Measurement & Evaluation

A New Donor Code of Conduct

Donors are in an ideal position to stem the flow of poorly thought-out or inadequately planned technology-for-development projects.

It was Albert Einstein who once said that insanity was “doing the same thing over and over again and expecting different results.” He could well have been talking about today’s technology-for-development (ICT4D) community, which has a habit of trying the same things over and over again in the hope that someone might eventually get it right.

How we define success is, like most things in the development world, open to debate and interpretation. When called upon, most of us could likely name a few projects we’d define as successful, but donors often chase scale not success. The irony is that, when asked to name ICT4D projects that achieved scale, most people draw a blank. In “Innovation: The Missing Middle”—a recent paper by Dan McClure and Ian Gray—the majority of interviewees struggled to name a single ICT4D project that had gone to scale. The only two that got a mention were FrontlineSMS and Ushahidi, and there were questions even around those.

In two decades of ICT4D initiatives, why have so few gone to scale? What number would represent a reasonable return for the vast amounts of money spent and vast number of pilots that have taken place?

Attempts to understand this and then put things right are as old as the discipline itself. The academic community unpicks, analyzes, and critiques wider development efforts—often for it’s own purposes, it has to be said—and occasionally its findings even feed back to the practitioners who do the actual work. Literature also regularly emerges on the topic of ICT4D best practice—a voluntary code of conduct (of sorts) for people looking to deploy technology in their work. The problem is that most people ignore best practice because, well, you don’t really have to follow advice if you don’t want to, and few academic papers make sense to people outside of academia.

So where does that leave us?

Given that the vast majority of projects would never get off the ground without some source of funding, donors are in an ideal position to stem the flow of those that are poorly thought-out or inadequately planned. So, what if all major philanthropic foundations signed a Donors Charter that encourages much greater scrutiny of any technology-based projects they might be considering funding?

The charter would be available online, offering considerably more guidance and transparency for anyone looking for money for their project. Critically, the charter would require that project owners answer a number of set questions before funders consider their grant application. These might include the following:

Preliminary questions

1. Do you understand the problem? Have you seen, experienced, or witnessed the problem? Why are you the one fixing it?

2. Does anything else exist that might solve the problem? Have you searched for existing solutions?

3. Could anything that you found be adapted to solve the problem?

4. Have you spoken to anyone working on the same problem? Is collaboration possible? If not, why not?

5. Is your solution economically, technically, and culturally appropriate?

Implementation questions

6. Have you carried out base research to understand the scale of the problem before you start?

7. Will you be working with locally based people and organizations to carry out your implementation? If not, why not?

8. Are you making full use of the skills and experience of these local partners? How?

Evaluation and post-implementation questions

9. How do you plan to measure your impact? How will you know if your project was a success or not?

10. Do you plan to scale up or scale out that impact? If not, why not? If yes, how?

11. What is your business/sustainability model?

Transparency question

12. Are you willing to have your summary project proposal and any future summary progress reports posted online for the benefit of transparency and more open sharing?

Not being able to answer these questions fully and reasonably needn’t be the difference between funding or no funding—donors would still be allowed wildcards—but taking potential grantees through a process like this would serve three purposes:

  • It would force implementers to consider important issues before reaching out for support, resulting in a reinforcement of best practice.
  • It would help the donors themselves focus resources and dollars on better thought-out projects that are therefore less likely to fail.
  • It would help stop the vast amounts of replication, failed pilots, secrecy, and near-zero levels of collaboration with project owners in the field.

My belief is that the adoption of a charter would do more to put an end to the frustrations, problems, and inefficiencies people in our sector regularly write, talk, and complain about. We should make better use of all that negative energy, and we owe it to the people we seek to help to do everything we can to avoid not only making mistakes, but making the same ones over and over again. Not doing so would, in Einstein’s words, be simply insane.

Tracker Pixel for Entry


  • Karl Brown's avatar

    BY Karl Brown

    ON September 25, 2014 11:25 AM

    Very interesting ideas Ken. I would add ongoing maintenance/improvements of the solution as an important question, as well as licensing - I covered this topic in a bit more detail here: The only constant is change, so for any technology we develop and install, we need to think about how (and who) will maintain it for the long term.

  • @Karl - Thanks for the comments, Karl. My focus with the initial 12 questions were more geared towards sense-checking projects before they get started, but you’re right - adding one that asks how they plan to support, maintain and improve the product once it’s out makes sense. As a donor yourself, it would be interesting to hear whether you thought something like a Charter was remotely possible.

  • BY Joel Selanikio

    ON September 25, 2014 02:21 PM

    ICT4D is centered on consultant- and pilot- and training-based (CPT) technologies because that approach is lucrative.  Asking people to sign a charter that goes against their economic interests is unlikely to be successful in changing anything.

    Luckily, the ICT4D providers stuck to that model will be driven out of that business by new organizations with completely new, Silicon-Valley-inspired business models and technology.

    As you and I have discussed many times, Magpi is one of the first examples of this approach. Magpi has more than 33,000 users worldwide, 99% of whom pay absolutely nothing for the service—which requires no consultants, programmers or training.  Even our paid accounts are a tiny fraction of the cost of the CPT tech equivalents.

    And the number of development organizations (or donors) willing to waste tens or hundreds of thousands of dollars paying programmers to set up some data collection system half as good as a free Magpi account are fewer with each passing year.

    I am absolutely sure that similar approaches will eliminate the CPT approach from mobile messaging and every other area of useful tech for development.  And good riddance, too, as the money saved may be applied to something more useful—like vaccines or school fees.

  • Thanks, Joel.

    I agree it’s going to be difficult to change old habits. Perhaps natural selection will prevail in the long term, but leaving it to chance is not an ideal option. A Donors Charter is one way, in my view one of the better ways, of beginning to proactively address the issues, but donors have to listen and engage. It’s worth a try, at the very least.

  • Gentlemen, this is what you seek:

  • BY peter van der linde

    ON September 26, 2014 04:19 AM

    Thanks for sharing! In my view the biggest issue with the adoption of it tools at scale, is the limited capacity that is available locally to use them optimally. And a lack of true understanding with funders. Serious ict for development tracks need serioous investments. Not in tools, but in a supporting capacity. Many governments we work with have very limited IT capacity available-when you look at nationwide exercises that is the bottleneck. How do we jointly make sure the ownership sits there sustainably. We should obviously make sure we avoid duplication of efforts in tool development, but should be carefull not to state that free tools are the solution. They are a small part, the smallest part-even.

