Innovation from Constraint (the extended dance mix)

I gave a talk a few weeks ago in Barcelona that was pretty well-received and widely blogged. Specifically, several bloggers have picked up and amplified the seven key points I offered in the talk – a list of principles that I believe characterize much of the best innovation coming out of Africa. I was deeply flattered that the blogger behind Design in Africa put these principles into dialog with advice offered by a couple of my heroes: Paul Polak and Amy Smith. And given the interest in the ideas expressed in those points, I thought I’d take the time (a flight from Dubai to JFK…) to offer a more thorough picture of the talk.

(This is probably a mistake. My guess is that the reason the seven points post has been widely circulated is that it was only a couple of hundred words, as opposed to the proxility I generally offer in this space. Turning something pithy and digestible into a Russian novel is probably not the best way to increase the currency of my ideas, but hey, evidently that’s how I roll.)

My friends in Barcelona asked me to address the question of innovation. How should social change organizations innovate? I realized that I didn’t have a lot of great examples of innovation initiated by social change organizations… but I had lots of great stories about ambitious and smart Africans innovating. This train of thought led me to think about approaches to innovation and, more or less directly, to how I failed out of graduate school.

I didn’t fail out, per se. I dropped out. But I dropped out because it was very clear that the graduate school I was in – the Integrated Electronic Arts program at Rensselaer Polytechnic Insitute – wasn’t a very healthy place for me to be. For one thing, the school was housed in the sub-basement of a communications building. There wasn’t any natural light in any of the spaces, and it had become something of a tradition for students to design “virtual windows”, usually computer monitors showing live or recorded footage of an outdoor scene designed to make the underground space less imprisoning and oppressive. (iEAR is no longer in a sub-basement and I understand it’s a much better program now, so please don’t read my failure to thrive as a comment on the current program.)

My real problem with iEAR was the problem of the blank canvas. I made some cool art as an undergraduate by playing with garbage. I somehow persuaded the environmental studies department at Williams College to give me a grant to spend the summer of 1992 wandering the garbage dumps, scrapyards and railroad tracks of northwestern Massachusetts and turning the junk I found into musical instruments. Then I’d pull out a four-track recorder and make strange, clanking music with said instruments and try to persuade my friends who were studying modern dance to build dances and perform on stage with me. It worked surprisingly well, well enough to get me an audition tape that first got me a scholarship to study in Ghana and later a scholarship to graduate school. Don’t knock junk art until you try it.

The great thing about making art out of junk is that it tells you what it wants to be. The decorative metal collar from under the glass bulb on a lightpost? Well, it’s too heavy and irregular to ring when you hit it, so it’s not a bell. But it’s shaped a bit like a dumbek and has a nice resonant cavity. Obviously, it’s a drum. All it needs is a skin made from the plastic top of a coffee can and some kite string to tension the membrane.

The tools we used at iEAR didn’t tell me what to do. Instead, they told me that I could do anything. You can put anything you want in that blank Photoshop window, limited only by your ability to draw, photograph or render it. Some of my colleagues thrived in this environment – my dear friend Daniel Beck used these tools to enhance the beautiful work he was already doing, making dark and beautiful films with a balky movie camera and a rusted tripod. Most of us didn’t fare so well, making art that was either absurdly overambitious or merely self-indulgent. I hated the stuff I was making… which rapidly turned into hating myself… which made me uniquely vulnerable to taking the first job offer that would get me out of the damned sub-basement.

(The job turned out to be to help co-found Tripod.com, which turned out to be a good thing indeed.)

Here’s what I learned by failing out of art school: constraints are good. They force you to be creative. If you’ve got a vast supply of precious hardwoods and carefully crafted musical hardware, you’ll probably spend a year failing to build a beautiful mandolin. (Trust me, I’ve done that, too.) But given some rusty wire, a couple of sheets of plywood and some thin sheets of plastic, you can make a jammin’ junk balalaika over the course of a long weekend. It won’t be pretty, it won’t look or work 1exactly the way you want it to… but the fact that you’re working within sharp constraints will force you into some creative solutions. (2″ PVC sawed lengthwise as a neck? Yep, that’ll work.)

