Beki Grinter

More on whether ICT4D fits into CS?

In C@tM, computer science, discipline, ICT4D, research on November 24, 2009 at 4:07 pm

Just found a paper by KEntaro Toyama and Muneeb Ali about whether or not ICT4D is Computer Science Research. Rather than answering that as a question, they present two cases.

The case for…

  1. Computer Science problems are routinely found in ICT4D. They observe that the solution to these problems have been published in AI, Networking/Systems, and Computer Human Interaction communities.
  2. The development of, for example, delay tolerant networks (good for ICT4D and also planetary-scale networks)
  3. The community is bound together by shared intellectual commitment to time in the field spent making sure that the systems meet the needs of the people that they are targetted for (this seems to me as a reader like a classical HCI argument)
  4. Common challenges include need for low-cost, power-efficient, interrupt tolerant solutions, for solutions that involve a UI, also challenges faced include literacy and inexperience with technology (I take this to mean both basic and technical literacy challenges)

They suggest that the case against is not one of whether there are interesting challenges for Computer Science within ICT4D, but rather whether the challenges that do exist form a coherent area of research and one that is not adequately addressed by the current areas of Computer Science.

  1. ICT4D is driven by applications. This, they then tie (not explicitly) to other areas in CS that are “socially-applied” such as HCI and argue that within the informal hierarchy of Computer Science, these areas struggle for acceptance. Applications research within CS is a second-class citizen to non-application focused areas of Computer Science research. (I have a bias of course, but I find this argument rather quirky because I see so much CS research as being motivated by application areas… for example, the emphasis on solving GreenIT challenges, or Healthcare as a domain to apply CS.) I guess there’s a difference between doing applications as part of the research and using applications as a justification for research…
  2. They do argue that ICT4D succeeds due to impact in application area rather than the merits of the technology.
  3. ICT4D is by necessity interdisciplinary, which the authors suggest is viewed with suspicion by fields. Interdisciplinary problems seem less legitimate. Also there can be the challenge of finding enough of a problem, and in solving it still needing other aspects of the problem to be solved within different fields before the solution is attainable. So, I take this as arguments about the scope of the problem, is there enough there there, and dependency, relying on other researchers outside the discipline to solve other parts of the problem. These are significant challenges for interdisciplinary research, although again I am surprised by these arguments because interdisciplinarity seems to be oft-held up as an example of the way to find important problems to work on in a type of “whitespace” that exists precisely because established fields can’t go there alone.
  4. I like this quote “As is the case with some work in systems research, its not clear that a core technical contribution is really valuable without building the whole system. Unlike systems research, however, “the whole system” might include non-technical components requiring social, cultural, economic, and political efforts, as well.”
  5. Growing evidence in ICT4D suggests that technology and technological innovation is not the bottleneck. This is a challenge, unless of course one defines technology the way that some others do as a socio-technical system rather than an entirely technical one which the authors do. Socio-technical systems research would surely say “that has always been true” that technology for itself and by itself has never been the problem or the bottleneck. It’s of course getting the technology aligned with the social system, etc… surely this has also played a role in adoption in the societies well served by technologies… so, again I would ask, has technological innovation in the absence of people ever been the bottleneck?
  6. Finally they suggest that the technical content of problems spans the various, existing, subfields of computing. So, it’s not squarely in one (and if it was then there would be no need for a separate field) but it’s also spread out. Any application area requires innovations in networking, systems, vision etc…

And so here they turn to a curious observation. ICT4D is struggling to gain identity and legitimacy within Computer Science. And it was not that long ago that Computer Science went through the same battles. They even suggest that Computer Science experienced the same battles, being to application specific and interdisciplinary. They suggested that what changed was that more people became interested and it became more relevant to society.

So, if this is true, then is one argument that ICT4D do nothing except keep doing ICT4D and recruiting bright minds to the field. But, of course relevance to society for Computer Science was measured by commercial marketplace success in the newly affluent postwar America. This argument also overlooks the important role that War, probably the Second and certainly the Cold War played in fueling the growth of Computer Science. Certainly in its funding. And that seems to be the real key, what will stimulate funding for ICT4D. Clearly some funding already exists. There are NGO’s and other non-profits who want to invest in this area. There is also funding from International Agencies such as the World Bank. But as I’ve said before, this is also out of alignment with the way that Computer Science has typically done business through the NSF and DARPA. And I am quite sure that the NSF would be the first to say that they are a Science and not an aid agency. One piece of work to do is to convince Science and Defense agencies that its not a question of aid, but a question of science and defense.

They suggest also looking to areas of Bioinformatics and so forth, other new areas of Computer Science that have gained substantial traction. This made me immediately think that I wasn’t surprised because Bioinformatics serves me. ICT4D at first blush serves a constituency that has not been such a strong focus of Computing, people who aren’t like me. Elsewhere I have commented that some scholars think that ICT4D faces a challenge that Computer Science has had strong attention on problems that its practitioners happen to personally experience (although I still can’t explain why that seems to make people think that smart alarm clocks are a good idea). And here it is again perhaps. Perhaps bioinformatics also has the serves science angle too, in that biology is a well established science.

Then there’s the name. No-one likes the name. Good names are hard to find.

They also talk about the need for faculty in CS departments to do ICT4D, as a way of establishing its critical mass. They provide a lot of reasons why CS faculty should do this. I’ll suggest another. If anyone thinks its an area that belongs in some way to CS more than other disciplines, get it now before the Information Schools who are growing and hiring in this space own it. Perhaps it shouldn’t be a competition, or doesn’t have to be, but the boundaries between Computer Science and Information don’t seem clear cut to me, so I think there’s more intellectual overlap than Computer Science would like to believe. And there are people who think that the future more generally for technology focused research belongs to the non-CS schools including Information Schools. Perhaps that’s a reason why CS faculty are so unhappy?

They also suggest that CS ICT4D needs a conference. I understand it’s going to get one, attached to the ICT4D conference. I know some people are worried about that, worried that it’ll take the computing research community away from the rest of the broader community.

And finally, well they have a few other things, but I want to pick on just one more… they suggest we’ll need time. Again they point to bioinformatics, a term coined in 1978 and suggest that it took a while for the field to coalesce.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: