what most of the internet cafes do too.
The computers were networked so we set up apt-cacher on the teacher's computer. The other software tip is that you will want to be able to block high traffic websites because internet access is so expensive and bad.
One thing which you might want to think about is if you'll have to pay tax bringing computers into the country. Uganda allows computers to be imported duty free. But for a while Uganda started banning people from bringing used computers into the country.
My sister-in-law's NGO is setting up a computer lab as well. They are bringing laptops from the US. Laptops are good because they have a battery built in so power fluctuations aren't such a big deal. The problem with laptops is that they can be stolen easily.
If you're bringing stuff from the US then bring a bunch of cheap USB keys for the kids. They will be very expensive locally.
Familiar like my job all the time, every day. It sounds like you are doing something similar to what I have done with the high school I work for in schools and orphanages in our local state, Nicaragua, Mexico, South Africa and (soonlike in a few weeks) the Caribbean.
While it’s impossible to speak to any specific instance, here are a few things to keep in mind:
1. Power: Do you have electricity etc that can support the lab? If not, is there a way to get it with little cost or no cost. Is the local power in stable enough to prevent long term damage to the units? That is, are there rolling blackouts or the like that could call for preventative measures to be taken?
2. Location: This isn’t just secure from people wanting to break in and steal stuff (which can happen depending on how desperate people are. But then if the community is heavily involved, it will help prevent this or at bare minimum spark response to it). This is secure from rain, wind, sun etc? Consider the elements, they are the enemy. Is the climate humid? Hot? Does the building flood? Does it have AC in any of the rooms? Consider what implications this could have to the long term use of the units and their placement.
3. Involve the school: What do THEY want? Assume nothing. Talk with the school, see what it is they want and need. Work for them. I’m assuming you have already done this but it’s always good to check. Long ago when we were first starting to do tech outreach, one of our leaders said something to the effect of “don’t try to change the culture. Give the tools, show them how they run and let them use them how they will.” She has been dead right everywhere we’ve worked and the only exception we’ve made to this rule is demanding that at least half of the training class be female and be made mostly of students. Involve the students. They aren't idiots, don't cripple their options. Yes, give them admin power. This approach gives the school’s members a stake in the lab and learning and will also help secure the structure. It can also be helpful if you can find local organizations to participate.
4. Support: who is going to be in charge of keeping the lab up? You can’t stay there forever. Find someone (or some ones) at the school who has raw potential or somehow has an edge above the rest in the tech department and train them on how to support the units. Don’t rule out students. Teach them the fine art of “frankenstiening” old units together if parts are not readily available. Software support is vital too. Consider a base image for all of the units. Something we’ve done in the past with some of our support contacts is give them a flash drive full of resources, installers, tech tools, guides, walkthroughs, Linux isos. you know, helpful things if you don’t have a stable internet connection. Having someone on site with a greater than average knowledge of the inner workings of the lab will also help prevent the lab from becoming a china doll, so to speak. You want things to break in the lab, you want it to be a rag doll. That means it’s being used.
5. Connectivity: Do you somehow have internet there? If not, would you consider doing an internal network? If you want internet, talk to local businesses, government or charitable organizations. You might find someone willing to run you a line or point to point connections. On a similar note, mirrored sites are awesome. A local server with all of Wikipedia, Kahn Academy, a pile of common domain classic books and more sites/resources with license agreements allowing such is a damn powerful tool to leave in a school’s lab.
6. Shipping/transport: You have any idea how you’re going to get the computers into the country? I’ve flown with switches and servers in my bag before and it is doable if not a bit back breaking and a little sketch in the eyes of customs. However, if you can find a secure and inexpensive means to ship them is always g
For this project I would suggest using the computers as a tool rather than a subject. What skills are most needed in the area? Is it a rural area? Maybe using the computers and the internet for weather info, commodity prices, and managing a farm would be a good use for them?
Depending on the area the students might not have the opportunity to go past high school. Who knows maybe one of the students will write a great farm management system because they doesn't like the one they are using or thinks he or she can do better. Way too often techies become so involved with the tools they forget that they are tools. Give them access to the universe but teach them what is needed there.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.