It turns out that great artists choose to constrain themselves all the time. Some of Picasso’s most moving works were made in his blue period, when he constrained himself – consciously or otherwise – to a limited, stark color palette. While I love buildings where Antonin Gaudi used a bricolege of colored tile, his most moving building, Sagrada Familia, shows the architect constrained to simple, smooth white shapes. And it wasn’t until I bothered reading up on Joan Miro that I realized that he was a phenomenally technical painter before he decided to constrain himself to expressive, colorful, childlike compositions that look, at first glance, like doodles. (Three artists with a Barcelona connection. That’s what we public speakers do – pander to the crowd.)

Innovation comes from constraint. And most of us aren’t smart enough to know what to do with a blank canvas.

This helps explain why there’s so much gorgeous African innovation for friends like Erik Hersman and Juliana Rotich to feature on Afrigadget. The soccer ball, pictured above, summarizes this neatly for me. If you’re a soccer-mad kid in Kibera, Narobi, you may not have a ball to play with. But there are lots of plastic bags littered around the neighborhood. Gather a few of them, tie them together with a scrap piece of rope and you’ve got a very cool soccer ball. Leave a tail on the rope and you can train yourself to be a better ball-handler, hacking the ball on your foot while keeping hold of the string, and tucking the string in to play on the field. Your product, in some ways, is even better than the one you’d buy if you had enough money to buy a soccer ball. And if you lose it or it gets stolen, you can build another one with a modest investment of time and resources.

You’ve heard the expression “If all you have is a hammer, everything looks like a nail”. It’s a compact way of making the point that you should use the right tool for a job, that you should make sure that you’re considering a question in the proper light before offering an answer. It’s a good expression.

But what about situations where you really only have a hammer. Does that mean you should ignore all opportunities that would call for pliers, or a philips screwdriver? Should you constrain yourself to living in the world of nails?

Not actually an African hammer. It’s a metaphor, people. Run with it.

The African hacker’s approach to this is to find a friendly blacksmith and hack your hammer. A little welding and you can put a prybar on there, maybe a compact axe. File down one of those prybar flanges and it’ll make a nice screwdriver as well. You’ve still got a hammer, but it’s a multifunctional hammer, an innovative hammer.

This, by the way, is how lots of innovation occurs in the real world. Eric Von Hippel at MIT’s Sloan School has written extensively about user-driven innovation. “Lead users” push the limits of what tools can do, and adapt them to solve the problems they’re facing. Companies that learn from these lead users can change their research and development cycle, building products that solve the problems their users actually face. Anyone who is interested in lead user theory could learn a lot from hanging out with African hackers.

So let’s try a challenge based on innovating from constraint. Let’s design a cooling system for market sellers to use to keep their vegetables cold. This is an important task if you’re a small-scale farmer in a hot country – as soon as you pick your crops, they start wilting and become less valuable by the minute. If you can keep your tomatoes, cabbages and carrots cold, you’ll sell more for more money, and you’ve got a better chance of feeding your kids and sending them to school.

Lots of people try to solve this problem by looking at inexpensive electric refrigerators. After all, dorm-sized refrigerators are already pretty cheap – maybe we could scale one down, remove some parts and make an “appropriate technology” developing world version… coincidentally opening up a whole new market for Maytag or Haier.

But that’s a poor solution to the problem. Even if you can reduce the cost from $100 to $30, it’s still way to expensive for the market you’re trying to serve. Plus, your market sellers don’t have electricity either at home or at work, so you need a generator – expensive – and diesel – expensive. And even if you can line up the generator, the diesel, the fridge, none of these things are made locally. If they break, you’re shipping in parts from overseas and asking local mechanics to repair technology they’ve not often worked with.

Our constraints: no electricity, local materials, built and maintained locally, with a price point under $5.

Here’s how you do it. (Here’s how you do it if you’re an extremely creative and innovative Nigerian engineer.) Make two clay pots, one smaller than the other so it can fit inside it and leave a gap. Fill the gap between the pots with sand. Thoroughly wet the sand. Cover the top of your apparatus with a wet cloth. A couple of times a day, wet the sand and the cloth.