    When chosing tools, we often see that donors make featire checklist. But they overlook elements like ‘who backs them’, is it sustainable? Open source? Do we want support (not for tool but capacity building), in the local language? Do we know what we want to do with information that is collected? Is the capacity there to analyse it? If the tool is free, how is continous development guaranteed? What is the business model, etc..

    Hope this is helpfull

  • i thinks its time for kickstater model of donor funding or some form of peer to peer donor funding by individuals in 1st world countries to 3rd world countries

  • Valencia, please see Global Giving and Give Directly.

  • @Wayan - Thanks for posting that. I did get pointed to the Principles the day after I’ve put together the Charter, and I’ve linked to them from the site. If donors have signed up to your principles (is there a list anywhere?) and they genuinely apply them to their grant-giving, then that would be amazing. It would see the end to funding replication, would encourage openness, collaboration and an approach which sees people with the problems being excluded from designing the solutions by the vast armies of outside consultants. Of course, we’ve all been saying this for years, and little has changed, so let’s watch that space. The proof is in the doing.

    @Peter - Thanks. It’s good to hear your thoughts.

  • @wayan There are inefficiencies in the current donor model, a lot of resources are wasted in the value chain that could benefit the people who are actually on the ground

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 07:30 AM

    @Valencia is another model, for impact investment overseas, and then there are other crowdsourced models like Kiva, etc. Otherwise, remittances (e.g. p2p capital transfers) today dwarf the official aid that is provided (,,contentMDK:21924020~pagePK:5105988~piPK:360975~theSitePK:214971,00.html) Many people are now studying how to leverage these remittances towards larger goals, and finding ways to iron out inefficiencies in those capital flows. But the remittances are happening, and if you want to move away from the donor model, find ways to give directly to groups or people on the ground doing useful work. Unfortunately if you live in the US, IRS regulations make it difficult to take a tax deduction for charities based outside the US (there are intermediaries however who will help handle this, for a fee).

    @Peter, I agree that capacity building is necessary, and at RF we supported the creation of several university programs to train the next generation of health informatics experts, and supported for example hiring and staffing of an eHealth unit in the MoH Rwanda. But I disagree that “free tools” aren’t important - what “free” means is not free of license fees, but the freedom to modify the code (e.g. open source). That means you aren’t locked into a particular vendor. Ultimately, you need to look at total cost of ownership - with a proprietary solution, what are the ongoing license fees and maintenance charges, vs what is the cost to train up local staff or hire a local technology firm who can do ongoing maintenance of a FOSS codebase? The answer isn’t always open source and it isn’t always proprietary - support, likelihood of changes, ability/flexibility to adapt to local languages, etc - these all weigh in. If a government needed a good spreadsheet tool I would have no hesitation recommending MS Excel; OTOH if they need a national health information system, is it better that the core platform be proprietary with expensive license fees and upgrade costs, or open source and let the local resources manage maintenance and ongoing support? We’ve had government representatives tell us that they preferred open source BECAUSE they see it as a route to local capacity building, because they have control and ownership, in a sense, of the codebase. And it’s not one or the other either. In any relatively sophisticated system, you will find proprietary and open source components. The iPhone and Google’s search algorithm are both full of FOSS. There are no easy answers.

    @Joel, I do think SaaS / Cloud-based models are very useful in some cases, and I think the expansion of Magpie is great. But it’s not clear whether SaaS models could work for information systems at a higher level of complexity - e.g. something like national level logistics management/supply chain, or national-level enterprise architecture platforms for health information systems.

  • peter dunn's avatar

    BY peter dunn

    ON September 26, 2014 07:32 AM

    Please see, although it is only a German based donation platform, it is the only existing donor platform the is 100% transparent

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 07:34 AM

    @Peter, cont’d, I also think hybrid models are quite interesting. For example RedHatLinux, which provides compiled/tested/robust/supported/warrantied version of the open source Linux operating system - but you pay for it, and you get support on it. There’s also the freemium model, or the “enterprise version” vs “community/FOSS edition” approach, which mixes proprietary/license approaches with FOSS. Ultimately I think for FOSS in global heatlh, this is the way to go - we need companies that are willing to build up support and maintenance infrastructure around these core FOSS tools, so that you won’t be locked in to a single vendor and can benefit from cross-country learning and innovation, but you can still assure a level of professional support.

  • @Karl - Thanks for chipping in on this. As someone with many years experience in the donor world, they’re much appreciated.

  • BY Joel Selanikio

    ON September 26, 2014 10:22 AM

    Hi Karl (good to hear from you),

    Re “it’s not clear whether SaaS models could work”, we are at this point very clear, as Ken indicates that almost all ICT4D tech up to now is very definitely not scaling, or working at all except to generate nice press releases.  So I’d say the onus is on the doubters (and the donors) to explain why we shouldn’t be trying this more.

    And it’s not just SaaS models that I’m talking about.  Even without the SaaS business model, there would be great benefit in simply utilizing the cloud/internet more to effectively share the costs of servers, etc.  Those in the SaaS business do that to be able to give away service, or lower costs, but that’s not the only business model that would work.

    As an example, I can’t imagine why any ministry of health should own and operate its own servers at this point, any more than they should operate their own cellular service. In my experience not only is it an egregious waste of money (for the technicians, A/C, etc) but they’re quite often failing.  MOHs don’t need to move to a SaaS model to just shift their data storage to a shared, rented space.  And that space could be Rackspace (like we use) or Amazon, or a regional provider.

    Regarding the “FOSS” business model, Red Hat is the ONLY completely open-source company I know that is economically viable. At some point we need to just acknowledge that the “all our code will be open source” model doesn’t work economically, and move onto what EVERY software company uses (including ours): some of our code is open and some is closed.  More on the uniqueness of Red Hat:

    Note also that companies that make a substantial portion of their income from training people how to use their software are severely dis-incentivized to simplify that software to the point where it doesn’t need training.

