Who Needs Infrastructure?

Last month I went to Haiti to help out with an IT project in Petit-Goâve, a medium-sized town about seventy kilometers west-southwest of Port-au-Prince, on the northern shore of the Tiburon Peninsula, opposite Île de la Gonâve on the Canal de Sud. The project’s objective is to create, or rather restore, a computer lab at “College” Harry Brakeman (actually a primary and secondary school, hereafter “CHB”), and provide greatly improved internet access, via wireless links, at five sites (including CHB) in Petit-Goâve owned by L’Eglise Methodiste d’Haiti (EMH). The epicenter of one of the larger aftershocks of the January 2010 earthquake was directly beneath Petit-Goâve.

Numerous ongoing projects for the EMH throughout Haiti are being funded by United Methodist Committee on Relief (UMCOR) and staffed by United Methodist Volunteers in Mission (UMVIM), but my personal involvement is not occurring as a result of direct involvement with any of those organizations. I have for many years been attending an informal Friday lunch group that for the past decade or so has included Clif Guy, who is the CIO of United Methodist Church of the Resurrection in Leawood, Kansas, generally known as “COR” throughout the Kansas City metropolitan area, in which it is by several measures the largest single church – big enough to have its own IT department (larger than most church staffs altogether) and a CIO.

In mid-January I returned from a solitary and somewhat monastic sojourn in New Mexico and the trans-Pecos region of Texas to 1) get back to work at Sprint; 2) bury my just-deceased 18-year-old cat; and 3) talk to Clif about opportunities in Haiti, which he had mentioned several times over the previous year. Two months of frantic preparation later, which included among many other tasks the filling out of a “Mission Trip Notification of Death” to specify the disposition of my corpse, I was landing at Toussaint Louverture International Airport.

Now, after all that buildup, I am not going to use this posting to describe my experiences in any great detail. I will merely mention some sensory impressions, and then go on to speculate about a highly unusual but nonetheless plausible development path for Haiti and other (to use the discreet expression) less-developed countries.

Visual: due to a combination of high population density, minimal road/street infrastructure, and few people wealthy enough to own personal motorized transport, linear density of both pedestrians and passengers on “tap-taps” and motorbikes is at least two orders of magnitude higher than almost anywhere in the US outside of a few very high-density urban areas. The impression, especially on someone from a much lower-density metro area like KC, is of impossibly large swarms of people in Port-au-Prince and continual busy-ness everywhere, even at the crack of dawn in rural areas.
Notwithstanding the horrendous deforestation one sees from the air, in Petit-Goâve itself there was sufficient vegetation to limit lines of sight to the order of 10–100 meters. I only got a few glimpses of the mountains (which rise to 1400 m only a few kilometers inland) or the ocean (except when I walked down to the beach one day).
There are many brightly painted storefronts and other buildings, although the pervasive dust, oddly similar to what I encounter on my visits to the trans-Pecos, obscures some of the coloration unless it is washed off.
The people are attractive due to a combination of low median age (early 20s), generally adequate nutrition, and nearly incessant hard physical labor. The women, particularly, are surprisingly well-dressed. The appearance of schoolchildren and people attending church is simply immaculate. Given that quite a few of them are still living in tents, it is utterly nonobvious how they pull this off.

Auditory: ubiquitous generators running somewhere; motor vehicles, mainly small motorcycles, continually honking their horns as much to communicate as to complain; roosters at all hours (continuous from 3 AM until after sunrise); occasionally other bird calls, though not many (possibly related to deforestation); being called “blanc” by curious locals; surprisingly non-obnoxious music from a sort of nightclub next door to where the team stayed the first night; gorgeous singing in church; lots of electric fans (which we used as a pink noise source to help us sleep – that, and earplugs!).

Olfactory: burning garbage (frequent); occasional whiff of marijuana; some barnyard odors depending on wind direction and proximity of animals; vaguely Cajun and generally quite good cooking; dust, again; vile stench in visibly more distressed areas, mainly from piles of rotting garbage in the streets – this was in Carrefour, a suburb of Port-au-Prince.