It won’t turn the mountains on your can of Coors Light blue, but it will keep your vegetables below 20C even if it’s 45C outside. In fact, the zeer pot works better the hotter and drier the day is. It uses the principle of evaporative cooling. As the water evaporates, the more energized, fast-moving molecules evaporate first, leaving the cooler, less-energetic ones behind. Your molecules left behind are less energetic on average, which is to say, cooler… and the vegetables inside that inner pot will stay cooler too.

(Before you try this at home, remember that this works in places that are hot and dry. If you’ve got 80% humidity, there’s going to be lots less evaporation than in a town in the Sahara, and this isn’t going to work nearly as well. Also, don’t try it with glazed pots – it works lots better if the clay jars are porous.)

The principles of evaporative cooling are well understood in the developing world, but Nigerian engineer Mohammed Bah Abba did a wonderful job of turning a physical phenomenon into a product that local artisans could build and local merchants would use. The pots are in wide use in northern Nigeria and also in southern Sudan. They cost roughly $2 for a set, and the market women who use them report that they make 25-30% more money from selling their vegetables.

The takeaway from the zeer pot story is the same as from the soccer ball – what you’ve got is more important than what you lack. If the problem of cooling vegetables turns into the absence of electricity to power your refrigerator, you’re going to solve the problem badly. But if youve got sun, sand, clay and some creativity, you might have a brilliant solution.

If we can use the sun to cool things down, surely we can use it to heat things up, yes? There’s an amazing wealth of designs for solar ovens available on the web. You can find plans for ones that look like umbrellas, ones that are unassuming glass-topped, foil lined boxes, ones made with tin cans as cooking chambers. Given the level of innovation and creativity surrounding these ovens – and the power of the sun in Africa – you’d assume that everyone would have adopted solar cooking.

Nope. Actually, solar ovens are a tough sell, even in places where firewood is extremely scarce. In refugee camps in Darfur, firewood is literally a matter of life and death. Men who go off to seek firewood risk being killed by fighters from other of other factions. Women risk getting raped. So families send young girls off, hoping they’ll be too young to be raped, or young boys, hoping they’re too young to be kidnapped and recruited as soliders.

Surely these folks would like a solar oven, yes? That’s the logic behind several projects designed to put solar ovens in refugee camps. But friends of mine who work in those camps aren’t especially enthusiastic about their prospects for success. You see, culture matters.

The woman cooking in the picture above is my friend Rose. She cooked for the geeks who lived in Geekhalla, the office and house we maintained as Geekcorps Ghana HQ. It was a really nice house, with electric power and a fully-equipped kitchen with a four-burner gas stove. So why’s Rose cooking over charcoal? Well, that’s how you make banku. It requires a particular kind of low, slow cooking with a great deal of stirring to get the consistency right, and Rose found that she could only make the banku she liked to eat and serve by cooking over charcoal in the backyard.

Stirring is the one thing you can’t do in a solar oven – as soon as you open the container where the food is cooking, you lose a great deal of heat. And stirring radically changes how your food cooks. I discovered this when a friend gave Rachel and me a crockpot. We eat a lot of stews in the winter, meat and potato creations that we cook on the stove on a lazy Sunday and eat for the following week. Our friend thought we’d enjoy a crockpot – we could put the ingredients in early on a workday, leave it alone all day and enjoy a stew in the evening. We tried it once and got a stew… that wasn’t stew. Everything was cooked through, but the starch from the potatoes hadn’t thickened the broth – basically, we got hot broth with cooked meat and vegetables in it. We hadn’t stirred, and it made all the difference. (We salvaged the stew by cooking it on the stove for about an hour, stirring vigorously… :-)

If people cook by stirring their food, they’re going to be highly resistant to cooking in a solar oven – the food won’t look or taste right. You’d think this would be the least of the concerns of people living in a refugee camp and risking rape to gather firewood. And you’d be wrong. My friends who’ve been working the Darfur camps report much more success showing people how to build more efficient three-stone fires, using three pieces of wood, and pushing them slowly into the fire, burning only the ends of the sticks. People are also having more success designing fuel efficient jikos, metal stoves usually made from scrap metal. A project at UC Berkeley has created a jiko that’s three times as efficient as a three-stick fire, but at $30 per stove, it’s still too expensive.

Don’t fight culture. Or, if you do, be cognizant of the fact you are, and that you may well lose.

