Crowdsourcing 2.0: Why Putting the Slum on a Map is not Enough for A2J

By Christina Gossmann

There was a time—not too long ago—when informal settlements the size of small cities were basically invisible. Beige-gray fields, intercepted by thin blue lines, signifying water, and several thicker, windy white lines that stood for major roads, would pop up on the computer screen when searching for infamous slums such as “Kibera” on Google Maps. The information void stood in stark contrast to the hundreds of thousands of people living in Kibera, ironically tucked away between some of the city’s most valuable and celebrated resources: the Royal Nairobi Golf Club, Ngong Forest and the Nairobi dam.


Over the past few years,participatory mapping has come a long way in Kenya and crowdsourcing geospatial data has entered the ICT4D vocabulary. Aside from reviewing some of the milestone initiatives in this blog piece, I would also like to make the case for a second generation of geospatial data collection—one with more purpose and accountability.

In the not-too-far past, Google-ing Kiberadid not reveal much information about the slum, but more significantly, information also lacked within the slum. In the eyes of the government, the slum did not exist or matter and only few stories, usually about gangs and murders with attention-grabbing sensational headlines, were deemed newsworthy. For relevant and current happenings, residents would therefore consult their social networks: neighbors, friends and family. As the extensive literature on social capital and intelligence has shown, who you know (rather than what you know) contributes enormously to slum dwellers’ complex networks of resilience.

But there are situations when those networks are simply not enough.

One of those situations presented itself on the night of the 30th of December 2007. After Kenya’s general election on the 27th of December, hopes—and polls—were at a peak for the opposition party’s RailaOdinga. But after a three-day delay, incumbent President MwaiKibaki was unexpectedly pronounced winner. What exactly happened next and who is to blame continues to be widely discussed, but what we do know today is that inflammatory text messages and emails had played a major role in inciting the violence that lasted for two months, killed over 1,000 persons and displaced 350,000. Most of what would later be called “ethnic cleansing” took place in informal settlements, including Kibera, the same areas with the least information. Nobody knew whether and when it was safe to step outside the house. After only several days, a small group of programmers released a software that would use the same tactics as the perpetrators—SMS and emails—to create an alternative information-sharing platform. Ushahidi, Swahili for “testimony,” mapped reports of crime and violence that could easily be submitted and accessed online or by mobile phone. This “politics of witnessing” has since spread as a crisis-mapping tool all over the world.

Figure 1: The Ushahidi Platform places reports submitted via SMS and email on a map.

Ushahidi shined the light on a long-existing problem that would finally be addressed: the lack of information on and for informal communities. Although constituting a significant urban demographic in cities of the Global South—and the majority in some, including Nairobi where an estimated 60% of the population live in slums—slum dwellers are often ignored in planning processes and budget allocations.

With the goal to change the situation by literally putting Kibera on the map, an international development practitioner and a programmer founded MapKibera in 2009. Through support from local techies who helped trainKibera residents in using OpenStreetMap(OSM) techniques—including GPS surveying and satellite imagery digitizing—Kibera began making an appearance. In the years to follow, the Voice of Kibera community news website and the Kibera NewsVideo Network journalism project followed, developing atop the MapKiberainformation on OSM.

Figure 2: Even today, few of Kibera’s resources appear on Google Maps.


Figure 3: Open Street Map displays various resources in Kibera including hospitals and schools.

MapKibera was the first of its kind. By training local residents in geospatial data collection and visual storytelling through photography and video, MapKibera significantly contributing to the democratization of media. It also made international news and brought much-needed attention to the Kibera slum. Kibera’sdata and founding members traveled the world. It was actually at one of those trips to the Massachusetts Institute of Technology Media Lab where I first heard about the initiative. MapKibera’s approach and legacy for the ICT4D community cannot be denied.

But is information enough? The most recent decision of the City County Council of Nairobi not to include informal settlements in the new Master Plan of the city—the first since 1973—indicates that it might not be. With a saturated map available on OSM, neither the city nor residents seemed to make much use of it. The City insisted on a dearth of quality information, while residents already knew the locations of basic mapped amenities such as schools, taps and pharmacies in their neighborhoods.

What was needed was information that would allow slum dwellers to claim political agency and their citizen rights. The result was the Spatial Collective, a Mathare-based social enterprise, founded in 2012 by several experienced participatory mappers. Similar to MapKibera, the Spatial Collective benefits from Kenya’s 70%+ mobile phone penetration rate and widely available cheap Internet service to tap into an already existing information system to access local knowledge. They use this information to map slums’ resources but also their most basic needs. Crime and rape reports, for instance, allow for specific interventions such as installing lamps for safety.

