O'Reilly    
 Published on O'Reilly (http://oreilly.com/)
 See this if you're having trouble printing code examples


World

Geeking in the Third World

by Richard Koman
05/12/2003

In 1993, Ethan Zuckerman was a Fulbright scholar working in the West African nation of Ghana. He was amazed that virtually no one in the country had Internet access: "There was, in the whole country, I think, one person who was regularly online, and he was getting online by dialing AOL in the U.K., and checking email once a week," Ethan says. After the Fulbright, Ethan returned to the U.S. and became tech employee No. 1 at Tripod, which was later bought by Lycos. In 1999, at the height of the dot-com bubble, he looked around to find himself far wealthier than he ever imagined he would be and found himself thinking about Africa again.

Ethan Zuckerman, with elephants, in Africa

(click for larger image)

Returning to Ghana, he was "thrilled to discover there was a nascent Internet revolution going on, and the main thing holding it back was the absence of experienced IT talent within the country. Anyone who wanted to set up, say, an e-commerce business was going to hit some technical hurdles pretty quickly." Thus was born the idea for Geekcorps, a volunteer organization dedicated to helping developing nations meet their IT needs. By working with companies in developing nations, First World geeks can help countries create sustainable IT cultures. In 2001, Geekcorps joined forces with the International Executive Service Corps and now operates as an independent division of that 38-year old nonprofit.

Zuckerman will be one of the participants in O'Reilly's Geek Activism Summit at OSCON (the O'Reilly Open Source Convention) this July, in Portland, Oregon. The summit is by invitation only, but a conference session and public BOF dedicated to the topic will be scheduled. Richard Koman recently talked to Ethan about Geekcorps and its efforts to help create an IT culture in Ghana.

Richard Koman: So you found this nascent Internet industry in Ghana in 1999. What was driving the Ghanaians to get online?

Ethan Zuckerman: There were two very real reasons for Net connectivity in Ghana. One was communication with the diaspora. So many Ghanaians live in Europe and the U.S. that email is a very effective way of bridging that gap. The other thing was the notion that there could be a market for Ghanian goods and services worldwide, and that market was going to be a lot more reachable online than it would be from any other medium.

But it was a very weird time because you'd find a cyber-cafe and there would be computers and staff but no electrical power, or computers and power but no telephone lines, or everything you needed but no one to plug things in and make them all work together. And across the board I felt you had an abundance of entrepreneurs who were willing to try things but they had a real lack of skill sets. So that was the problem I was interested in: Could we find a way to do skill transfers between people in the IT industries in the U.S. and Ghana?

Obviously, the project expanded from there. While Ghana continues to be a flagship presence for us, we also have a large presence in Mongolia. We have smaller programs in Rwanda and Jordan, and we're doing some work in Armenia and Bulgaria. At this point we work in a dozen nations in total.

Koman: How did you wind up in those countries?

Zuckerman: There are two answers to that question and they focus on different aspects of the problem. One answer is that you're looking for countries that are already developing something of an IT industry, whether that's a domestically focused industry that will help deal with local needs, or whether it's a regional player, or whether it's someone who looks like they have international ambitions. You're looking for evidence that someone has started IT businesses, that people are succeeding and running IT businesses, so that there's something to work with. We don't start businesses on our own. We basically work with people to build up their businesses. We look to see who has a lot of ISPs, who's got loose ISP licensing laws, who's doing regional software development. We get very interested in that. That's one answer.

The other answer is that, because we are an NGO (non-governmental organization) that doesn't have a core endowment, we work to find sponsors who support our work. In many countries, we work where we can get USAID (U.S. Agency for International Development) support; that's been our best and most consistent sponsor. In some cases, we are able to use money from foundations or from private individuals to open up a presence and then look for ongoing sponsorships from USAID or the U.N., or something along those lines. And that's what happened with Ghana. We ran that program for about 18 months without a single large sponsor. And then we got to the point where the program was so successful that we were able to find sponsorship for it.

Koman: So, a lot, if not all, of your funding is coming from USAID?