The fine folks at Wildlife Direct, one the projects that focuses on bettering life for people and wildlife in Virunga National Park in eastern Democratic Republic of Congo, turned their efforts to the problem of charcoal. This is one of the most pressing problems facing the developing world. Charcoal is an environmental nightmare. You get it by cutting down living trees, digging pits and partially burning them. Then you take the resulting coal and put it into sacks and sell it to women, who bring it home to cook with. They often cook inside, creating fumes and smoke that damages their children’s health.

Basically, the only good thing you can say about charcoal is that it’s cheap and people like to use it. Except that it’s not cheap anymore. As the authorities in the DRC try to prevent destruction of gorilla habitat, they’re protecting the forests from logging. This means lots less wood to turn into charcoal, which means rising prices.

The folks at Wildlife Direct tried to do the right thing. They knew that plant oils – extracted from local crops like palm nuts – could be used as cooking fuel. So they worked with a pair of German manufacturers to design a very beautiful stove, optimized for plant oils. It didn’t work. It’s not clear why – the web account of the project shows a great deal of enthusiasm for the stove… then moves on to another technology in a couple more weeks, not mentioning the pretty German stoves. My guess is that a) they didn’t work, b) they worked, but they broke and were hard to fix or c) they worked, but it was hard to get a supply of plant oil. One way or another, they violate the principle of locally made, local materials, local repairs.

The technology the Wildlife Direct folks moved onto is biomass charcoal. Dr. Amy Smith helped convince me a few years back that biomass charcoal is pretty much the coolest new technology being pioneered in the developing world. To make biomass charcoal, you take agricultural waste (the stalks from maize plants for instance), squeeze the water out of it and form the resulting bits into briquettes. You dry them out, and burn them as you’d burn charcoal. It takes serious experimentation to get it right, but the results are pretty amazing.

You end up with “charcoal” that’s produced as a fast-renewing resource, that repurposes agricultural waste and that, remarkably, burns cleaner than traditional charcoal.

There’s another great thing about biomass charcoal – it creates jobs. It takes a lot of people to work the press to produce this stuff, to prepare the “mix” for pressing, to dry the finished briquettes. But you can work within existing market mechanisms – people who make their living selling charcoal can make their living selling biomass charcoal, and people who made their living illegally logging and creating charcoal can create biomass charcoal. And the wooden presses can be made and repaired locally.

Two more lessons here. First, embrace market mechanisms. If people are making money making charcoal, they’re going to be deeply unhappy if you put them out of work… possibly unhappy enough to point an AK-47 at you, as happened when Virunga rangers tried to put charcoal dealers out of work. It’s much better to figure out how to leverage the mechanisms that already exist and put them to work to achieve your goals.

Similarly, there’s a lot to be said for learning how to innovate on existing platforms. People are already familiar with charcoal. They’ve got stoves that burn it, and they know how to cook with it. It’s lots easier to give them better charcoal than to convince them to cook differently. Charcoal is a very promising platform because it’s well understood and widespread – if you can make better charcoal, you’ve got a good chance of making a major, widespread change.

The platform that’s seen perhaps the most innovation in Africa is the bicycle. With a little bit of hacking, you can turn a bicycle into a cargo vehicle, loading it down with frightening quantities of bananas. Add a wheeled cart behind and a bicycle is an ambulance. Add a tiny one-stroke engine and it’s a motorbike, capable of propelling the rider over long distances… and she can start pedaling when she runs out of gas. In Uganda, bicycles become phone booths, with wireless phones attached to metal boxes mounted on the handlebars.

Because bikes are so common in Africa, they’re very well understood. Because they’re well understood, they’re hackable. A technology that’s precious, strange and scarce won’t get hacked – everyone will be afraid of breaking it. Tech that’s common, repairable and replaceable will.

Bikes get really interesting when you let them change function. In my single favorite video on Afrigadget, Peter Kahugu shows off his knife-sharpening bicycle in Nairobi. I love the video because I owned the same damned bicycle when I lived in Ghana in 1993. It’s a heavy, Indian-made beast, but it’s indestructable, and has some cool features. One useful feature is a huge kickstand that surrounds the back wheel. This lets you park the bike with the rear wheel off the ground… which allows you to ride the bike and spin the rear wheel without the bike going anywhere.

