A few days ago I came across an article from SciDev.Net, published on 13 February 2009 by journalist Katherine Nightingale, titled “Rural Internet, not online but still connected.” I’m quoted in it as Geekcorps’ Mali country director, a period of my professional life that goes back to my years setting up ICT access projects in West Africa.
Nightingale’s piece explored a paradox that was very much alive in the digital development world at the time: what if the best way to connect rural areas wasn’t a permanent connection, but an asynchronous one? Rather than laying miles of cable or relying on a constant signal, several organizations, United Villages in India, Wizzy Digital in South Africa, and Geekcorps in Mali, were experimenting with systems where data physically traveled: on a USB stick strapped to a motorbike, in the bus serving the village, or through an overnight connection at off peak rates.
The article described the project we ran in the Koulikoro region of southwest Mali, using a “desert PC,” a computer designed by Geekcorps to withstand the heat, powered by a single 22 watt solar panel and connected to the world via a USB stick carried once or twice a week to a telecentre in Ouelessebougou. The local cybertigi offered basic office tools, photocopying, digital photography, and email.
In the article, I explain that the service was shut down after a year due to insufficient demand. The reason wasn’t really technical: fewer than a quarter of people living outside the capital can read and write, and rural Malians, while genuinely interested in connecting with the wider world, did so mainly through radio and mobile phones, not email. Geekcorps then shifted its focus to offline resources, including a French language version of Wikipedia updated yearly and distributed on CD or USB stick.
The best designed technology only takes hold if it meets the usage patterns, literacy levels, and social infrastructure that actually exist on the ground.
This article is over fifteen years old, but I still find it instructive. It documents a lesson I’ve run into in other forms throughout my career, including in my current work on AI governance and human rights at ITU-T. You don’t connect a population by imposing the channel that seems obvious to you. You start from what people already use and build around it.
It’s also a small personal pleasure to find a written record, published at the time, of those early field years in Mali. Before Geneva, before conference rooms, when the job meant keeping a computer running under 45 degree heat with a solar panel.
Original article: SciDev.Net, “Rural Internet, not online but still connected,” 13 February 2009