  • Donor programs in the ICT space should focus on 2 things
    1. Developing local skills capacity in ICT (programming, electronics e.t.c)
    2. Supporting tech innovation & entrepreneurship amongst those with the ICT skills through entrepreneurship competitions & entrepreneurship meetups

    *As soon as we have more local people with ICT skills they will develop solutions to meet local needs especially if the entrepreneurship is encouraged through it competitions and having local technology blogs

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 10:52 AM

    @Joel, I have no dispute with your points re: MOH owning servers; however often times there is a desire (or an obligation) to host data in country, but for that purpose of course generic data centers, when they exist, should be used. I don’t know if people are regularly proposing that MOH own their own servers. That said, the Thai National health Security office, which has a fairly mature and robust IT system for management of millions of claims, manages their own server farm - should we critique their decision? I’m not ready to do so, as I haven’t looked at the math and their alternatives. Sometimes the data center infrastructure, costing, access, etc is not an ideal match. Our best path would be to provide Ministries with the best advice, but then empower them to make the decision about where their information/services are hosted, as there are many options. I’d also point out that with the recent security leaks from Apple’s photo cloud, people are, I think, rightly concerned about security of cloud-based infrastructure; there are many roads to Rome.

    re: hybrid models, I also agree, and mentioned that in my post above - the winning models will likely have some sort of hybrid nature (e.g. licensing fees for compiled binaries, some “enterprise features” which are kept proprietary, etc). I don’t think anyone is proposing that RedHat is going to kill Microsoft (or that OpenMRS will kill Epic), but that’s not the point! If proprietary software firms arrive with amazing, easy to user/configure/modify solutions for community health workers in rural villages in India, then let the market select their software. I just don’t see these BOP markets as being very attractive to the big tech players yet, especially the sort of sophisticated solutions that in-country stakeholders are demanding.

    IF it does become attractive, then I also believe that private capital should fund such private software development, and if you look at the many startup accelerators and investors around the world, capital is flowing into technology, and some of that tech may focus on populations we care about. If so, great, but donors don’t need to get into that business unless it’s through impact investment. Donor funding should ideally be targeted at social good, where the social benefit outweighs the private benefit. This to me translates to, at least in most cases, a strong case for making such new development open source, so that other donors, and other NGOs, and other governments, can use and share in that technology.

    There will remain - IMHO, for some time - a viable middle ground in between simple, easy-to-configure tools (that could be proprietary or FOSS), and enterprise-scale proprietary systems sold to middle-income countries - that middle ground is for relatively sophisticated IT systems that are targeted at populations or countries that big private sector firms are not interested in. That is the sweet spot where FOSS solutions have had impact to date IMHO. But there will never be a purely open source system in health - as there will always be add ons (e.g. analytics and visualization packages, robust commercial database systems, even the OS on the laptops). Nonetheless, we should not underestimate the utility of FOSS in this domain, there is still room for FOSS and we should encourage it when it makes sense.

  • BY Joel Selanikio

    ON September 26, 2014 11:02 AM


    We definitely have a lot of common ground here.  A few points, though:

    1 - If the Thai ministry wants to run its own server, and can do so, that’s great (as long as my tax dollars aren’t paying for it).  In most poor countries I’ve been to, on the other hand, ministries do a terrible job of running that stuff—AND it’s always more expensive.  So if they’re paying for it, it’s up to them, but it would be great if donors stop subsidizing inefficiency.

    2 - As the to desire/obligation to host data in-country, that is a luxury.  Again, if a country wants to do that, and can pay for it, great (just as if India wants to spend money on Mars orbiters while kids are growth stunted, that’s their decision)—but I hope the donors can stop allowing countries to choose expensive luxuries like that.  No donor would agree to fund Maseratis for distributing vaccines (or a Mars orbiter), even if the country expressed a desire for them, or indeed insisted that Maseratis were an obligation.

    3 - As to proprietary software firms arriving and providing solutions, have you considered that philanthropic support for all the failed ICT4D projects of the past 20 years has itself crippled the market for private solutions?  Anyone who funded such a project (or continues to do so) is in effect dumping *apparently* cheap technology just as handing out second hand clothing can strangle a local clothing industry in the cradle.

    So I know donors can’t necessarily make things better. But they could start by not making things worse.

  • @Valencia - I’m with you on all you say. Technology has given people in developing countries/emerging markets a great opportunity to solve their own problems, but we have to let them. I wrote a little about this for BBC Future a while back:

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 11:19 AM

    @Joel, it’s a tricky line to walk. “but I hope the donors can stop allowing countries to choose expensive luxuries like that.” - yes, I agree, we would never agree to fund Maseratis, but plenty of donors have bought the classic “Project 4x4”.  But a grant is not a contract (even though some donors behave like it is, I’ve never perceived of it in that way) - a grant is a gift, and the best grant IMHO is an intersection of something the grantee wants to do anyway, with something that helps you, as a donor, pursue your overall strategy for impact.

    It’s an imperfect art, but it’s not as easy as saying “No, we will only pay for hosting in Florida”. We are not project managers, we are portfolio managers, and ultimately the decisions about how the money is spent once the grant starts comes down to the local management by the grantee (Indeed, we are restricted from even interfering in this process by IRS regulations - we cannot, for example, specify that a grantee must contract with entity Y to do their hosting - it’s called earmarking and we’re not allowed to do it).

    A grant with a foundation is a huge difference from IBM contracting small entity X to do work Y, both in intent and in the legal and normative structures that surround it. The Ministry of Health X is not a contractor, fulfilling the wishes of the Foundation, nor should it be. They have their own goals, their own approach, and we must respect that - even if they end up making choices that we may not agree with. How far we’re willing to support decisions that we don’t necessarily agree with is a very delicate process - but I always think, who is closer to the problem? What if I’m wrong, and they’re right? There is no simple solution here, between country empowerment and doing what you as a donor think is right.

    Some donors are more controlling than others, of course, but that leads to additional reporting burden on grantees. So it’s a balance, ultimately you want to give a grantee a sense that they are driving, you’ve agreed together on the eventual destination, you are putting gas in their tank, and there are some lines on the road they should not cross (like buying Maseratis).

    As to your last question, I don’t know the answer to that. I have seen in person, and heard stories, of proprietary vendors acting in a predatory fashion with certain governments in Africa for example. There’s just more potential for abuse when you have a set of decision-makers who don’t have a sophisticated understanding of software licensing, ownership, etc, so when donors directly support proprietary vendors (as many have), they are setting up these governments for vendor lock-in which can sometimes be very problematic.

