Projects galore

November 27th, 2008

Ha, the bluff is called! I now have to actually commit some constructive ideas in writing! This will have to be followed by a second post thoroughly critical of all that I propose, otherwise I’d betray my mission in the world of ICTD.

Currently, I am not working on any specific project, and i’m in the pre-dissertation phase of graduate life that resembles the phase kids go through when they are 7, also known as the ‘will I be a ballerina or an astronaut’ phase. So my thesis topic goes from ballerina to astronaut without any logic or committment. So, all the projects I am proposing are a. developed in the perfect engineering tradition of not asking anyone, just thinking things in one’s own mind and b. imagined as possible starting points for a semester in Berkeley, followed by field work.

Given these limitations, I focused on 1. ideas that create partnerships locally and explore, again locally, possibilities for projects or for learning in view of future field work and 2. on possible developments of existing projects.

Ok, now that I’m done with the disclaimers, here’s my ten brief ideas, in no particular order:

10. agriculture: go out in the great California fields to do an ethnography of farmers, to find out what are their information needs, and when and how information is shared or not shared. I have no idea of whether or not this will be relevant to projects in Africa, but it builds topical familiarity which I suspect is generally missing.

9. agriculture: get in touch with food importers and traders who work with small farmers in developing countries, and ask them what they are looking for in their suppliers and what is missing. This might help developing ideas for applications that may not be obvious when user-centered work is done in the field, but that may be very beneficial. See post about middle men.

8. education: there are a few OLPC floating around the iSchool… start putting the ‘developing’ part of ‘DRCA’ in practice with the OLPC? Or create content? Or work with organizations that are organizing OLPC pilots (like Waveplace, although I don’t know anything about it – but they seem to work with teachers too – see post about middle men, again)

7. health care: Hmmm. Here I actually don’t know what to say. The American health system seems to me very peculiar, and not a good example to export; and the projects we have heard about that landed on developing countries with an American health care system perspective were not good. So this gets moved to the legacy projects…

6. legacy projects, or: continue working on successful TIER projects. Aravind Eye Hospital – Soonesh has graduated, maybe they need more workers? Or the multi-mice project at Microsoft research, which, if I remember right, is now becoming a separate non-profit. Getting involved in development would be a good way to start a project here, and continue it with field testing.

5. connect with TIER/iSchool people who are doing field work while the semester is on here: Melissa Ho will be in Uganda, and might need remote developers; Janaki is doing her field work in India; eventually, I will go back to China. Clearly, these projects leave out what a lot of people seemed to be interested in, i.e. user requirements and user-centered design. But I don’t see how this can be otherwise, at least until the flying car and/or teletransportation are invented.

4. the middlemen: quite a few presentations this semester pointed out how the relationship between headquarters and field offices was difficult, and sometimes fraught with misunderstandings. Is there any way technology can make these relationships easier? There could be some user-centered design here (interviews and observation of HQ people), followed by summer fieldwork and fall development.

3. a very local project: build relationships with scholars, local NGOs and local institutes that are doing projects (development projects, not ICTD projects) in the field, and look if there are areas where technology can provide support or solutions. So, instead of having a solution and looking for a problem, try the reverse…

2. do some hands-on work so that everybody can learn the basics of, I don’t know, how to set up a wireless tower (including mixing cement!), or whatever other mix of low-high tech is useful to know in the field for both engineers and the social scientists. I am not sure what this would actually be, but there seem to be common threads among presenters…

1. just to avoid reinventing the wheel, do a training workshop where people learn how to train people when doing ICTD projects, and on what. Not ICTD at all, but it seems like one of the most useful tools that every ICTD practitioner should learn (and there’s plenty of training manuals online, including the unmissable human-elephant conflict mitigation at Eldis!).

This will take the entire semester to read, so I’ll stop here. A final suggestion for next year’s DRCA class – the audience this semester was quite varied, some people new to the field and just curious to find out what it’s all about, some wanting to learn more and move on to actual projects. Perhaps a way to reconcile the different sets of interests and goals is to offer the class on a variable units option – those who just want to come to the talk can take it for 1 unit, those who want to develop a project for 3 (and the middlemen for 2!)

a boring post

November 17th, 2008

This is going to be a hard post to write, because I think the Aravind Eye Hospital telemedicine project is great, is developing organically, and is using technology intelligently as a tool within its greater mission. Which is an encouraging way to end this class (gee, technology sometimes does have a meaningful role!), but makes for a very boring blog post.

All I can perhaps say is that Sonesh did a very good job of explaining the challenges of training people to support the technology, or at least to be able to report what is wrong with it. Although he was the most articulate of our speakers in detailing the challenges he encountered, these challenges sounded very similar to what other people reported: explaining technology to people unfamiliar with it, language (even when the language is allegedly the same – and he made a very good point about adapting to the local terminology, even when for us it’s unusual), turn-over of trained people, etc.

A module, or even just a couple of classes, on how to train people and on what would be an excellent addition to the planned ICTD year-long course. Maybe a meta-project about the projects we heard about this semester would be to identify the common challenges related to training and dealing with local authorities, and put together a training plan, to be accompanied by a specific ‘train the trainer’ workshop. We could even put together a fancy kit with pretty graphics and pictures, just like Ideo!

Middlemen are people too!

November 2nd, 2008

Many ICTD projects are centered around the elimination of middlemen from the chain that connects producers with consumers (or patients with doctors, etc). Case studies (yes, the ubiquitous Kerala fishermen for example) also highlight the increase in profits and decrease in costs when the producers have direct access to the market.


A majority of people in the developed world now work in the tertiary sector (primary = agriculture, secondary = industrial production). Silicon Valley is all about the service economy. We are all in the service economy. And the service economy is, essentially, the middlemen economy. We have reduced agriculture to a tiny, very efficient section of employed and GDP; we have moved most production to developing countries; we occupy, mostly happily, the sector in the middle, which has become a huge source of growth. So, wouldn’t it make sense to do the same in developing countries? In a developing economy, middlemen are in a very precarious position: they don’t control production, nor distribution. So, to survive in the middle, they have usually developed a set of skills that is mostly based on knowledge of circumstances, in knowing better than the farmers how the market works, and better than the distributors what farmers are producing. Can’t this knowledge be leveraged to introduce changes that will benefit the entire chain? If the middleman is the one with the ICT device that will improve agricultural output, he will have a direct interest in circulating the information at least through the circle of the farmers he buys from. No doubt, he will also have an interest in making sure that other farmers and other middlemen won’t get to know that information, but so do farmers, and they have a smaller circle of acquaintances that middlemen. Not that I have done any case study on this, but I think it’s something worth considering.

This was a jolly post for my standards, so I feel obliged to end with a link to a depressingly entitled article on the journal Development in Practice (behind UCB libraries firewall): “How international NGOs could do less harm and more good,” which features the usual litany of issues that people familiar with INGOs know: NGOs are accountable to donors, usually back in a Western country, not to local people or local governments (right, we already established those are evil, I forgot); they pay their staff too much compared to local wages, and therefore poach skilled people away from public sector jobs (I remember the UN in Afghanistan hiring teachers to be drivers, not exactly a way to reconstruct a country); they create dependency from foreign funds and foreign expertise, without training locals. So a key part of the sustainability of ICTD projects should be training – we’ve talked about this during this semester, but it’d be interesting to discuss how to incorporate that as part of the actual workplan of ICTD projects.

Headquarters never get it!

October 27th, 2008

Very interesting presentations last Tuesday, where we got to hear about technical applications, but also about business plans, leveraging of existing resources, the human element, and how HQ never understands anything. Oh, did that ring a bell… it’s the ultimate dilemma of anyone working for any type of decentralized organization, but it seems always more painful in the non-profit sector. The view from the periphery is: the center just doesn’t get it – they sit in their nice offices with fast internet connections in Starbucked cities, and expect us to fill their stupid forms/follow their stupid rules/make their lives easier, without realizing that their requirements are wasting our time and preventing us from doing our real job. The view from the center is: you want money for your project? Then you’d better start filling in the forms, because without accountability there’s no grant/no charity navigator ratings/no giving. And we haven’t found yet any way of separating accountability from forms of one kind or the other. In the non-profit sector, this is known as the ‘who’s your master dilemma’ (ok, I just made that up – the definition, not the concept) that has been explored by, among others, Ian Smillie (he of Mastering the Machine, the must-read book of all ICTD people).

