Harvard Forum – What do we need to know? · Global Voices
Ethan Zuckerman

Canada's International Development Research Center and Harvard's Berkman Center are convening a conversation today and tomorrow at Harvard on the future of information and communication technology and development (ICT4D). Global Voices will be participating in the event as a media partner, and I and Jen Brea will be twittering and live-blogging the event. You can find out far more about who's around the table and what we're planning on talking about on the Global Voices special coverage page, which includes links to the background papers prepared by participants.
We're here in part so that you can have a voice in the discussions. Please feel free to post questions on Twitter, using the #idrc09 tag, or as comments on Global Voices posts – we'll try hard to work those questions into the coversation here at Harvard. You may also want to use Berkman's “question tool“, which will be used to put questions to the panelists at a public event this evening.
Yesterday's conversations at the Harvard Forum on ICT4D orbited two general themes:
- the need to include conversations about inclusion of women, the poor, the marginalized into dialogs about ICT4D
– a debate about whether we embrace the success of the mobile phone as a tool for development or ask for more capabilities than we're able to gain on mobile networks.
Today's conversation starts with discussions of “knowledge gaps”, open questions we need to answer through research so we can understand what's succeeding and failing in our field.
Clotilde Fonseca of the Omar Dengo Foundation suggests that we focus on creating effective indicators of impact. Educational projects often have difficulty expressing their impacts in language understood by development banks. Success stories are dismissed as anecdotal and not scaleable. Evaluating impacts just in terms of results on standardized tests, the standard evaluation framework, aren't considering “ecologies of learning.”
Beyond evaluation criteria, we need to work on the development of standards, especially standards for teacher development. Scaling up projects from pilot phases to replicable states involves massive teacher development – this, in turn, requires us to ask questions about whether teachers are learning the skills and tools needed to scale and expand these projects.
Fonseca worries that we aren't sufficiently studying “learning communities”, the power of collaboration, networking and sociability for education. These techniques are increasingly recognized as key to learning, but we're not putting sufficient research into the value of networking and communities to education.
We need to broaden our views of what technology can mean for development. We tend to have limited and restricted views of what technologies are and can do. “There's lots of magical thinking,” and a tendency to use a simplistic model – technology and development is the product of infrastructure plus content. She worries that while we understand what infrastructure is, we might not fully understand what content is and needs to be. The interventions suggested post-WSIS tend to be very technocentric and may overfocus on infrastructure over questions of content.
To allow a new generation to learn 21st century skills, we need to face cognitive issues, and learn how the mind actually functions. We need education to create learning skills. It's been risky for governments like Costa Rica to address these issues, but it will be critical to solve these problems to fully embrace potentials for a digital future.
Laurent Elder of IDRC offers three concrete questions about knowledge gaps.
- We're trying to create not just a knowledge society, but an inclusive, equitable knowledge society. Does openness help us achieve these goals? We worry that we've seen with the rise of the mobile phone doesn't necessarily eliminate inequality – we're seeing the GINI coefficient increase in countries with high mobile phone penetration. If we're trying to increase inclusion, do open principles, open content licensing and open innovation help? We don't know yet.
- IDRC sponsored a great deal of research and interventions around telecentres. There's a debate about whether these telecentres were successful. Now IDRC is trying to determine whether building interventions (build our own telecentres) or incentives (support the construction of telecenters or other projects) is more succesful.
- How do “knowledge turns” – the cycle from hypothesis, testing, results to new hypothesis – affect different fields. In the semiconductor industry, knowledge turns take about 18 months, making this a very fast field. The health industry has a knowledge turn of about 8-10 years. Can we embrace these faster-moving cycles? How do we spur innovation at this pace, and what are the consequences of moving this quickly?
Mike Best takes on the emerging cleavages within the ICT4D field. He notes that we're in danger of building unhelpful disciplinary walls, and that this wallbuilding contributes to the “common tendencies for this field to jog in place.”
A recent Doha conference on ICT4D raised the idea that we may want to split the ICT4D field into at least two camps. The computer scientists worry that their fields don't see ICT4D as real computer science. In the hopes of raising the profile of this work, they're planning an ACM special interest group, and considering a CS-only conference in conjunction with the next ICT4D conference in London. This, Mike argues, is a really bad idea.
Computer scientists tend to build ICT4D projects with this method: I decided to build this thing. I worked on it, I adjusted it. I took it to Ghana. I asked ten people – nine of them liked my thing. Computer scientists tend to dismiss work that doesn't fit this paradigm, and especially work that doesn't include fundamental technical innovation. Social scientists wonder whether fundamental techological innovations are really required for ICT4D work. “For either group to think they don't need to sit at the same conferences together is worrisome.”