  • BY Joel Selanikio

    ON September 26, 2014 12:15 PM

    Karl, all valid points, thought I think you’ve got more control and more conditionality than you let on. Or you could have.

    And whether there are private companies that are unscrupulous has no bearing on whether donors dump apparently free but disastrous technology on those unsophisticated decision makers.  That free tech the donors have been paying for hasn’t been the slightest bit better than that provided by some unscrupulous company. 

    And, indeed, were donors not squashing the private markets for technology by funding misguided projects (essentially all of which later collapse) there would be more companies and more competition and service would improve and prices would drop—regardless of scruples.

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 12:19 PM

    @Joel: “That free tech the donors have been paying for hasn’t been the slightest bit better than that provided by some unscrupulous company.  ”—citation needed. It’s not black and white, and there are plenty of examples of useful software produced that is making a difference.

    I think more private sector engagement in these markets would be excellent, and I think it’s starting to happen.

  • BY Joel Selanikio

    ON September 26, 2014 12:50 PM

    Sorry, Karl, not providing any citations.  I’ll quote Ken’s article:

    “In ‘Innovation: The Missing Middle’—a recent paper by Dan McClure and Ian Gray—the majority of interviewees struggled to name a single ICT4D project that had gone to scale.”

    If you’re still thinking that ICT4D is making any goals, we can’t have a serious discussion about it.  And that’s twenty years of watching projects fail—while the techies who did them become acclaimed “experts”—talking, brother. wink

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 01:01 PM

    @Joel, that article was more about scaling technology innovations. And that said, I think it’s a fairly terrible methodology - e.g. “We asked a few people we knew, and few of them could name more than a few examples” - in terms of health technologies that have scaled there are a number, a very famous and simple one being ORT in Bangladesh, to say nothing of vaccines or other medical technologies. In terms of digital tech, you have famous examples like mPesa (originally funded by DFID), or examples like DHIS (now running national-scale HIS in a number of countries) or OpenMRS (with several million patient records across dozens of sites), or even Magpie (which is, or was, I believe, donor funded at some point). I think it’s quite possible there’s a power law, whereby the bulk of pilots and tech approaches will fail yet some will succeed, and it’s also quite possible that the failure rate is higher than commercial software investment, but more robust study is needed beyond “Uh, i can’t think of any”. Bhoomi is a relatively low-profile and earlier example, but it has demonstrated impact for the specific domain it focuses on.

    In any case, many of the critiques leveled at ICT4D initiatives are fair and I think as donors and implementers we can all do better. I think Wayan was right to point to the ICT4DPrinciples, which I’d like my organization to sign up to, as one example of a positive forward approach.

  • BY Joel Selanikio

    ON September 26, 2014 01:03 PM

    OK, Karl.

  • @Joel @Karl - Thanks for the lively discussion and for taking the time to share your thoughts.

    There’s plenty up for debate, but one thing I think we can all agree on is that, overall, investments by donors in ICT4D initiatives have to date given a very poor return, and that the status quo of outsiders coming up with ideas and then trying to impose them on end users in developing countries doesn’t only fail most of the time, but it works to the detriment of local solutions providers.

    As far as the Principles go, I don’t yet see how these are going to work in practice, and don’t quite see what’s changed, other than that someone’s put them all in one place. I thought we were supposed to be doing many of these things already. If donors are signing up to them, I’d like to see how that influences what they fund, and how that impacts best practice for practitioners.

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 01:35 PM

    @Ken, sometimes changing institutional practice is like redirecting an ocean liner. You tweak the rudder but the ship only changes direction a few miles later. Remember, it takes a lot more than one enlightened program officer at a major bilateral or multilateral to shift institutional approaches on funding technology. But actually agreeing on these principles, which were first elaborated at a GreenTree meeting several years ago, is an important first step IMHO.

    As to your second question, the same issues would apply to your charter! Ultimately, it is normative approaches that are most likely to work well here - e.g. endogenously developed norms that “enlightened” members of the donor community pressure their peers into adopting, which then translate into changes in contracting and grantmaking practice. Merrick, I’m looking at you! smile

    “the status quo of outsiders coming up with ideas and then trying to impose them on end users in developing countries doesn’t only fail most of the time, but it works to the detriment of local solutions providers.” - I think that’s a pretty harsh critique and doesn’t really match to reality, which is much more complex, and I think you know this.

    Use of user-driven requirements generation and agile methodologies is gaining traction, and that leads to the opposite of “outsiders coming up with ideas”. Indeed, the argument made by FOSS proponents is that low-friction nature and ability to make changes locally (even fork the whole codebase) makes it more likely that local players will adopt and tweak a solution to their specific needs, rather than having to call the solution provider (which in many cases is will not be a local entity, due to the ways large organizations contract).

    I remember vividly visiting the “tech” room at the PIH hospital in Rwinkwavu, a few hours outside of Kigali - the OpenMRS team from PIH was on site, with the doctors, making changes in real time to the solution. This was not outsiders talking to insiders - this was people on the same team - clinicians and technologists - making adjustments, adding new queries, all in response to specific needs on the ground.

    Also, FWIW, If we take the “let’s create local solutions” idea to an extreme, it means each country - including very poor countries - must come up with their own web browsers, database engines, and so on and so forth. Obviously, that’s ridiculous, and no-one says that suggesting a web interface is an imposition on end-users nor that bringing MySql or Apache is a form of digital colonialism. What’s more interesting to me is, the fact that there are commonalities across countries even as you go up the service stack, and these commonalities can be potentially be exploited to get economies of scale and amortized costs of deployment. Joel takes advantage of that in designing a generic data collection platform, and you take advantage of those commonalities in designing a generic FrontLineSMS. This amortization and productization (which can itself happen “in country”, it needn’t be written in the leafy suburbs of Boston!) must be balanced against the need for a fully custom solution for every district health office. Again, it comes down to total cost of ownership and fit to function.

  • @Karl - I’ve got a headache.  raspberry

  • BY Joel Selanikio

    ON September 26, 2014 02:32 PM

    ICT4D Glossary

    in-country stakeholders
    impact investment
    agile methodologies
    crowdsourced models
    low-friction nature
    major bilateral
    enterprise architecture platforms
    grant making practice
    predatory fashion
    specific domain
    requirements generation
    hybrid models
    service stack
    BOP markets
    amortized costs of deployment
    fit to function
    solution provider
    power law
    digital colonialism

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 02:35 PM

    @Joel, ouch! smile

  • BY Joel Selanikio

    ON September 26, 2014 02:35 PM


  • BY Joel Selanikio

    ON September 26, 2014 02:36 PM

    (this would be a lot better over beer)

  • I’ll buy the beer (and popcorn)!

  • Karl Brown's avatar

    BY Karl Brown

    ON September 26, 2014 05:15 PM

    Not exactly ICT4Dev, but a broader look at technology in development, including business perspectives:

  • The interesting thing about this whole discussion, other than the fact that it’s obviously interesting, is the lack of consensus on some core ICT4D issues. It would have been nice to have some other voices join in, but open debate is sometimes another challenge we have.

    The question of what we can practically do to fix long-standing problems still remains. My answer, as a starting point, was to take the obvious step - stop the flow of funding to poorly thought-out and planned projects. Only donors can do this. The Charter was my starting point. In my mind, it would clearly make a difference if widely adopted and strictly followed, and it could evolve over time.

    It would require some major shifts. As Karl rightly puts it, we’re trying to turn a cruise liner around here (although it’s more like an oil tanker if you ask me).

    What would your first step to resolving many of the issues be? And I mean, a practical step which would without any doubt begin the process of change we so badly need?

    Anyone? One thing, achievable.

  • BY Joel Selanikio

    ON September 27, 2014 07:50 AM


    Again, I don’t think that your suggestion—of signing a piece of paper—IS the “obvious step.”  I think that this promotes the fiction that philanthropic donors are in any way a significant source of funding for technology, when they simply aren’t.  Or that they are central to development of useful technology for development, which they aren’t.

    It’s like Bill Easterly said about “development aid” in general: it’s a drop in the bucket compared to the budgets of even poor-country governments. For ICT4D funding, compared to the amount of money being spent by commercial companies developing and purchasing tech, the disparity is even greater.

    We don’t need to change what the donors do, because what the donors do with regard to technology doesn’t matter.  Donors didn’t produce any of the most useful technology tools for those fighting poverty: Google Maps, text messaging, email, Facebook, twitter, mobile phones, etc.

    If you want to produce software that helps those fighting technology: have an idea (hopefully influenced by deep understanding of the activities of those you’re trying to serve), imagine a business model and a pricing model that will:

    1 - keep your lights on
    2 - make prices affordable to those who need your technology
    3 - will manifestly save time or money (or increase productivity) for your customers
    3 - give you enough revenue to keep improving the product

    and then start providing it.

    Donors may or may not be a source of seed funding, but since the costs of memory, servers, internet access, programming have all plummeted, maybe you don’t need donors for that.  Tech operations that cost $100,000 twenty years ago costs $5000 today (or might be free). 

    Think about it: if the cost of getting your idea was $5000 and not $100,000 . . . would you be talking about donors at all? No.  So start developing the idea, try to find a customer who’ll pay for it AND provide useful feedback (something no donor has ever been able to do, since they’re not users of your software), and make it happen.


  • @Joel - All going points. In response:

    1. Agreed that tech products funded by donors may be insignificant in the greater scheme of things, but when they fund poorly-planned projects it’s often the target communities who suffer. Local solutions providers may also suffer, if they’re pushed out of the market by these outside ‘experts’. And when the donor funding results in 100’s of SMS platforms or mobile data collection tools being available, regardless of their quality that still muddies the field for the good ones.

    2. With costs, agreed - if your idea only costs $5,000 to develop then you won’t go to a donor (or are at least far less likely to). But a $5,000 project won’t have hordes of experts jumping on planes as part of its delivery plan, and in my experience people who build things with very small initial budgets tend to live off passion and adrenaline, and do it because they believe in it. These projects start off small, realistic and only grow if they get traction. Those are the ones we want.

    3. Regarding the “signing of a piece of paper”, right now everyone verbally agrees to best practice - they certainly all say it at conferences, and it certainly gets all the tweets. But even then, very few people follow it. Signing something would represent a step forward from that (for what it’s worth) but as I’ve said the Charter has to be acted on, not just supported. The proof is in the doing.

  • @wayan I’m still interested in a few things about the Principles you’ve been talking about:

    1. Who are the principal audience? Is this to remind solutions developers what they should be doing? Or for donors to sense check proposals?

    2. Are they going to be enforced in any way? If not, what’s different about this than all the other sets of ‘best practice’ we’ve seen over the past decade?

    3. Who’s ‘signed up’ to the Principles, and what does ‘signing up’ actually mean?

    4. I’m curious who else was consulted beyond the giants of the development community listed on the site? There seems to be a lack of any grassroots voice, or any of the smaller organisations who probably have a lot to share from their experiences.

  • BY pertti lounamaa

    ON September 28, 2014 03:03 PM

    One criteria for scaling as a measure for success seems to be missing from the discussion so far:  Is the user organization ready to scale? 

    We have worked both with commercial cloud solutions and with FOSS and although it is an important choice with a lot of implications which one is used, essentially both can either succeed or fail depending on whether the target users are ready to scale. The readiness depend obviously on whether there is a real need for the solution, which in general is not an issues. A second obvious question is the education level or general competence in using ICT, which can be addressed by suitable training. The difficult question is the general maturity of the organization and its processes.  In many developing countries the organizational context is simply so immature or even chaotic that continued systematic use of a new ICT solution is just very unlikely to take place.

    To put this in another way, perhaps donors should first ensure that sufficient long term funding exists for the overall user base to continue working in a systematic and sustainable way. Only in such contexts can ICT scale.  A good example of where this is happening is lower income countries rising to middle income countries and thus being able to build more mature organizations e.g. in health care and education, the typical target user groups for ICT4D.

  • BY Joel Selanikio

    ON September 29, 2014 07:51 AM


    Unfortunately, regarding FOSS, you are ignoring issues of cost.  Commercial solutions, by definition, have a business model that provides revenue—revenue needed to improve and support the product. Other than Red Hat, no FOSS organization has ever been able to create a viable business model other than one based on training and support.