The dilemma is: NGOs’ constituencies are [fill in who the NGO says it serves in its mission statement: the poor, the children, women, etc]. Lately, the rhetoric of ‘treat your constituency as your client’ has really taken hold, in an attempt to bring a concept of service, rather than mere charity, to the development field. So, the NGOs’ constituencies are in fact their clients, and NGOs, like any enterprise, should be accountable to their clients. Isn’t that the very basis of capitalism the modern world the service society what we should do? Second, less cynical try: if an NGO doesn’t do what its constituents need/expect it to do, then there’s no point, is there? But, argues Smillie: the poor/children/women do not audit how NGOs use their resources to do projects, nor do they complain if they don’t get what they are supposed do. After all, you don’t look a gift horse in the mouth. However, donors do look a gift horse in the mouth, and require follow-ups, reports, forms that show how the money they gave was spent, what kind of deliverables and improvements there have been and so forth. Which inevitably translates into a mountain of information that tries to make very complex situations understandable, logical and actionable.

Or, as an NGO worker we interviewed in the course of our sitrep project put it: “We had this money [from a donor] for recreational activities in schools, and I got a huge matrix to fill in with attendance records and all this information, so I spent all my time trying to get the school to produce nombers (note: this was in Sudan, a country not exactly known for its book-keeping practices). Enrolment went up x per cent, so recreation activities are good. The money for activities ends up being money for baseline studies, which they don’t want to pay for. All this looking for causality…”

NGOs compete for funds (and for clients, but that’s another story), and funds go to those who show they are using money well. There have been proposals to move the competition to the service-point delivery, e.g. have different organizations offering the same services in the same areas (instead of dividing up geographical areas as they tend to do now) and then let the clients decide which one offers the best services. This is the next logical step to the introduction of the concept of client, and one with which I rather disagree. When the market is brought in in certain areas, it can bring more efficiency, but it will certainly bring customer exclusion (some customers are too much trouble/too much of a minority to serve) and the elimination of certain services that may be effective but don’t have market(ing) appeal (will you choose the NGO that gives you a cell phone to learn English, or one that offers you a low-tech, terrifying teacher?).

Another post about a problem and not a solution. My last post of the semester will be bound to be a piece of genius that brings together the perfect solutions to the issues I’ve raised so far, all neatly packaged in a 300 words essay.

Oh, and just to go back to the initial topic of this week (last week’s presentations), i enjoyed them very much, especially the Bodas for life which provided a great example of leveraging existing technology and existing social relationships for a sustainable solution; and I’m in awe of people who go to a country they’ve never been before, and manage to tackle the problem they’ve been assigned, find a good and sustainable solution, and learn the local language in the meantime…

It’s all politics! (or: the gap in the middle)

October 20th, 2008

There were three threads that TIER skillfully wove into its workshop program: practical applications, a few of which have been presented in this class; policy eco-systems and infrastructure, or how do ICTD project fit in the bigger picture of state policies, economic systems, etc; and meta-ICTD, or what is ICTD research and how to do it successfully.

The part that was missing for me was the middle layer between the application created to address a specific problem in a specific place and the general political/economical/social system that surrounds this place and this problem. In other words, what happens between Matthew Kam’s Millee project and the educational decisions of the Indian state? Can ICTD intervene in the middle, where it seems that the system often breaks down – for example, can ICTD do something to address teachers absenteism, or their own lack of training, rather than target students?

Most ICTD project I am familiar with are focused on eliminating what is usually portrayed as the evil middle man – the distributor in agriculture, the teacher in education (OLPC for teachers, please?), the para-medical personnel in medicine – in order to empower the end-user, without reflecting much on what this middle layer could be adding to the process. This perhaps reflects what the workshop keynote speaker Tim Unwin described as the individual-focused approach of the US versus the communal-focused approach of Europe.