Zuckerman: At the moment, it's about 90 percent. And that's a big shift. When we started, it was largely privately funded. Raising money from the general public, raising money from tech corporations, specifically raising money from foundations, has been extraordinarily difficult. This year, in particular, we are finding that a lot of foundations are deciding not to give. They simply lost so much money in the market that they are taking a year or two off from grants or new grants in the hope of allowing their assets to recover. In general, raising money for international development is one of the hardest things you can do. If you look at how Americans give to charities, we tend to give locally. And locally doesn't always mean geographically local. Locally tends to mean socially local. So we give to our churches, our communities, our alma mater. But we tend not to be nearly as good about giving towards issues that we don't have a personal connection to, which makes perfect sense. But it means that raising money for economic development for developing nations is a tough thing to do. Which is why so many groups like ours get a lot of support from bilateral or multilateral aid agencies.

Koman: Talk about some of your accomplishments overseas. What benefits have you realized in building IT infrastructure in these countries?

Zuckerman: In many cases, it's less about IT infrastructure than strengthening IT businesses. Just to be very clear, we are not going over with duffel bags filled with 10BaseT. We've stayed very far away from the hardware side of things. We are trying to help build up companies that are already competent at what they do. At this point, we've already completed about 100 projects using over 60 volunteers. Many of them end up working on two or three projects, serially or in parallel. A couple of examples that give a sense of the variety of things we do:

We worked with a company in Ghana called IDN, which is the third-largest ISP in the country. IDN largely sells its services to cyber-cafes and fairly large businesses. Their challenge is to get a decent amount of connectivity to another facility. And this is tricky because the Ghanian phone system is pretty bad. So what these guys figured out they should do is skip the phone system entirely and use Wi-Fi, and essentially throw up a parabolic antenna to create a signal that way. And you can quite comfortably do good data transfer at 10 kilometers that way. The problem, it turns out, is that the equipment on the client side is pretty expensive. IDN's clients had to buy a bunch of things, a server, a dish, and so forth, and they ended up roping all the pieces together. The problem is that with all the import duties and such, the cost is about $4,000, which is really outside the range for most of their clients.

Hacking in the Third World

(click for larger image)

IDN looked at this problem and said, "Jeez, we may need our own piece of hardware." They contacted us to see if we had a very accomplished networking and Linux hacker, which we did. He worked with them to design a unique product, which they are now selling, called the Sawa series of routers. The Sawa box is just a customized Linux box. It grabs a Wi-Fi signal. It provides IPs for everyone in the cyber-cafe. It's got a local email server. It has a good firewall. Also, it's in a single box, and its being manufactured and distributed in Ghana. Since coming up with that product, and because of our help in implementing it, IDN's sales have gone up more than 50 percent. The cost to a client has gone down by a factor of five. Here's a product that has real implications for how you provide connectivity in the developing world, and we had a real hand in putting it together. So we're pretty psyched about that.

Another example, which gives a good sense on how we work on governance projects, is a project we did in Rwanda with the Department of Justice, which had to do with the genocide mediation process, the Gacaca process. Gacaca is Rwanda's way of trying to hear the more than 100,000 court cases that are a result of the 1994 genocide. There are an enormous number of people in prison, more than 100,000, and a fairly desperate need for justice. A lot of work is being done on the legal side, making sure that there's a swift, efficient legal process that draws heavily from traditional Rwandan aspects of justice.

But there's also a really strong IT need. You need a court docket system for all of this, which we helped build. We got involved by helping the Department of Justice identify Rwandan firms, and then we sent a Canadian volunteer to go work with those firms for two months, helping them get a sense of how to work with very large databases in SQL. Not only did we complete the project, but that database firm is in a lot better shape to do more work for the Rwandan government or for others in the region. It's a nice dual impact. We completed the project but we've also transferred some very valuable skills at the same time.

Koman: So a big part of your focus is in helping to build an IT business ecology?

Zuckerman: That's a great way to phrase it. We often use the phase "digital independence." We want to help countries get to the point where they're self-sufficient with their own IT needs. My feeling is that almost every country in the world is going to have IT needs in the near future. Whether that's e-government--trying to make governmental systems more transparent--or whether it's integrating your economy into the global economy, there is an IT need within literally every economy.

One of the key challenges facing a lot of governments is, if you don't have the local talent to do this, and if you wind up importing that talent, it's very expensive and becomes a vicious cycle. You brought in people to build your system and then they break and you have to bring in more people to fix them. Unless you stop at some point in the process and say, "Wait a minute, this is too important to outsource, we really need to build this competency ourselves," you can wind up with what a friend of mine in Rwanda refers to as "poison-pill IT systems;" systems that people are locked into using but they have no way to maintain themselves. Digital independence is getting nations to a point where they can take on these needs. And being good, free-market capitalists, our belief for how to do this is that countries are going to need a group of competent IT companies that can take on this problem.