    Unfortunately, if your revenue is derived from training and support, you are dis-incentivized to simplify and thus improve your product (and thereby reduce the need for training and support, and thereby reduce your own revenue).  FOSS software, as a result, is almost always more complex to use, and much, much more likely to require expensive programmers and trainers to learn it and maintain it and use it.

    This is a critical difference with grave implications for cost and therefore sustainability.



  • Karl Brown's avatar

    BY Karl Brown

    ON September 29, 2014 08:08 AM

    @Joel, I’m not sure where these broad generalizations come from re: FOSS and viable business models. I think comparing FOSS to commercial software in terms of business growth and market cap is a poor way to understand how software is built and used and what impact it has, and very much underestimates the power and utility of open source. The market cap of Apache is probably quite small (actually, it has NO market cap!) - the Apache Foundation runs on about $500,000 year in expenses ( and yet Apache captures ~65% of the market share globally for web servers. ( Other open source projects that have had significant market impact or are market leaders in their domain include wordpress, joomla, drupal, mongodb, hadoop, mozilla, to say nothing of core internet infrastructure components like sendmail (harder to measure, but likely north of 15%, most surveys put it above Exchange) and BIND. The notion that FOSS == expensive programmers and trainers isn’t borne out, but it is true that the closer you get to the core stack, the more companies globally seem to prefer FOSS. It’s also true that FOSS licensing allows and encourages much better collaboration across company boundaries, vs competition - which is what we need more of esp with cash-strapped health ministries trying to manage 1,000 competitive flowers blooming. I do agree hybrid solutions will dominate, but when putting together core health infrastructure for a national government, open source should be given serious consideration.

  • BY Joel Selanikio

    ON September 29, 2014 08:21 AM

    Sorry, I should have been more clear.  I generally draw a distinction between two types of “open source” approach, and unfortunately, the tendency in the development community is to NOT make that distinction clear (or, perhaps, to be unaware of it):

    1 - open-source sharing—the idea behind this is that open source can often be a useful component (usually a background component) of any software. Almost all commercial software has open-source components (Apache is a good example); many commercial manufacturers, such as Apple, have some part of their code that is open source.  Like sharing of property in general, most people like to share some of their code, but not all of it—just as they think sharing is nice, but also believe that they should be able to own private property (whether that is computer code or something else).

    2 - open-source communism—the idea behind this is that there are unique and wonderful benefits to making ALL computer code available to anyone for free. Those purported benefits include “sustainability” and “scale”, though in my experience those benefits never materialize.

    - - -

    When I talk about FOSS, I’m talking about software created by the latter group, the open-source communists.  Again, aside from Red Hat, I don’t know of any examples of a sustainable business model for this kind of software (where all the source code is available for free)—and as I mentioned, there are grave problems with software providers who make most of their money from training and support (or grants).

    Obviously, group #1 (which includes basically everyone else in technology) is capable of creating sustainable business models.

    Thanks for pushing me to clarify.

  • BY Joel Selanikio

    ON September 29, 2014 08:34 AM

    Karl, I should have noted one other thing. You state that ” It’s also true that FOSS licensing allows and encourages much better collaboration across company boundaries, vs competition”.

    This isn’t true at all: the main method used in commercial technology today to encourage collaboration is the use of APIs (by definition, sharing only a small part of the code).  Thus Yelp (and Magpi) use Google Maps API to produce maps, etc—without the need for Google to give us ALL their code (Google likes to share some things, but they’re not communists).

    Open-source communism (FOSS) doesn’t encourage collaboration, because great programmers don’t wish to collaborate on software products that have no business model, and therefore which cannot reward them for their work.  This “tragedy of the commons” has been well-documented in economic theory.

  • Karl Brown's avatar

    BY Karl Brown

    ON September 29, 2014 08:39 AM

    @Joel, APIs are not the same thing as collaboration towards enhancing a single codebase, but, the comments section here is getting a bit too long to debate the details! gives a glimpse, showing commits from ~91 different contributors, wh0 represent likely over a dozen different organizations. That’s the bounty of the commons! smile

  • See @Joel and @Karl, this is why we need you two to sort this out in front of a live audience that can join in the conversation. Much more fun than these empty lobs without context.

  • BY Andy Narracott

    ON September 29, 2014 10:28 AM

    @Ken - thought provoking article. Many thanks.

    My question on Twitter (@AndyNarracott) was whether this would stifle innovation. I agree there should be more transparency in grant giving and building on the successes and failures of others. In practice I think this would be very difficult. If donor-funded innovators were to post up their ideas ready to be shot down by the naysayers or ‘experts’ then this would do more to stifle innovation than stem the flow of poorly thought-out or inadequately planned projects. 

    Especially with the move to venture capital approaches by major donors ( where 80% are expected to fail in the VC world, not through lack of planning at the outset, but how they react to the numerous barriers faced by any promising solution on the journey to scale.

    However, if the code of conduct platform was to build on the ideas of others in collaboration and mutual respect, then there you have a powerful platform for generating high potential solutions to poverty. Especially if it could funnel the experience from around the world on an ongoing basis, rather than just at the outset.

  • @Andy - Thanks for the comments and for engaging in the discussion.

    For me, posting details of projects online isn’t primarily so they can be shot down by ‘experts’, but so people doing similar work can learn what others are doing. The current system is too closed, leading to replication and missed opportunities for collaboration.

    With the Charter itself, its main purpose is to get projects to sense-check whether or not they’ve asked themselves the right questions when designing their project, and by doubling up as a check-list it helps donors make sure they’ve done so, too. If you’re not looking for donor money then you obviously don’t need to do any of this, but most of the bigger, failed projects need buckets of money to get going, so they’ll have little choice.

    I don’t think a Charter would stifle innovation, either. Most of the really interesting stuff in the aid world is happening ‘outside’ of the aid system anyway, and often by engaged individuals and entrepreneurs who have personal experience of the problem they’re trying to solve. If they don’t go the donor route then they, by default, need to keep costs low, know what they’re doing, and be very committed for the long term. Many friends who I’ve met over the years started off with no funding. FrontlineSMS, which forms most of my own experience, didn’t get funded for two years. It says a lot for people, and their projects, if they’re still around after a year or two without financial support. And if these people do fail, they’ve not wasted millions of dollars in the process.