Tactile: dust, yet again; rough concrete surfaces everywhere due to lack of lumber for construction; daily temperature range year-round is low 20s to mid-30s Celsius.

Taste: yummy cane-sugar-sweetened soft drinks; locally grown coffee to die for.

The stereotypical take on Haiti is that it’s “depressing,” and it is certainly shocking in many ways, even more than a year after the earthquake. The only place that really kind of got to me was the above-mentioned Carrefour, which (rather portentously, I must say) means “crossroads.” But the built-up areas have electricity, albeit of low quality – it was routinely interrupted, and we measured voltages in the 90s – and “cyber cafés” are common, though again the quality of service is wildly unacceptable by global standards; we measured packet losses well above 10% on many occasions, whereas 1% is considered bad in the US.

(Electricity in Haiti, incidentally, is North American: 220V 3-phase coming in off the pole and 120V/60Hz in ordinary-looking outlets inside. I think this goes back to the 1915–34 occupation by the Marines, during which a fair amount of infrastructure got built, unfortunately with corvée labor. That particular unpleasantness has passed from living memory, and Americans are generally admired in Haiti today.)

And the physical environment is actually rather pleasant much of the time. The climate is that of a midland-American July, without the violent thunderstorms; high winds, lightning, and hail are practically unknown there except during hurricanes, which is to say for a couple of days every few years. The sea breezes in Petit-Goâve were delightful. The countryside is beautiful and the topography, at least from this Midwesterner’s viewpoint, spectacular.

Along with Internet access, mobile phones are common, and like many other third-world countries, the land-line network is minimal – only about 100,000 lines before the earthquake, certainly many fewer now, but mobiles are in the hands of nearly half the population (ie at least 40 wireless phones for every wireline one). The buildout of a land-line network to a couple of million additional households will never happen; it’s been technologically leapfrogged.

After a few days in country I began wondering if there would be other examples of centralized infrastructure being bypassed by the equivalent of mobile phones. Probably not an original idea, but anyway, here goes:

Individual household-generated electricity seems like the most obvious possibility. Insolation in Haiti is at least as great as that of the southwestern US, where a horizontal flat plate intercepts over five kilowatt-hours of solar energy per square meter per day on an annual basis. At that intensity, my electricity usage, which although less than ten kilowatt-hours per day is far greater than the average Haitian’s, could be easily met by ten square meters of 20%-efficient photovoltaics, assuming adequate storage technology. Most Haitian households could probably function on two or three square meters of PVs. It is overwhelmingly likely that over the next decade or two, the installed cost of such a system will drop so much faster than the cost of conventional power plants and power lines that most of Haiti will never have a conventional electric power grid.

Water purification could follow a similar path. What comes out of the tap is unsafe to drink; purified water is produced in small plants and delivered by truck. Haiti receives over 1,300 mm of rainfall each year, and individual usage would average well under 100 cubic meters per year. It is not difficult to imagine small rooftop storage tank/purification units obviating the need for water lines in most places, again becoming cheaper faster than conventional infrastructure ever could. Something analogous would treat wastewater.

So: no phone lines, electric lines, water lines … no gas lines, either, especially if electricity is cheap enough – a combination of weak-to-nonexistent institutions and galloping technology producing a surprisingly physically unconnected population.

Now here’s where it gets really weird: what about roads?

The ones they’ve got are pretty bad, although not as bad as I expected. The very widest streets in the Port-au-Prince metropolitan area, population 3.5 million, are not quite as wide as a 4-lane street in the US (there are, perhaps, ten traffic lights in the entire city). Most of the relatively major arteries are the width of a typical residential street in the States. There are a handful of paved highways out in the country, all two-lane and occasionally buckled, whether by the earthquake or otherwise shifting foundations is not clear. Bridges are not always intact – we had to ford the Rivière de Grand Goâve. And this was along a heavily-travelled and populated route; in many rural areas nothing much bigger than a motorbike will pass.

