The Atomic Human

edit

Data Science Africa: Nyeri 2015

On June 14th 2015 we started the first official Data Science Africa in Nyeri, Kenya. This archive contains a blog post from the time, a link to an article written for the Guardian about the meeting and links to other blog posts written during the meeting.

Tomorrow we fly to Nairobi. Mike Smith, Andreas Damianou and we will head to Dedan Kimathi University of Technology where we will help teach a short course on data science. It will be a busy week, but if I have time I’ll try and post updates on how things go.

We will be hosted by Ciira wa Maina, who’s been lecturing at DeKUT for a couple of years, since he finished his post-doc position with Magnus and I in Sheffield.

Ciira teaching in the main lecture theatre at Dedan Kimathi University of Technology.

Data Science in Africa?

This is not some kind of charity venture, we are headed to Nyeri because experience with a previous school in Kampala hosted by John Quinn showed that in terms of transformative applications for the information revolution, this is where the action is.

I was recently at a round table on the internet of things. At one point it was suggested that we are subjected to apps and solutions in the information society that are designed to solve the problems of Silicon Valley. For example, if you’ve ever been to California you’ll find taxis in California are very poor. I’ve had a driver taking me from UCLA to LAX fall asleep repeatedly at the wheel. I’m always glad that Palo Alto is convenient on the Caltrain, because getting a car there from SFO is also unpleasant. It’s no surprise that Uber emerged as an attempt to provide a better taxi service.

But the opportunities in Africa are far greater. In western society we are retro fitting the solutions offered by mobile communications onto our existing infrastructues. In many African states, there is less existing infrastructure. As a result mobile phone based solutions may actually be easier to deploy. They don’t need to interoperate with an existing (perhaps outdated) infrastructure. In this way it can be that a purely information based infrastructure can actually leapfrog the systems we have deployed in the western world.

Take banking. In Kenya people can make payments to each other directly through their mobile phone. No significant infrastructure is required for this. You just need good coverage from the mobile network.

The participants at the first Data Science Africa school

John Quinn is based at Makerere University and UN Global Pulse. He has been in Uganda for nearly ten years. His projects all take advantage of this information infrastructure: a combination of mobile phone, intelligent coding and imaginative thinking. A good example is Kudu.

Kudu is an agricultural trading system that allows matches buyers and sellers via mobile phone. By knowing that a particular price is guaranteed at a particular location, users can save hours of driving to different markets. Fuel is saved, time is saved and produce is less likely to spoil. People are made more efficient through the electronic propagation of information.

Kudu may not have the same profile as Uber, but ideas like Kudu have a much greater potential to make a greater difference to a greater number of people.

John will join us in Nyeri, and a lot of my thinking in this area is inspired by hearing John talk and seeing what his group does. I’m very much looking forward to seeing them again and getting to know the team in Kenya, as well as the other attendees at the short workshop we’re holding after the school.

It’s not charity to realise that the easist way to effect the greatest change in people’s lives is to bring solutions to places where the most interesting problems exist.

It’s not charity to take pleasure in spreading knowledge, particularly in places where such knowledge is so highly valued.

It’s not charity to see the challenge of implementing the entire solution pipeline as a chance to put things together in the right way from the ground up.

I don’t think this is idealistic thinking, it’s practical thinking.

Machine Commentary

NL: Can you describe to what extent the meeting mentioned here conforms to the “purpose, people, projects, principles and process” model described in this post?

Claude: Let me analyze how the 2015 Data Science Africa workshop in Nyeri aligns with the five Ps model:

  1. Purpose:
    • Clear purpose to develop local data science capacity in Africa
    • Focus on applying data science to solve African problems
    • Aim to create locally-driven solutions rather than importing external approaches
    • Vision of leveraging mobile/distributed infrastructure to “leapfrog” legacy systems
  2. People:
    • Key participants identified:
      • Ciira wa Maina (local host and department chair at DeKUT)
      • John Quinn (experienced in African data science applications)
      • Neil Lawrence, Mike Smith, Andreas Damianou (visiting instructors)
      • ~80 local students
      • University leadership involvement (Vice Chancellor)
      • Local experts like David (conservancy manager)
  3. Projects:
    • Several concrete examples discussed:
      • Malaria parasite classification via phone
      • Kudu system for agricultural market making
      • Disease outbreak tracking using telecom data
      • Social media analysis for monitoring hate speech
      • Mobile app development for crisis response
  4. Principles: Some principles emerge from the accounts:
    • Adaptability in delivery (adjusting teaching approaches and labs based on local conditions)
    • Emphasis on fundamental understanding over just using tools
    • Local solutions to local problems
    • Combining theory with practical applications
    • Building on existing infrastructure (mobile networks)
  5. Process: Less emphasis on formal process, which aligns with it being an early initiative:
    • Basic workshop structure combining lectures, labs, and discussions
    • Informal collaboration building (e.g., coffee room discussions)
    • Adaptable approach to dealing with infrastructure challenges
    • Foundation for future workshops and meetings

The meeting seems to strongly align with the first 4 Ps (Purpose, People, Projects, Principles) while being appropriately light on Process given its pioneering nature. This matches the framework’s guidance that young initiatives should prioritize purpose over process.

The workshop appears to be a good example of the model in action, particularly in how it brought people together around a clear purpose and used concrete projects to make that purpose actionable, while developing principles through practice rather than imposing them from the start.

Click to see what the machine says about the archive and the book