    I covered a lot of this ‘innovation outside the system’ in my recent book, “The Rise of the Reluctant Innovator”. You can find out more about that, if you’re interested, at

  • I just wanted to quickly jump in to say a big thanks to everyone who’s joined the discussion so far. We’ll struggle to make any progress if we don’t have these kinds of open, honest conversations.

  • I have not gone too deep in the issue of ICT 4Ds but i will more time is required to collect mistakes where necessary

  • BY Tony Roberts

    ON October 2, 2014 07:10 AM

    Ken, so if I follow your argument correctly, you are you saying that all ICT4D is useless (except your own FrontlineSMS and Ushahidi); that all academics are a complete waste of space; and that all funders have got it wrong? And that if all funders would just use your template everything would be fine?  ; )

    I agree with you that asking sensible questions like those that you list is part of good practice. Not enough practitioners go through this process. My experience however is that funders do have such processes in place.

    Perhaps like, Chairman Mao, funders wish to see a proliferation of flowers bloom. I don’t want to see funders have any increased power, nor do I wish to see them us their power to enforce a form of monoculture on ICT4D.

  • @Tony - Thanks for the comments. It’s good to have you share them here - not everyone seems to want to.

    In response:

    1. Not all ICT4D is useless, but I think you’d struggle to find many people who work in it who are happy with what we’ve achieved. And look at it’s ‘flagship’ programmes - Internet kiosks and OLPC among them - and how well they’ve gone down.

    2. FrontlineSMS and Ushahidi were mentioned in someone else’s research, which I was quoting. You’ll have noted my caveat when I added “and there were questions even around those.”

    3. Not all academics are a waste of space, but few would disagree that much of their work is little use in the world of practitioners and implementers. Their audience are, generally speaking, journal editors and other academics. And most practitioners I know struggle to make much sense of what academics write, which is no real surprise since they’re not the audience.

    4. Many funders do get it wrong, I think. They should take as much of the blame as the project initiators when they fail (for whatever reason, including poorly designed, applied, thought out or implemented ideas). Perhaps the fact that most donors hire programme officers who have never designed or built an ICT4D initiative before might have something to do with it. Or innovation experts who have never innovated.

    5. I’m not saying everything would be fine with the Charter, but it might be a start if we asked the really obvious questions before we start. The donors, for me, are just the obvious ones to do this. Self-regulation hasn’t worked, and most best-practice has been ‘conveniently ignored-practice’.

    6. As I wrote here - - on problem with letting a thousand flowers bloom:

    “... it confuses the end user who is expected to make sense of all 75 mobile data collection tools that end up on offer. The policy of funding many in the hope that the odd one shines through – the so-called “let a thousand flowers bloom” scenario – belongs to an earlier era. Today, we know enough about what works and what doesn’t to be far more targeted in what is funded and supported.”

    Happy to hear your thoughts on any of this!


  • BY Samir Doshi

    ON October 2, 2014 01:52 PM


    Thanks for continuing this discussion which you also touched upon in your SSIR article last year. The ICT4D Principles that Wayan pointed to—and was also described in Chris Fabian and Robert Fabricant’s article last month (—directly addresses many of these questions. It emphasizes that ICT4D projects should be squarely focused on working for and with communities, doing appropriate ethnography and M&E to create evidence-based learning and decision making, consider scale and sustainability, be open source/collaborative and more as outlined on the website (

    I should also state that the Principles have come out of years of conversations and forums from donors, practitioners and feedback from the ground, including much of your own feedback. The Principles have been signed on as recognized best practices by many in the donor community, including the Gates Foundation, UNICEF, USAID, SIDA, OCHA, UNDP, Global Pulse and more. The reason that they are framed as Principles is that they are evolving and open guidelines of how to operate in good faith and as a community, but not to be prescriptive of how to implement development interventions. 

    A few of us are writing a separate blog post from perspective of different donors, private sector partners and others that will present the Principles more in depth, unpack many of your pertinent questions outlined in the donor charter. We will also describe our initial thinking on how to operationalize these principles within our organizations.  We’re acknowledging the culpability of the donor community (with the understanding that the Principles are not only a donor effort, which would be unrepresentative and unsustainable) in the ICT4D space and are trying to continue the open and constructive dialogue.

  • BY Tony Roberts

    ON October 2, 2014 02:34 PM

    Yeah, fair point Ken, maybe not a thousand ; )  but no donor-enforced mono-culture either?

    We all agree that challenges exist but an opening gambit of attacking ICT4D practioners and academics and donors may not be the best way to bring people into the conversation?

    Part of the problem is recognising that none of us have all of the answers. Maybe a more inclusive process would help build bridges - like a collaboratively developed voluntary code built *with* practitioners and academics and donors?

  • BY Joel Selanikio

    ON October 2, 2014 02:37 PM

    I love the idea that the way to build good technology is to build a sufficiently large committee of all the “stakeholders”, then get them to adhere to certain philosophical principles.

    That sounds so promising, I’m surprised that Silicon Valley doesn’t try that approach.

  • Karl brown's avatar

    BY Karl brown

    ON October 2, 2014 05:30 PM

    I wanted to respond to some earlier comments about donor’s efforts in this space being useless - esp this by Joel: “We don’t need to change what the donors do, because what the donors do with regard to technology doesn’t matter.  Donors didn’t produce any of the most useful technology tools for those fighting poverty: Google Maps, text messaging, email, Facebook, twitter, mobile phones, etc.”

    I think there has been so much myth making out of Silicon valley in the last few decades along with unrepentant celebration of the genius of the private sector that donor contributions - even to things like google - are forgotten in the mists of history.

    In fact, the famous google page-rank algorithm that launched google was funded by an NSF research grant.

    Indeed, many of the fundamental technologies upon which all of the awesome and terrible mobile for dev apps are built on were originally supported by govt funded research - “All basic technologies that make our mobile phones “smart” can be traced back to governmental initiative and funding. Just a few: microprocessors; RAM memory; hard disk drives; liquid-crystal displays; lithium batteries; the Internet; cellular technology and networks; global positioning system (GPS); multi-touch screens.”