As we forge ahead with increasingly popular crowdsourcing initiatives, I would like to extend my recommendation to emphasize not only data collection but purpose-driven data-collection that targets one particular goal at a time. A foreign-founded and –funded initiative, the Spatial Collective has drawback of its own. But if there is anything we have learned from ICT4D projects by now, it’s that nothing’s perfect. 

Christina Gossmann is a Master of City Planning student at UCBerkeley. Before returning to graduate school, she worked as a freelance journalist and researcher in cities of the Global South. Email her at christinagossmann@gmail.com and follow her at @chrisgossmann.

22 comments:

  1. Thanks for your fabulous reflections, Christina. Along with many other things, I pick up three major bold points from your reflections. Firstly, the suitability of agility for ICTD, if not for the whole IS spectrum - as you well reflected with respect to the top ups of Spatial Colective social enterprise over Ushahidi platform - which is further in line with, for example, Hans Rosling's insight of the quest for searchable, presentable, and usable data only after having accessible data (open databases). Secondly, while I wish to learn more about the practicability of the projects in question the viability of involving locals for such "techie" implementations in light of resources and technical competency, the all round value of participation and local content appears to be embraced in the projects. Thirdly, I am being convinced that there is an increasing mismatch between technological "innovations" and their respective value returns to the society. Why are Ushahidi data and its innovators are touring the world, including MIT as a world class innovators? Is that because they delivered Ushahidi solution that avails data which is not used by the City Councils and locates the amenities to locals? Or is it because Ushahidi is the first of its kind and praised by someone as "innovation"? Or is that because there were not other similar innovations that served the society as Ushahidi does?

    ReplyDelete
    Replies
    1. Dear Habesh,

      Thank you for your very insightful comment. I can sense a bit of frustration in your words with the technological determinism that goes with pretty much every international development tech-project. I feel it too! Technology and crowdsourced information will not necessarily solve the underlying problem that slum dwellers lack access to public services, jobs and education BUT--and this is a big but--if there is no data, negotiating with the the government to demand access to those (guaranteed through the constitution) services becomes almost impossible, I would argue. However, it is true that the effect on the community is unclear. As far as I know, Ushahidi in its initial post-election violence project and MapKibera did not collect baseline data to measure any kind of effect their intervention may have had on the slum communities. Spatial Collective, the last organization I described in my post, IS now actually collecting baseline data to see how, if at all, their interventions affect the communities. But since they only started in 2012, they are still a pretty new organization, so it's hard to tell what impact they will have. All their projects are in its beginning stages.

      To answer your more specific questions:
      I don't know if Ushahidi was traveling the world (although they should, because their platform has proven to be very useful for crisis-mapping all over the world!), but it is true that I met one of the MapKibera founders at a presentation at MIT. It is not unusual, and not unhelpful, I would argue, to present your innovation/app/project at different forums, especially at U.S. universities, think tanks, etc. in order to raise funds and get feedback. It inspired me to visit, research and write about crowdsourcing, and many others too, I'm sure. Do you disagree? If yes, why?

      Delete
    2. Thanks again, Christina; and sorry that I have edited my post display name to reflect who I am to the network.

      I am happy that you share my prime concerns on tech-led projects. However, I am still being convinced that "innovation" is seen as something that didn't happen before rather than something that can be implemented in efficient way and deliver more to the beneficiaries. In this case, for example, I am sure that locating a school or tap is not crucial for dwellers as having basic education or clean water is. I used to know where all this facilities were while I was residing in slums in Addis, Ethiopia. I totally agree in that sharing whatever "innovation" - with its limitation- with the broader audience will benefit.

      Delete
  2. Great blog post Christina! Another similar project from India that we discussed in class: http://www.tnscb.org.in/RAY/map/slums.html

    ReplyDelete
    Replies
    1. Dear Seema,

      Thanks for your comment! I checked out the project you posted in India, and I would like to find out more about it. I could not find a legend to the map, but I'm assuming the blue areas correspond to different slum areas? How did this project start, what was/is its mission and how has it developed over time? Is it also doing targeted mapping, like the "solution" I am recommending (via Spatial Collective)?

      Delete
  3. Thanks a lot for your very interested blog post Christina. In advance sorry for my bad English!