And this is connected with another trend I have noticed in many presentations, where projects that are commendably taking a user-centered approach often have a slide in the conclusions that says: “Don’t underestimate politics.” Short-hand for: people behave in funny and strange ways, and you need to be friendly with the mayor if you want your project to go through. Unfortunately, politics is often interpreted in a narrow way as elections, political parties, and C-SPAN, which represent only the mechanics of it. Politics is what should be at the basis of every ICTD project: why is there a need for development? What kind of development? Are we liberal democrats (individual freedom is the utmost goal of the political system) or social democrats (solidarity and equitable – note: not equal, that would be communism – distribution are at the center of the political project)? By creating an English-lesson plan delivered by cell phone to children, or by creating a telemedicine center that is hosted by the entrepreneur of the village, we are making political decisions about the type of development we are envisioning, not only technical decisions about what is the most effective way of delivering services. So maybe the ICTD curriculum should begin with a short introduction to political theories and development…

Evolution and revolution

October 13th, 2008

The idea that has really stuck with me from the designing appropriate technology talk is the idea of evolution – small, incremental changes to an existing system – versus revolution – a completely new system, new idea, new implementation. I find this a fascinating topic, because of its connection with creativity, and because of the important consequences it has for people who want to bring technology to environments that are utterly different from the technology ‘native’ ones. 

Last year, (fittingly, as I was an intern at the UN Global Alliance for ICT and Development!) I saw the fabulous exhibit Design for the Other 90% at the Cooper-Hewitt Museum in New York, which featured technological innovations mostly created in developing countries, rather than for developing countries. The inspiration for the exhibit was a comment by Paul Polak of IDE: “The majority of the world’s designers focus all their efforts on developing products and services exclusively for the richest 10% of the world’s customers. Nothing less than a revolution in design is needed to reach the other 90%.” Interestingly, most of the artifacts in the exhibit were not revolutionary, but rather evolutionary, and of a very pragmatic evolution: water filters, irrigation systems, pumps, latrines, etc. The One Laptop Per Child was featured too… and it looked pretty out of place… mostly because it wasn’t an evolution of existing ‘things’ that people use in the countries featured, nor a revolution that took the ideas and functions that a computer is based on, and revolutionized them completely to adapt to new places. (ok, this was my STS moment). 

Both evolution and revolution require fresh eyes, the capacity of looking at the ordinary and see a new opportunity. Being a fan of industrial design, I admire tremendously this ability… and am still ambivalent: is it a learnt talent, or is it innate?

Technology for technology sake

September 29th, 2008

I’m about to have another STS moment. I’ve been thinking about Tuesday’s presentation all week (with short pauses for eating and sleeping), and I hope that I didn’t get the point of the project, because if I did, then it was all rather depressing. As I understood it, the initial motivation was to make family planning more available and widespread in villages, to curb the natality rate that is a big problem in Uttar Pradesh. So, a franchising of telemedicine centers was set up, with the telemedicine center tools (mostly distance diagnostics) being the incentive for people to use the center, and therefore be instructed about family planning. But the family planning goal that was so much part of the initial motivation sort of disappeared in the implementation. And the follow-up study on the use of the center, even though preliminary and informal, did not focus at all on family planning. And the end result was that people used the telemedicine center, but for the most part got diagnosed, rather than cured; not much is known about family planning; nothing was said about the social dynamics of getting someone who had US$ 3,000 to invest in charge of the center (we were assured that they were all people already active in the community, but I couldn’t help wondering what caste and/or social class were these people, and whether or not this was an issue on the kind of people that they attracted as patients). Nothing was said about what kind of improvement – or even just change – there was compared to the situation before these centers came along, i.e. mostly a network of healers and pharmacists with varying degrees of training and a combination of traditional medicine and random Western medicines as tools at their disposal. I mean, it’s all private healthcare, since people have to pay at these centers and they had to pay whatever other ‘specialist’ they used before, and it’s all people untrained in Western medicine, who are however using tools typical of Western medicine. So if I understood all this right, it seems that the ethical problems are much bigger than the issue of giving people access to diagnosis but not, at least directly, to the cure; and that the importance of the local context has been taken into consideration, but only insofar as it fits with the technological agenda of the project implementators. In other words, the context didn’t seem to have been studied with the idea of ‘is there a problem – what is the problem – what is the best solution to address it’ (or even a short version of ‘the problem is family planning – what can we do to address this problem’). The context seemed more like ‘there is a lack of family planning – what can we do with telemedicine to address it’?