Koman: And building the IT ecology is a core part of long-term growth, GDP growth for the nation...

Zuckerman: I think that's absolutely true and I think that's true even for countries that don't have the stated aim of exporting IT services. I think what's become shockingly clear is that, to do business in the 21st century, you need to understand that a global business culture is an Internet-connected culture. And that's as simple as, if you're going to export to someone, you're going to have to tie into an Enterprise Resource Planning or Supply Chain Management system. Or here's something even simpler: If you're running a hotel in Ghana, people want to see your rooms on the Web, and they want to send email to you and get an answer. And that requires some sort of local IT ecology to support it.

Koman: What's the experience like for your volunteers? How long do they spend on assignments, what do they do, what is life like for them?

Zuckerman: We've looked for ways to support our volunteers but also to make sure that they're not coddled and that they get out and see the culture. We also make sure they're able to work relatively efficiently. In our Ghanian and Mongolian program, we deploy our volunteers in teams, and they go through about a week's worth of training together and generally live together. Those tours of duty are officially four months; a lot of people wind up extending them to six months.

And we make a point of paying volunteers quite badly. The reason for that is if you're somewhat fiscally constrained, it forces you to go out and have lunch with your colleagues and figure out where they're able to have lunch for 50 cents for a plate, rather than $5. It keeps volunteers from automatically becoming expats. In most of our programs, we pay about $400-$500 as a living allowance, and that has to cover food, transportation, and beer. We've discovered in Mongolia that we have to allow a clothing allowance because the winters are so harsh. We have to give people $100 so they can gear up.

The experience is very similar to working in a tech company in the U.S. and Europe, except that you're in a radically different business culture. In the bigger programs, we've structured it so that there's some group travel; there's a chance to get out into the countryside; get some other experiences. We are very aware that the reason people are doing this is to experience the country they're in, so we're careful to not work people to the bone, and we make sure that they get out and see things. On the other hand, we're really aware that when we have someone in a country, that's a pretty scarce resource and we really want to take advantage of their time in that country.

Koman: What's the personality profile of your volunteers? Are you getting a sort of uber-nerd or someone who takes a more workaday approach?

Zuckerman: We have more than 1,500 people in our database right now. I think the median age is about 28 or 29. We're generally dealing with someone who has 5-plus years of tech experience in the real world. These are people who are coming out of the corporate world. The database represents people from several dozen nations at this point. It tends to run about 60percent from the U.S., but we have a lot of Canadians, a lot of Europeans, increasingly South Americans. We've had some great luck with Chilean graphic designers.

IDN staff

(click for larger image)

I recruited the first group we took to Ghana the way I would have recruited geeks for Tripod. I looked for type-A workaholic supergeeks, and I discovered that type-A workaholic supergeeks have a really hard time with the developing world. After figuring this out, we started recruiting for flexibility and a sense of humor. We looked for people who we felt were a lot more likely to roll with the punches and were less gung-ho but more flexible, people who were able to look at a challenge and shrug it off or laugh it off.

Koman: So, you're looking at how someone deals with losing power and losing an hour of work.

Zuckerman: Yeah, scenarios like, 'I've been working with this one guy, teaching him C++, and it turns out that his family in the north of the country is sick, and so he's disappearing for a month. What do I do now?' And we're looking for people who say, 'Well, these other guys are cool, maybe I'll work with them for a couple of weeks.' What we're not looking for are people who are going to stomp their feet and say, "Goddammit, this program has to get finished!"

We're very lucky. We have a very good recruiter who was a Peace Corps volunteer and has a very good read on who's going to make it and who's not going to make it. We spend a lot of time on the phone, sometimes in person, certainly over instant messenger, before we say yea or nay. Because the database is so big, 1,500 people, and because we do no more than 50 assignments per year at this point, we can say no to people--a lot--and we do--a lot! As a result, we've never had an early termination, and that's pretty cool. Most overseas programs don't have that record.

Do you have what it takes to be a geek volunteer? Find out more at the Geekcorps web site and get involved today.

Richard Koman is a freelancer writer and editor based in Sonoma County, California. He works on SiliconValleyWatcher, ZDNet blogs, and is a regular contributor to the O'Reilly Network.


Return to www.oreilly.com

Copyright © 2009 O'Reilly Media, Inc.