    It’s true that information can’t resolve everything. Nevertheless it is better than no information at all. About Crowdsourcing, your article reminds me another blog post I have read some month ago. It was about an app (I will try to find the name) to fight against food loss in some African countries. The idea is that today a lot of foods in Africa are lost during the transport because the transport conditions are very bad: the poor road infrastructure creates some vibrations that can damage the fruits and vegetables, the weather and the rain can also make the fruits rotted and the duration of the transport may be very very long sometimes (too long) because the drivers have often to stop during the trip to have to pay bribes to police…

    Thus, this app provides to the drivers some very useful information about the best road to take to go from A to B thanks to crowdsourcing information (where is the worst road, where is the police, where there is some accidents or dangerous climatic events etc.). Of course the app work in real time and with GPS position, so when a previous driver has a problem, he can alert all the other users directly with the app and sends an alert.

    ReplyDelete
  4. I found the app! It is called "Cheetah: : http://www.copernicus-masters.com/index.php?kat=winner.html&anzeige=winner_esa2013.html

    ReplyDelete
    Replies
    1. Dear Antonin! Thank you for your great feedback and your English is perfect!
      I agree with you that information is important and better than none, but I also believe that if you do involve local residents, by training them, etc.--like in all the cases I described in the blog--you raise expectations. In that case then, if you do not deliver some kind of change (improvements in slum dwellers' livelihoods, more public services, less crime), your future operations might be greatly endangered. Residents might no longer want to participate--and you fail in meeting your main goal: improving the lives of slum dwellers!

      I like the idea of "Cheetah" and it is trying to tackle a very serious problem in Sub-Saharan Africa, and many other countries in the Global South--transportation difficulties due to low-quality infrastructure. Do you know this app's uptake rate? How many people are using it and what do they think about it? Is it working?

      Here is another interesting transportation-related initiative you might find interesting. It's focused on mapping informal transportation systems, matatus, in Nairobi: http://www.digitalmatatus.com/

      Delete
  5. Great post Christina! These platforms have been valuable in bringing out those stories from obscure places which were ignored by the mainstream media.

    These crowdsourcing initiatives can be used to highlight stories ranging from bad governance to crisis mapping. Here is another interesting post that supports your recommendation on purpose driven data collection: http://willtradeforelephants.tumblr.com/post/62807471344/using-tech-to-make-kenyan-problems-harder-to-ignore

    ReplyDelete
    Replies
    1. Thanks for your comment and for the link you shared! It looks like the author you mentioned is making a very similar argument to mine, using data and evidence from Ushahidi and the Spatial Collective. I'm curious: how did you come across the blog?

      Delete
  6. This comment has been removed by the author.

    ReplyDelete
  7. Great post, Christina. There are a couple of slum mapping initiatives targeted at collecting data about slums and mapping them on a city wide map with colour codes indicating the density of population in those slums. However, the city municipality does not want to either assist these initiatives or use the mapping in any way. I think its a great idea to crowdsource and geographically map such information (Ushahidi) and I agree with your recommendation of having a purpose driven mapping; but there should also be a way to either make it accessible to the intended audience (who could be low literacy users) directly or have a governing body that can undertake actions on behalf of the intended audience. In the absence of either of the two, I feel that such mappings only end up being pretty visualisations.

    ReplyDelete
    Replies
    1. Dear Priya, thanks for your comment! I definitely agree with you on the latter necessity (have a governing body that can undertake actions), but not necessarily the former (make it accessible), depending on who you mean by the audience. In this case, I believe that some of this mapped information was intended at decision-makers rather than slum dwellers. In this case, it seems that their acknowledgment (rather than that of slum dwellers) might take precedence. This is, however, a quite specialized case. Usually in ICT4D projects, I would agree, reaching the marginalized is necessary in order to graduate from simply being a "pretty visualization."

      Delete
  8. Awesome post Christina! The topic of crowdsourcing and influencing high-level decision-making made me think of laborvoices.com. It’s an organization that collects reports of workplace abuse, makes them transparent, and influences companies that are higher up in the supply chain to negotiate better workplace conditions. I think that open and transparent data goes a long way in bringing issues to people’s attention, whether for policy-making or awareness raising. On a related note, what are your thoughts on how organizations like MapKibera could raise awareness about or mitigate the current slum displacements in the upcoming world cup in Brazil?

    ReplyDelete
    Replies
    1. Dear Timothy, thanks for your insightful comment! laborvoices.com seems like a very interesting initiative that aims to change labor conditions by democratizing information. The only difference might be that it is not necessarily geospatial, is it? I could not tell from their website.

      Regarding Brazil, I think MapKibera and Ushahidi are great mapping initiatives to uncover displacement and evictions that are taking place in preparation to the World Cup and the Olympics. I was actually working as a reporter in South Africa during the 2010 World Cup, covering some of the related displacement of homeless people and shack dwellers and I think geospatial mapping would have made a great difference in communicating the gravity of the situation. However, once again, the audience is the key. Who would we try to tell this information? The government? The international community?

