XVIII.6 November + December 2011
Page: 64
Digital Citation

Lessons learned in prototyping systems in developing-world contexts

Sena Allen

back to top 

This article was written by a master's-level student reflecting on his first year of research work in a developing-world project. Captured here is what he wished he had known before starting. We hope it serves to better prepare other students in the same situation. —Gary Marsden, Editor

Research in ICT for Development (ICT4D) can be a very rewarding experience, but it is not without its challenges. During the course of my master's degree, I have witnessed many events of which I wish I'd had some foresight to better prepare me for the long, complex journey of an ICT4D project.

These challenges are unique and are almost exclusive to developing countries. Factors such as fear due to high crime rates, lack of familiarity with technology, lack of infrastructure, and political instability all greatly affected my experience as an ICT4D researcher. This article explores some of my experiences and sheds some light on what sorts of events can occur during similar projects—perhaps it will even mitigate similar challenges for future ICT4D researchers and help make their lives easier by showing them what to look out for.

My project entails creating and observing the effects of a content-management system with a situated display—a large screen that uses Big Board, a service that transfers data to and from the board and a mobile device via Bluetooth [1]. It works by displaying separate images for each data item. For example, there could be a picture of Nelson Mandela representing a file containing images and information about his life. An individual would then take a picture of this image and send it to the board via Bluetooth. The board recognizes the image and proceeds to send all related files to the phone that sent the original picture. The system was envisioned as an innovative way to freely disseminate information to individuals who had access to mobile phones without special client software installed.

There are a number of user groups within our research environment that could make use of this technology to provide relevant information to low-income residents. Providing a system that can manage these groups' access to placing material on the board is the focal point of my research. One of the major stakeholders is an NGO that works with local teenagers (which might want to put educational content as well as organizational announcements onto the board). A second set of stakeholders is a group of librarians who can provide educational content, community news, and perhaps some entertainment content for the youth and community to consume (see Figure 1). The final stakeholders are the local youth, who represent the target consumers of all the content created by the first two groups.

The project is located in a township outside of Cape Town, South Africa, called Khayelitsha. The situated display is located in a local library within the township. The largest single township in South Africa, Khayelitsha boasts a very young population. In fact, a population register update to the national census states two-thirds of the population in Khayelitsha is younger than 30 [2]. Like many other townships in South Africa, it also has a relatively high crime rate.

back to top  Observations

In order to study media usage, we employed a technology probe methodology and deployed a large situated display into the environment during the design phase of the project. It's interesting to note that when the display arrived, the very first comment some of the stakeholders made was not about how best to use this new technology, but rather "How are we going to secure this thing?" Such technology stands out in the environment, and people have been conditioned to treat technology as an extremely valuable resource. Though the security of the system was important, I was more interested in how to manage the content of the system and how best to design it. Yet at every meeting with the stakeholders, someone raised a concern about whether we should bolt the display to the wall or purchase a padlock and chain it to the wall. Eventually we obtained the chain and lock and secured the display against a pillar, as shown in Figure 2; the librarians insisted that I hold onto the keys since they had heard of other librarians being victims of crime.

To better understand the media consumption of the teenagers, the wider research team (of which I was a member) disseminated 20 feature phones among specific teenagers in the community and observed their usage. We observed many practices among the teenagers, such as phone sharing and Bluetooth media sharing, as well as others. But we also noticed a certain level of fear among the teenagers regarding their phones. They were not entirely comfortable using these brand-new phones anywhere except home and maybe school. They said they were "too afraid of it getting stolen," and even carried their older phones as decoys in case they were robbed. One teenager also expressed interest in damaging the face of the phone slightly to "make it look older." When questioned about this matter, all the teenagers felt the same way about now being potential targets of crime. Hence, they had creative ways of hiding the phone in socks and undergarments when moving from place to place.

Although the teenagers managed to keep their new phones safe, the situated display in the environment was victim to crime during the technology probe deployment. The device requires a Bluetooth dongle to run; the display is useless without it. The library had been subject to crime and vandalism before, and a librarian was quoted as saying, "Some people here have the urge to steal anything with no regard for the community." The Bluetooth dongle, located in the back of the display, resembled a flash drive. The librarians suspected it was stolen as a result of this misidentification. As a result, I resorted to gluing a new dongle into the USB port, essentially damaging the hardware of the screen for security reasons. But unfortunately, this did not help: the second Bluetooth dongle was stolen as well. After the second theft, I had to make use of a computer with built-in Bluetooth capabilities (together with a security cable tied around a pillar to make sure the computer did not get stolen) and hope that this level of security deterred would-be thieves.

In addition to having a high crime rate, violent events do occur in the area. The NGO we worked with, Ikamva Youth (, was victim to a tragedy during the course of my research. A politically affiliated group mistakenly attacked the NGO office. The staff was harassed, and later a petrol bomb was thrown into the office space. According to the NGO, "In addition to destroying the office, the attackers threw stones into the Nazeema Isaacs Library and torched the adjoining Zimele Pre-Primary school" [3]. The NGO suffered property damage, equipment damage, and data loss, among other things (see Figures 3 and 4). Luckily, no one was physically harmed, but the shock of the event did take its toll on some of the workers. My research was affected since the NGO workers were some of the primary users of the system. Their operations were severely restricted by the event, and therefore they were not as available to use the system as they would have been, since they had to sort out building a new office. I had to take a patient approach and work around the schedules and limitations that were newly imposed on the NGO workers.

Other challenges we faced were technology related. For example, when demonstrating a high-fidelity prototype of a potential system to some of the stakeholders as part of a participatory design meeting, I used a netbook. Unbeknownst to me, they had never used a trackpad before, which affected their comfort levels during the prototype walkthrough. It took some time to adjust to using a trackpad, and there were feelings of uneasiness throughout. If one were to repeat a walkthrough, I would suggest mimicking as closely as possible the environment in which the users are comfortable, paying special attention to minor details such as using a mouse instead of a trackpad. After I witnessed this event, all future prototype demos took place with technology the users were familiar and comfortable with.

One must consider the thoughts and feelings of everybody one works with in the field and make sure any technology introduction is not creating an unnecessary burden on anyone.

The Ikamva Youth NGO makes use of a computer lab donated by the Shuttleworth Foundation (, a nonprofit organization that teaches computer literacy and provides some technology services to the NGO members. All of the computers use Ubuntu, a Unix-based operating system, and therefore all classes are geared toward using this OS. During my research I helped facilitate one of these computer-literacy classes. The participants consisted of 10 to 12 11th-grade learners (roughly ages 16 to 18). The students struggled with abstracting certain things like file structures on the Unix-based computers, as for all but one of the students (whose family had their own desktop) it was their only access to a desktop computer. On the other hand, all the youth were extremely proficient with their phones, to such an extent that we thought they would be able to type much faster on their phones than on a desktop computer. All of the students also knew how to use Bluetooth for file sharing if their phones supported this.

back to top  Overcoming These Issues

By looking at these difficulties, we can find ways to better prepare for the unique challenges present when undertaking an ICT4D project.

Patience and tactfulness are crucial when forming relationships with NGOs or people within the area. One must consider the thoughts and feelings of everybody one works with in the field and make sure any technology introduction is not creating an unnecessary burden on anyone. Consultation throughout the design process is vital to gathering people's thought processes and comfort levels, and therefore one should consider using participatory design methodologies for the duration of a project. For example, after the bombing incident, I had to be mindful of the NGO workers, making sure I was not putting pressure on them due to my research constraints, as they already were already coping with a great deal. Sometimes research agendas have to take a backseat to long-term relationships.

Foresight and careful planning are also important attributes for deciding when to go to certain areas. For example, visiting some neighborhoods late in the evening might not be the safest idea. Another example is to avoid events such as political elections and rallies. Although a researcher might be there to try and help the community, it is easy for misunderstandings to occur—one should always be prudent in areas where political violence might happen.

Physically securing equipment is also an issue; contingency plans for securing any items that one would bring should be made, and items should be insured.

When carrying out experiments, it is best to use equipment that is as close to what the participants are used to as possible. Demos and walkthroughs should be done in a familiar environment, and the participants should feel comfortable expressing any concerns they might have.

Creative methods to overcome infrastructure limitations often must be carried out; for example, bandwidth and Internet speed limitations might be present, and one might need unconventional means to solve a problem, or sacrifice certain functionalities as a result.

back to top  Conclusion

It is tempting to become disheartened when experiencing problems in an ICT4D research project such as those discussed here. There are undoubtedly many challenges, but the rewards for helping out a community are great, and it is highly motivating to actually see the excitement such projects can generate within a community. So while it may be more straightforward to undertake a more traditional ICT project, such as designing a ticket-booking system, the benefits of undertaking an ICT4D project greatly outweigh the potential setbacks. I hope that my experiences can be a lesson to future researchers undertaking ICT4D projects, and though experiences such as petrol bombings and rampant theft may not happen to others, it is useful to know that these events are possible and to keep them in mind.

back to top  Acknowledgements

I would like to thank the staff at Ikamva Youth and the Nazeema Isaacs Library, who have kindly hosted and supported my work. I would also like to thank my collaborators, Marion Walton and Silke Hassreiter from the University of Cape Town, for creating this opportunity and supporting my work. This work was supported by grants from the National Research Foundation of South Africa and Nokia Research Kenya.

back to top  References

1. Maunder, M., Marsden, G., and Harper, R. Creating and sharing multi-media packages using large situated public displays and mobile phones. Proc. of the 9th International Conference on Human Computer Interaction with Mobile Devices and Services (MobileHCI 07). ACM, New York, 2007.

2. South Africa. Sub-Directorate Population Development in the Research and Population Directorate. Department of Social Services and Poverty Alleviation. The Population Register Update: Khayelitsha. 2007.

3. Interview acquired from (2011).

4. Image acquired from (2011).

back to top  Author

Sena Lee Allen is currently pursuing a master's degree in computer science at the University of Cape Town, specifically in the field of ICT4D. He has lived in Africa all his life, and his heart is firmly set in that continent.

back to top  Figures

F1Figure 1. Young users interact with the display in the library.

F2Figure 2. Lock and chain securing the situated display.

F3Figure 3. The Ikamva office after the bombing [4].

F4Figure 4. Melted computer screens [4].

back to top 

©2011 ACM  1072-5220/11/11  $10.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2011 ACM, Inc.

Post Comment

No Comments Found