Self-driving off-road vehicles reached an adequate state of development several years ago, thanks to the DARPA Grand Challenge. How inexpensive and reliable will they be in another 10–20 years? Once again, I perceive a race, one that will not be won by 20th-century methodologies in a country with endemic institutional weakness but full of energetic people. Most of the road network in Haiti may never rise above the level of dirt tracks. It may not need to.

8 thoughts on “Who Needs Infrastructure?”

  1. And in 20 years or less most third-world towns will probably have at least one shop with a second- or third-hand desktop fabricator, powered off the solar panel and downloading open-source software through a cheap satellite dish, and turning out all sorts of small objects and components. The centrally-manufactured equivalents, if they are still manufactured anywhere (probably in government-subsidized legacy factories) will never be delivered to little third-world towns.

  2. For some applications, it would probably be better to use thermal energy from the sun directly and skip the electricity phase. I’ve seen old washing machines, probably from the 1920s-1930s, intended for use on farms without electricity–they were directly powered by small gasoline engines. In a country with a lot of sun, you could take the same approach, with a small steam engine using a solar dish to boil the water.

  3. I’m glad you decided to write about your trip to Haiti.
    About autonomy and its advantages in certain niche circumstances: recently I read about this project, an architect daughter designing and supervising construction of a summer home for her retired father on an island off the coast of Maine. There is a very interesting part in the article, concerning water purification means and self-reliance on electricity, based on solar roof panels; note also, in comments, very low cost of construction. In condition of mainland (which a big island like Haiti must feel like, compared to the Ragged Island) the delivery/installation costs are lower, plus local prices on materials and expectations of lifestyle are different, the cost of building something similar as a mass-market flexible model for places like Haiti might be lowered more.
    Also, another food for thought: Swedish company that produces transparent glass roof tiles as means for HVAC system. Granted, they advertise it primarily as improved heating, and their market is, naturally, in the North, but every heating might be turned into conditioning…

  4. I don’t think photovoltaics will really kick off, even in the equatorial 3rd world. It really doesn’t matter how much sunlight they get or how efficient and cost effective the photovoltaics become. People need electricity when they need it and not just when nature can provide it. Unless you have a practical means of storing enough electricity to meet even 3rd world minimum standards for wattage and reliability, such systems will be non-starters. A lot of people have tried but they always fail for just that reason.

    I do think that solar powered water purification could be very practical because you can easily store the treated water. Recently, some Africans found out that just putting water into clear plastic bottles and leaving them out on a corrugated metal roof for a few hours nearly completely sanitized the water from the solar UV. It would be fairly easy to make a formal system using plastic tubing and perhaps a photovoltaic driven pump and ozone generator.

    I do agree that decentralization will be the trend in the 3rd world over the next few decades. People in the 3rd world are poor precisely because they are disorganized compared to the industrialized world. They just don’t have the cultural and political experience with reliable, large scale and long term organization required to operate an industrial tech base. Giving them the tools to raise their standard of living based on the smaller organizational units they do have would produce better results than trying to force evolve them through 500 years of Western cultural evolution in just a couple of decades.

  5. Somme technologies reduce the cost of chaotic societies and reduce the number of people needed to deliver some core capabilities such as phone service. But the ability to get by without phone landlines should be contrasted with the value data landlines. Fiber optics can transfer orders of magnitude more data. So while radio bandwidth has gone up fiber optic bandwidth has gone up far more.

    As for water: Try living on rain water in desert regions. Not going to work. Also, making all the water pure enough to drink doesn’t cost that much. The fact that Haiti can’t manage it does not speak well of Haiti.

    While I like technologies that reduce our dependence on centralized infrastructure the value of centralized infrastructure isn’t going down overall. There are economies of scale from centralized infrastructure that boost living standards. A nation without pure city water and without reliable electric power is a poorer nation.

    Even some of the technologies that reduce reliance on central infrastructures work better with central infrastructures. For example, solar PV costs less in big installations than it does on roofs. Also, concentrating solar power with hot salts for storage does not work in small installations. The insulation costs do not go up as fast as storage volume goes up.

    Also, if you want to live independent of a large society you can’t do that in a place as densely populated as Haiti. Your fate is too intertwined with that of the larger society in a place like that.

Comments are closed.