In an effort to go beyond the STS syndrome, I also tried to think how I would have approached the issue. I think that I would have started field work inspired by the idea that the goal is to see if technology has a place in the equation, and if so how it would fit in the environment, but keeping in mind that technology is only one of the possible tools. Then I would have targeted a specific group, rather than the village as a whole: who needs family planning tool (assuming that this is the issue we want to address), what do they do now, why, what kind of medicine/doctor (in a broad sense) they use, and why. And then see if family planning can be woven into existing practices, before escalating to a telemedicine center that doesn’t sound sustainable.

Incidentally, the sustainability is not only financial, but also on the supply-side, i.e. doctors. I have read only one evaluation of a telemedicine center, in Crete, that looked at the behavior of doctors on the diagnostic/operative side of a distance system, and it concluded that their in situ work took invariably precedence over the telework, which was considered peripheral and not directly significant for them, partly because of the lack of direct contact with the patient. Is it such a surprise that direct contact with other human beings would be an important factor in a doctor’s motivation? Are we envisioning a system where doctors become call-center workers? What is the social impact on the way doctors work that a franchise of telemedicine centers could have?

Mostly, it is ironic that just as Western medicine is finally realizing that curing the single illness without understanding the situation of the patient as a whole is not a solution and is more costly in the long-term (see the other study my sieve-brain doesn’t remember the details of, that showed that male doctors are more productive in terms of number of patients seen per day, but female doctors – who spend more time with patients – have a lower rate of patients returning for follow-up visits), we are also seeing the elimination of all human-contact as a positive development, as a poignant quote from the Telemedicine in Rural India article illustrates: ““In terms of disease management, there is [a] 99% possibility that the person who is unwell does not require [an] operation. If you don’t operate you don’t need to touch the patient. And if you don’t need to touch the patient, you don’t need to be there. You can be anywhere, since the decision on healthcare management is based on history and interpretation of images and chemistry.” All humanity is taken out of illness. Reminds me of something Tania Li writes in The Will to Improve: Governmentality, Development, and the Practice of Politics: “Questions that are rendered technical are simultaneously rendered nonpolitical. For the most part, experts tasked with improvement exclude the structure of political-economic relations from their diagnoses and prescriptions. They docus more on the capacities of the poor than on the practices through which one social group impoverishes another.” (p7). But this, as they say, is another story.


September 22nd, 2008

Last week I was speaking of motivations for working in this elusive ICTD field, and, right on cue, we had a guest speaker who was very clear about his own motivations to set up the Jhai Foundation and to focus, despite himself, on Southeast Asia. That, I appreciated. And his remark that 80% of everything that happens in his projects comes from locals and only 20% from the Foundation was also a sign of his commitment and intellectual honesty.

What I wasn’t sure about in his approach (but it’s a more general issue) is the emphasis on “finding the de facto leaders, mainly women, and work with them” – this has been a trend within development agencies in the past 10 years, but it can create more problems than it solves. Empowering people who are outsiders or in a subordinate position doesn’t happen overnight, and elevating them to positions that they don’t originally occupy in the local social hierarchy can mean that they don’t have enough power to influence the village, or to be agents of change. There is a very interesting reading that talks about a woman, maybe in Peru, showing other women that they needed to boil water before drinking it, but because she was of the wrong social class, they didn’t listen to her. (oh if only my memory wasn’t a sieve). This doesn’t mean that projects should always be run by the head village, that is an elderly male, but it means that one has to consider what is the downside of empowering the disempowered… which brings me to the next question I have: are local governments and institutions always the enemy? I know, local governments are more often than not the origins of a lot of problems, rather than the solution; and I am undoubtedly brain-washed by a class on development and modernization that I’m taking right now, where it’s the state and the macro-forces of history that shape the destiny of a country, but: can there be a compromise between these two approaches? (ok, back to the Science, Technology and Society corner now, where I can criticize everything without ever being constructive)