That’s the key to Peter’s business, as he uses the power of that back wheel to turn a belt which turns a grindstone mounted on the handlebar… which lets Peter sharpen knives and scissors, a trade that pays him a solid wage, roughly $10 a day. His business is entirely portable, which lets him bring his services to his clients. And he can take his tools home at the end of the day. The bicycle is still a transport tool, but it’s also a power generation tool.

William Kamkwamba has never met Peter Kahugu, but they’re spiritual kin. William got interested in electrical power living in a Malawian village that’s disconnected from the electric grid. One of the few ways to generate power is to use a bicycle dynamo, a simple tool that captures the motion of a spinning bicycle tire and uses it to rotate a coil of wire within a magnet. The current that results can be used to power a light for the bicycle… or, really, anything else, if you can just figure out how to keep pedaling the bike.

So William attached windmill blades to the sprocket of a bicycle and put the assembly on top of a tower made of wood. The wind turns the sprocket, the chain turns the rear wheel, which turns the dynamo, which produces sufficient power to light his parents’ house with Christmas lights and power two radios.

(That William did all this with a grade school education is part of the story. Or that he had to manufacture all the switches and acoutrements himself, usually by heating and shaping PVC pipe. Or that his windmill got written up in the Malawi Times… which got him written up in some Malawian blogs… which got him into Afrigadget… which helped get him to the TED Africa conference… which got him written up in the Wall Street Journal… all of which means that well-wishers around the world have helped raise enough money to send William to the best highschool in South Africa and, in the long run, to a top engineering university. But that’s another story.)

The mobile phone is the new bicycle. They’re amazingly pervasive in sub-Saharan Africa – there’s at least 100 million handsets in the region, and people in very rural areas are often able to access phones, even if they’re borrowing them from friends or renting time on them from a local entrepreneur. Because they’re pervasive and well understood, they’re hackable. And they’re changing function. So while there are lots of cool stories about peole using phones to get better prices for their fish or to share information on market prices for maize, the stuff that flips my wig is when people use phones to replace cash.

Until you’ve lived in an all-cash society, you have no idea what a problem cash is. The guy in the photo above is Stophe Landis, who was our program director for Geekcorps in Ghana. This is not the photo of a drug deal going down – Stophe was simply trying to give each of our volunteers a weekly stipend of about $50. At that point, the largest bill available in Ghana was worth a bit less than a dollar. So paying the stipends involved handling large bricks of cash.

Imagine buying a car in this kind of cash society. I don’t have to – I acted as “bagman” for a friend years ago, carrying two huge shopping bags of bills so she could pay $3500 for a used Corolla. When you don’t have checks or credit cards, you pay cash for everything. And when inflation devalues currency quickly, you carry huge stacks of bills.

The phone’s changing all this. To send money from the city to a family member in the village – say to Auntie Grace in Dzolo-Gbogame – you traditionally took a stack of bills, handed them to a taxi driver who was travelling to the region and said, “Hey, give these to my auntie, will you?” It worked better than you’d think… but not all that well. It’s much safer to travel with the cash… but that gets pretty time consuming if you’re travelling home every time you’re trying to send money to your family.

It’s easier today. You buy a couple of phonecards. You call the woman in Dzolo-Gbogame who owns a phone. She rents the phone to other villagers, collecting fees in cash and using it to pay for mobile phone minutes – it’s a good living. You tell her, “I’ve got fifty dollars in mobile phone credit here. I’m going to read the numbers off to you so you can top off your phone. Once you verify that the numbers are real, go give $48 to Auntie Grace and tell her to call me to tell me she got the money.”

That’s the system called “sente” in Uganda, and known by different names in different corners of the continent. It relies on the fact that prepaid mobile phone cards have the remarkable property of turning money into information – your $50 turns into a series of numbers, which can be turned into $50 worth of value at a later point. Information is lots easier to move than money, so you can get money to Auntie Grace without getting into a cab or trusting the cab driver. (You do have to trust the phone lady, but the fact that Auntie Grace can confirm the transfer goes a long way to making that a more secure transaction.) The system works so well that creative phone companies are formalizing the system – M-PESA, the celebrated e-cash system that’s become so popular in Kenya is really just a formalized form of sente.