We're making major mistakes, Mike worries. We tend to view the access to knowledge field as if “knowledge is a reified thing over there amd our job is to offer access to it. Schools, in this cartoon, is where children as empty vessels have information poured into them.” This may be a straw man, but it's too common a point of view, and it's a dangerous one.
We're failing to be a progressive field – we fail to stand on the shoulders that have come before us. And since this field is only a decade old, we've failed to stand upon each other's shoulders. Most projects end in failure – absolute failure, sustainability failure or partial failure. That's not the problem – problem is our failure to learn from our failures.
Mike offers four suggestions to help save our field:
- We need to return to our interdisciplinary roots and read each other's literature. It's a problem that we're all rewarded for writing, not for reading, our collective literature.
- Avoid technofetishism
- Find patient money that can support our work over time – Most projects Mike has worked on are 18 months or under.
- We need to find shared problems and methods especially in the realm of evaluation and assessment. Much as David Hilbert put through key problems in mathematics, we might want to identify the “Hilbert problems” in our field.
Onno Purbo makes it clear that he's an activist, not a researcher. He's both, actually, and he's been one of the key figures in building open, community wireless networks in Indonesia. These networks are designed to save the expense of buying technology from the outside world. “You can use kitchen tools to create a network,” he tells us. “These networks are easily replicable in communities, but its a surprise that it's possible to do these things. People don't believe it's possible until they see it on TV.”
Purbo sees a profound need to make information on community networking accessible to Indonesian communities. We need to translate from English into local languages. He's able to measure success by looking at Google Trends and comparing searches for networking information using English and Indonesian terms – the interest in the Indonesian terms is increasing over time, suggestion more people comfortable in Indonesian are seeking this information.
One area where Indonesians are producing and sharing knowledge is around the idea of the “healthy internet”. Parents and schools are interested in providing access to the internet, but filtering out pornography  – they share tips and techniques through blogs that discuss “healthy internet”. He tells us that there are now 2 million blogs in Indonesia on this topic, and a weekly blog award for the best writings on the topic.
Purbo's wife focuses her work on ICT for women. She helps run a training program that spends three days teaching women how to operate office applications in Linux. The problem isn't the course – it's getting women to be able to take three days off from their work to take the training. Hivos has funded a salary for women participants, but this isn't a sustainable model.
Purbo's latest project involves using the internet within Indonesian schools. Only 4,000 or 240,000 have internet access, so the tools of choice are blogging platforms run from LiveCD or LiveDVD linux distributions, allowing for community publishing within a school, rather than on the live internet. (He offers us a distribution, but warns that it uses the Indonesian translation of WordPress.)
Finally, Purbo lets us know why he's videoing our proceedings. “People in Indonesia are more inclined to learn from video than from text.” He asks that groups like IDRC consider offering incentives for video creation rather than for creating more texts.
Alison Gillwald reacts to Laurent's provocations suggesting that open standards are neccesary, but not sufficient, to create innovation. On the idea of incentives versus interventions, she suggests that there are worthy activities – community media in minority languages, for instance – that can't ever be profitable but are still worth doing. Addressing Mike's questions about research, she notes that it's very hard to find African scholars writing about ICT4D – “the African academic ethos is highly uncritical.” We need to fund local policy interventions that have community involvement, and this might help create local scholarship to analyze the success of these interventions.
Rohan Samarajiva worries that the policy progress we've made is modest, and short term. “The real achievement would be long-term, enlightened policy,” not oriented towards quick wins.
David Malone wonders what we're missing in our discussions. He notes that we've focused heavily on mobiles, but hardly considered satellite television, which has also been a dramatic force for transformation in much of the world, especially the Arab world. He notes that Egypt's media environment has transformed almost entirely – no one watches state-controlled media anymore – they watch Al Jazeera. But this hasn't translated into activism on the ground, perhaps because activism on the ground doesn't pay.
Anita Gurumurthy is concerned about Laurent's question regarding interventions versus incentives, seeing an incentive strategy as overfocused on market mechanisms. She wonders if telecentres have failed because they were too early to provide services and content really useful to poor users. She points out that technologies are transforming public sphere, letting people come into the public sphere in new ways, and suggests that these capabilities go beyond the simple analysis of market supports.
More live-blogging by Ethan Zuckerman
- Day 1-
1) Update from the Harvard Forum on ICT4D
2) Markets, Mobiles and the ability to make culture
3) The complex world of ICT and gender
4) Are we satisfied with what we've got?
5) ICT4D and, and, and
- Day 2 -
1) What do we need to know?
2) Faith and focus