      Delete
  9. Hi Christina,

    Thank you for the interesting blog post. One of my main interests is in choosing innovations that work in one African country and applying them to another. Of course a lot of changes need to be made because no two regions are exactly the same, but the slums have a lot of similar challenges. I will be working in Khayelitsha, South Africa in the summer and innovations like MapKibera and Cheetah appear to have a lot of potential outside Kenya too. Though the Blog was not about Cheetah, I was really intrigued by this software and thank you Antonin for this post. This innovation tackles a daily problem in the third world - tracking kombis(matatus) - and my first guess is that it would be well received. The problem, as Seema mentioned, comes with making this service available to the intended audience.

    ReplyDelete
    Replies
    1. Thanks Sidee for your comment! Exciting that you'll be in Khayelitsha (!), doing what exactly? I'd be curious to learn more! Yes, participatory geospatial data collection has been collected and used quite extensively internationally since it first made an appearance in the mid-2000s or so. You can check out where Ushahidi has been used. It ranges from Haiti to the Philippines! But those initiatives are usually focused on crisis situations. But what to do about daily challenges as the one you mention about transportation? Especially if the intention is not necessarily to communicate which bus is where at what time to users (since they probably know already), but to establish a kind of legitimacy for informal transportation systems. That at least has been my experience in doing some research in mapping informal transportation systems in Kenya. Did you check out the digitalmatatus initiative that I recommended Antonin to have a look at?

      Delete
  10. Cool post Christina. It definitely seems like having some data is better than having no data at all. Your topic of crowdsourcing and the mentioning of MapKibera remind me a lot about a similar use by the travel agencies these days. They’re a for profit company that borrows the same idea by heat mapping the photos taken at each popular location around the map and overlaying it on Google Maps. This app allows people to view popular sites and it’s pretty similar to MapKibera in a way. This idea is relevant to ICDT but it makes me think that it can also be used for other areas such as finding trends in the agricultural industry or to track the airfare prices at different times of the year depending on the amount of traffic to a location. I think all of this transparent data gives leaders more information to discuss understand the situation a lot better.

    ReplyDelete
    Replies
    1. Dear Johnny, thanks for your comment! Interesting that MapKibera, Ushahidi and Spatial Collective made you think of how ICT4D practices can be applied in the for-profit sector. This indicates the extent to which ideas that originate in the Global South can actually travel up, towards the North and West, rather than always being on the receiving end of the international development ideas spectrum.

      I also think that transparency is good, although in some cases, especially those informal practices (like informal transportation systems) that are semi-(il)legal. More info becoming available to authorities could also lead to more reinforcement of practices that substitute non-existing services among the (urban) poor.

      Delete
  11. Hi Christina,

    I have to agree with everyone who says that having data is better than not having it. Even if it is not being used now, perhaps it will be used later. Reading the end of you post I wondered if the crime and rape reports are just allowing specific interventions, or actually prompting it? It is certainly amazing progress if these purposeful and targeted interventions are truly being carried out because of the described app. Do you know if they were aware of and made use of the Ushahidi Platform mentioned earlier in your post? One source of skepticism I've observed is how even in the field of Information Technology many practitioners do not seem to have information about the other initiatives going on in an area and efforts get duplicated and sometimes confused.

    ReplyDelete
    Replies
    1. Dear Crysta, thanks for your insightful post! I totally agree with you that correlation does not equal causation and that there is (too) much replication and lack of coordination among ICT practitioners, even on the ground.

      The Spatial Collective was intensively engaged in raising awareness around the connection between crime/rape and safety lights. Whether lights were installed due to their direct efforts or due to their effort that caused another intervention that ultimately led to it, I don't know. There might be a excluded variable bias here for sure.

      In the case of Ushahidi, Map Kibera and Spatial Collective, I personally know that they are very much aware of each other. I would say they all borrow from each other and do not replicate each other's work. This is not always the case though. I think you find such replications and confusions especially in the realm of infrastructure development and also in areas where on-the-ground efforts are accompanied by initiatives that originate in very large institutions like the UN.

      Delete
  12. Dear all,

    Thanks again for all your insightful comments and thoughts for food! I thoroughly enjoyed the discussion this past week and hope that we can continue in the months to come, as we learn together.

    I'd like to share with you that this blog post (with several more cynical additions) was re-published on South African magazine The Con here: http://www.theconmag.co.za/2014/02/16/crowdsourcing-2-0-why-putting-the-slum-on-a-map-is-not-enough/ and that one of the founders of MapKibera, Mikel Maron, actually commented on the blog post, somewhat in defense of MapKibera. Check it out!

    ReplyDelete