International Conference on Social Implications of Computers in Developing Countries, 2013

Ocho Rios, Jamaica, 19-22 May 2013

Conference website

By Caitlin Bentley SSI Fellow and PhD Candidate at Royal Holloway University of London

Highlights:

1. There is greater acceptance within the ICT4D research community at IFIP WG 9.4 that we need to move from research that explores the ways in which new ICTs are accepted, appropriated and resisted towards exploring the links between ICTs and development outcomes.
2. Researchers engaged in co-design or participatory design methodologies are struggling with scaling-up and transferring these valuable projects to different contexts.
3. ICT4D researchers are multi-disciplinary and use a variety of innovative research methodologies that are not well-supported by off-the-shelf commercial qualitative analysis tools in particular. There is a need for greater awareness of the variety of low-cost tools available and there is a special need to create links between this research community and producers of qualitative analysis software.

Event report:

ICT4D is a multi-disciplinary area of research and practice that has traditionally focused on the design, development and application of ICTs to solve international development problems. The theme of this conference was ‘Into the Future: Themes Insights and Agendas for ICT4D Research and Practice.’ The general feeling amongst those that I met and spoke with was a general dissatisfaction with narrow views of development as economic growth and the impact that this has on the rationalisation of ICTs as a critical element in increasing economic growth. The keynote, Saskia Sassen, for example talked about approaching electronic interactive domains instead as ecologies which creates space to acknowledge that technologies are not neutral and have within them social and cultural logics embedded. Along with  the track on ICTs and the Capabilities Approach chaired by Dorothea Kleine and Yingqin Zheng, there are many within the ICT4D community at this conference that are departing from the traditional view of ICT4Economic Growth in favour of a view of development that is concerned with how ICTs can support the expansion of human freedoms. At the same time, there was also concern that we need to be able to create links from ground level case studies to learn what is happening in the field at a meta-level. From a research perspective, a paper presented by Gomez, said that ICT4D researchers are not typically reporting research questions, methods and approaches that they use to explore research problems which means that it becomes incredibly opaque to discern perspectives and make connections across studies. There was also a panel on theorising ICT4D research that also made the point that we come from different academic traditions with different vernaculars whilst we often aim for our research to eventually be operationalised in policy and practice. This means that as researchers we need to be careful with language and make more efforts to create these connections across studies.

Health information systems and technologies used to support these were another common area at the conference but I did not have time to attend many sessions on this topic save for one that was immediately before my presentation and the second keynote presentation. This keynote, by Sundeep Sahay, explored technical challenges from a humanistic and ‘moral’ position. Explaining that healthcare workers are often over burdened by the introduction of new ICTs that rarely address their actual information and decision-making needs and these ICT systems are instead used as monitoring and control mechanisms. In this sense, the biggest challenges seem not to be whether to use mobile, cloud-based computing, open source, etc. but instead how to make visible the political power dynamics at play. Regardless of this, ICTs and their use in solving grand public health challenges represents a major avenue of research and practice in ICT4D.

Another area that I was interested by was the use of software to document indigenous forms of knowledge. One presentation by Heike Winschiers-Theophilus, explored how they engaged in a co-design process with a community in Namibia. The presentation was concerned with what aspects of a co-design process are transferrable to other contexts, i.e. how much of a relationship does a designer have to engage with a community to adequately understand indigenous representations of knowledge. They had developed a sort of drawing game where community members would draw a concept (cow, house, marriage, etc.) until another member of the community could guess what it was. Then they would design these representations in 3D modeling software. Considering each community has different concepts and representations of these, the software has to adapt or change considerably for each community. This represents a great challenge for them. The team is willing to share their software and although I did not get a chance to speak with Heike for more than a second, it seems that they are not quite sure what to do with their software, i.e. they haven’t published it, and they probably haven’t thought about what to do with the various forks of their designs. I’m not entirely sure though. I hope this example conveys the complexity of software design within our field, not only in terms of versioning, but also in terms of variety of use or utility.

The last area I wanted to write more about are methods. There were a few presentations by the New Mine Lab at the University of Lugano. Co- and participatory-design, participatory video, and etc. are another common area of research within ICT4D. These methodological techniques are not  well-supported by off-the-shelf software. Sara Vannini for instance, presented the results of her study on creating social representations of community media centres in Mozambique through photo elicitation. They ended up printing all of the photos, grouping them, and then analysing interview text in comparison through nVivo. She said that the process was very long, and very tedious because it was extremely difficult to create connections between categories of photos and analyse interview texts based on those categories. It was difficult to change and manipulate categories and subsequent analysis because the software is not designed to handle photo elicitation (it is a method where photos are representations of participants’ ideas rather than a representation of ‘data’ per se). Overall Sara said that she would never use nVivo again, which to me is a concern that we currently don’t have adaptive tools that support ICT4D analysis methods. The director of my research centre, Dorothea Kleine, explained also that her project had to pay exorbitant licensing fees for Atlas.ti so that her research collaborators in Chile and in Brazil could contribute to the analysis phase of her project as well. These software constraints are considerably constrictive for ICT4D researchers in general. Many of use aren’t aware of free alternatives or have the skills to know what software designs would be useful to us. In this sense it is challenging to look for alternatives.


Overall, I think that the Software Sustainability Institute can learn from our domain of research particularly in the realm not of software as research instrumentation, but as software as a research product, research in our field demonstrates that questions need to be asked concerning why, how and for what ICTs including software should be a point of focus. Likewise, our community could learn from the SSI in terms of gaining confidence in engaging in software design and selection that suit our needs as researchers. It is an area of research that is generally fundamentally concerned with software sustainability and so I feel there are excellent opportunities for cross-fertilisation of ideas and collaborations. One concrete idea would be an assessment tool that could help researchers evaluate the sustainability of their research tools and products that would enable them to link this evaluation to the benefit of increasing software sustainability.