So anyway, what are my motivations to study China and ICTD? China: great book I got from my grandmother at Christmas (I was 8 or 9), with legends from exotic countries. China had the best, and I have been looking for dragons in streams ever since. ICT: all my science fiction readings promised flying cars by the year 2000 at the latest, and darn, I don’t see any around… I waited until 2000, but then I was too old to go to flying cars engineering school, so I thought that ICT at the iSchool was second best. Hey, I said that motivations had to be clear, not that they had to make sense…

Culture-neutral applications and the White Man’s burden

September 15th, 2008

To pick up where the conversation stopped in class and give it some more nuance: there are no culture-neutral applications, if nothing else because the same applications are interpreted differently in different cultural contexts. Researchers should be aware of this, but more as a background constraint, so that they can build something that is usable in a specific context, and then respond to how what they built is appropriated by local communities. Local conditions are a limitation, but also an opportunity for further development, and the specificity of the local can enrich the application in ways that would not happen in different environments, and are not predictable.

Which brings me to the second important point of the day, that is the “short guide to surviving and even thriving while doing field research.” The suggestions were very realistic, but the really interesting issue to me was how to teach – or, less ambitiously, how to make people aware of – the fundamentals of field work. There seemed to be a consensus that this is not class material, but rather something that has to be lived. I am ambivalent: it is true that learning in class that it is important to eat all the food that one is offered won’t diminish the impact of actually having to eat semi-live octopus, and that not having access to potable water is something that cannot be simulated. On the other hand, what should and can really be taught – or even just mentioned! – is that field work is not about oneself, and yet working in any kind of development project is all about oneself. Opaque enough? What I mean is that to be successful in the field, one has to forget one’s own fixations/pathologies/pet peeves and adapt to the local environment, as a good guest does anywhere. At the same time, one has to be clear that we (people from more or less privileged backgrounds working in developing countries) are there to respond to our own interests, not to save the locals, lest we go back to a modern version of the White Man’s Burden. I don’t want to discount the feeling of wanting to do good, but one has to be clear about one’s own motivations in order to engage effectively with local communities. This is an issue that comes up often when talking with development workers, and yet it is rarely – if ever – mentioned away from the field, and can come as a surprise worse than the live octopus…

Good intentions and development

September 9th, 2008

Neal Lesh talk touched upon all the issues that face people involved in development projects, regardless of whether they involve technology or not: the tension between foreign experts and local communities, the different expectations and different goals of communities impacted by development projects and ‘benefactors’, and most tragically, the unintended consequences of development projects, aka the meaning well and doing evil paradox. There are two points that I found particularly interesting:

  1. the Electronic Medical Records Neal project, I think in rural Rwanda. It wasn’t clear from the talk, but it sounded like the EMR were simply a digital copy of the paper documents – in other words, that the information containted in the EMR was no more (and perhaps no less) than the one in the paper documents, which are still the backbone of the system. This reminded me of Sorting Things Out, the book by Geoffrey Bowker and Susan Leigh Star, and its chapter on ICD (International Classification of Diseases) as Information Infrastructure. They point out that ICD, and in general classification and information systems built around diseases, are Western-centric, and they often have the undoubtedly unintended consequence of putting the burden of data collecting on the already scarce medical personnel in developing countries, who usually don’t see the direct benefit of their work. Later on in the book, they say that “New infrastructures do more than support work that is already being done. They change the very nature of what is to do work, and what work will count as legitimate.” (p239). It would be interesting to discuss in the course of this seminar how new computing applications change the organization of work in the communities that adopt them.
  2. the second interesting point, which seemed like the implicit thread of the “Establishing Relationships for Designing Rural Information Systems” article, is how to ‘translate’ concepts like Research&Development and Localization for the ICTD field. Is ICTD research, especially the applied research that we are discussing in this class, the equivalent of the R&D that a multi-national company would do, that is products developed in few labs/countries, and then adapted to different circumstances? That seems to be a reasonable way to eventually exploit economies of scale, but it also seem to go against the ‘from the field up’ approach that has characterized most of ICTD projects, in contrast with commercial ones.

The latter point is particularly interesting to me as I am beginning to define my own research area – so I’ll try to think some more cohesive thoughts about it in future weeks.