If this seems like a surprising use for mobile phones, it is. But it makes intuitive sense to people who’ve lived and worked in the developing world. Serious problems, like the difficulty of living in a cash only society, aren’t always obvious from afar.

Another function change phones are taking on is becoming location-sensitive tools. In the same way that my phone talks to GSM towers to give me a rough idea of where I am, allowing me to navigate darkest Queens via my iPhone, Kenyan hackers are figuring out how to turn GSM phones into elephant tracking systems. It’s a good idea to know whether elephants are enroute to your farm as one elephant can eat a year’s crops in a single evening. If you know that elephants are on the way, you can stand in your fields with torches and chase the animals off.

What you need is for the elephants to let you know that they’re coming. SMS works well for this. Elephants are fitted with GSM-powered collars. These collars are aware of “virtual fences” – when the animal crosses into areas where it’s likely to encounter humans, the collar sends a message to a SMS gateway, which forwards a warning to villages who have mobile phones. Because villagers can effectively defend their crops, they’re less likely to attack and kill the elephants.

SMS is an effective tool for monitoring all sorts of large, dangerous mammals. You can make the argument that Morgan Tsvanagarai was able to challenge the first round of Zimbabwe’s presidential elections in no small part due to SMS. A change in polling law meant that every local polling station in Zimbabwe was required to post local voting results publicly. Zimbabwe’s opposition party, MDC, organized an effort to collect these results via SMS. As a result, the MDC knew, within a few hours after the close of polls, that they’d received more votes than ZANU-PF.

When private companies invested in building mobile phone towers, it’s unlikely that they were thinking about eliminating voter fraud, tracking elephants or even moving sums of money around. But infrastructure begets infrastructure. Once you’ve got an infrastructure that lets lots of people communicate, they’ll find ways to make it work as a currency system, for instance – they’ll build new infrastructures on top of it. (Think about how aspects of American car culture – drive in movies, diners, strip malls, big-box stores, just-in time inventory – have been built on top of the internet highway system.)

Here are the seven principles of innovation from constraint that I shared with my audience in Barcelona:

– innovation often comes from constraint
– don’t fight culture
– embrace market mechanisms
– innovate on existing platforms
– realize that problems aren’t obvious from afar
– understand that what you have is more important than what you lack
– build infrastructure on infrastructure

I’d add two other observations that I’ve realized in giving this talk more recently:

– objects need to become familiar and pervasive, then they become hackable
– the really amazing innovation happens when objects change function

I closed the Barcelona talk looking at how these principles were followed or ignored on three international development ICT projects: One Laptop Per Child, Kiva.org and Global Voices, a project I’ve helped lead. I’m not confident that I got those evaluations right. I suggested that OLPC was suspect because it definitively ignores market mechanisms and largely fails to build on what’s already present. You can argue that it innovates on existing platforms, using open source software, or argue that it designs its own hardware and operating environment. But questions like whether the project ignores or embraces culture get very difficult to evaluate, leaving me wondering if this is really an effective method of testing the success or failure of projects, or whether it’s more useful as a way for innovators to think about their work as they’re executing projects.

I suspect that the seven or nine “laws” I’ve offered here are likely augmented by dozens of other useful observations from far more experienced students of developing world innovation like Dr. Smith and Paul Polak. I’m not sure that introducing another seven or nine “laws” is useful, except inasumuch as it lets us draw some generalizations about innovation from constraint.

Making these generalizations forces us to understand that astounding stories like William Kamkwamba’s windmill aren’t isolated cases. They’re everyday miracles, the routine creativity that allows people with limited resources to live, thrive and survive in difficult environments. This sort of innovation is hard for people in developed nations to understand intuitively precisely because we’re free of the constraints that characterize life in the developing world.

Innovators who want to help people in the developing world need to find ways to understand constraint and the creativity that can come from constraints. This is why Dr. Smith asks her students to spend a week living on $2 a day in Cambridge, MA to understand what a serious constraint income is. It’s why I focused on bringing geeks to Africa to live for months at a time, hoping that proximity to these constraints would help them help counterparts working within constraints. It’s why anyone who wants to write software to help poor people needs to step away from the blank screen and into the world of constraint before writing a line of code. Your constraints are your best friend as an innovator, your best chance of creating something with lasting utility.