    Do you really think we’d have google earth if we didn’t have 30 years of govt funded space research and development of satellite technology? That the private sector piggybacks on, brands, productizes and scales that tech is the natural way of things but we can’t ignore who put in the first dollars.

    So we really need to ease up on the donor bashing and consider what are things donors can fund successfully, and what are things they should let other actors take up. But suggesting donors are insignificant in the ict4d space is ignoring a whole lot of history.

  • BY Joel Selanikio

    ON October 2, 2014 05:33 PM

    I’m not saying donors are insignificant in the ICT4D space, Karlito. I’m saying ICT4D is insignificant in the technology space.

  • @Tony - A “collaboratively developed voluntary code built *with* practitioners and academics and donors” sounds great, if you think you can get them all together. I guess someone would need to be the catalyst and driver for this, though. You volunteering?  ;o)

    @Samir - Thanks for sharing. I was under the impression from Wayan that donors were “signing up in their droves” so was naturally interested in who that was. It sounds like it’s the donors who were part of putting the Principles together, which is fine, but different.

    From my own perspective, if the Principles succeed in what they’re trying to do then I’ll be as happy as everyone else. We’re not in a race here, or in a competition, so if the Charter just becomes a slightly (hopefully interesting) side discussion to a better or more practical solution then that’s fine. We’re all working towards a more productive, effective ICT4D sector, or at least those of us who see it has a future in its current form. As you attested to, I’ve spent many years writing about how we might make better use of the incredible opportunity technology has presented us in the social/development sector. Sadly, in my 12 years working in mobile I’ve seen lots of talking and analysing, but little change in how ICT4D is ‘done’. We seem to have learned a lot, but done a far worse job of putting that learning into practice. And that’s everyone’s fault - implementers and donors, alike.

    I look forward to reading the post that you and your colleagues are putting together.

    @Karl - I hope you don’t see the Charter itself as a ‘bit of donor bashing’. It was designed to be a positive step, a proposal to unlock a decade of problems where many other solutions have either been ignored, or failed. My argument, as laid out in this post and on the website itself, gives two benefits of a Charter - one for implementers and one for donors:

    “Firstly, it would force implementers to consider key issues before reaching out for support, resulting in a reinforcement of best practice.

    Secondly, it will help the donors themselves by focusing their resources and dollars on projects which are better thought-out and less likely to fail.”

    As I said in my comment to Samir, if the Principles have more support and are a better way of taking us forward, then that’s great. There’s enough competition in ICT4D without adding to it.

  • BY Tony Roberts

    ON October 3, 2014 06:08 AM

    @Joel       ICT4D does not look up to Silicon Valley for inspiration.

    Silicon Valley’s White/fratboy-elite culture aims at market domination and sees the accumulation of vast private profit by venture capitalists and Founders as a sign of success. This way of thinking and being is responsible for creating many of the problems of inequality that development seeks to address.

    Human Development seeks to produce equal relationships based on social justice. Consulting stakeholders and especially involving the ‘intended beneficiaries’ of ICT4D in the determination of the means and ends of ICT4D is a key method in realising those relationships.

  • BY Joel Selanikio

    ON October 3, 2014 06:17 AM


    I don’t know whether the hearts of those involved in ICT4D are purer than the hearts of those in Silicon Valley. Perhaps you do.

    As Milton Friedman said, however, “One of the great mistakes is to judge policies and programs by their intentions rather than their results.”

    I know nothing, and assume nothing, about Silicon Valley’s intentions.  What I do know is that Silicon Valley, and other commercial technology producers, have in fact given us incredibly powerful and low-cost tools in our fight for global health and to reduce poverty.  I remember what public health was like before Hotmail, Google, Twitter, Skype, Google Maps—and the mobile phone.

    No one (least of all you I am guessing) would have predicted twenty years ago that commercial companies would produce technologies that could be provided for FREE to, for example, health workers around the world.  Technologies like Skype and free email that have revolutionized public health, and the activities of so many—and they were produce by BUSINESS, not by charity.

    You want people to make pledges and declarations, and to deliberately eschew the approaches of commercial technology—although it has given us so much and ICT4D has given us so, so little. 

    I would rather emulate the success of commercial technology as much as I can, and direct its techniques towards the advance of global health.  I’d rather skip the pledges and go straight to a working business model that creates and maintains useful software at a fair (and sustainable) price.


  • @Tony @Joel - If we think about the tools the non-profit sector most uses and relies on, it’s true to say that the commercial world wins hands-down as the provider. Joel listed just a few. There aren’t many comparable tools that have come from the ICT4D world, which generally exclusively targets those users.

    During talks I always use the example of mobile. The development community got very lucky when mobile phones broke through to become accessible to vulnerable communities. Jeffrey Sachs calls mobile “the most transformative tool we have for development” and, again, it’s a commercial offering. And a massively successful one at that.

    Does anyone have a list of commercially-developed tools vs. ICT4D-developed tools anywhere? It might be an interesting read and, if not, exercise.

  • BY Tony Roberts

    ON October 4, 2014 05:22 AM

    @kiwnaja @joel

    I am even less enamoured with Milton Friedman and Jefferey Sachs as potential role models than with the White elite frat-boys that we’re told we should idolise from Silicon Valley start-ups.

    It was Freidman’s ‘Chicago Boys’ and people like Sachs that did some of the early R&D for neo-liberalism and the Structural Ajustment Policies that directly caused the explosion of economic and social poverty experienced at the end of the last century. Sachs personally piloted what he called ‘shock therapy’ in Bolivia, before developing a fully-fledged White Saviour Complex and deciding that he had divined ‘The End of Poverty’ along with Bono.

    It is not clear to me why you wish to look to White elite men from the global North to guide you in your ICT4D work, especially when they are the arcitypal architects of the neo-liberal structural adjustment policies that created such inequality and (dis)advantage?

    Development, in my opinion, is not something that can be done ‘at’ disadvantaged communities by White prvileged elite men, not in their capacity as ideologues or ‘economic advisors’ nor by means of their inventions.

    If human development is to be a process of becoming able to determine and to solve your own problems, then it must be authored by those people themselves facing adversity. In so doing they may appropriate and repurpose technologies that were originally developed for-profit; they may use ones developed by not-for-profit entitites, and increasingly they should be able to innovate their own.

  • @Tony - I think you know my work well enough to know that I don’t “look to White elite men from the global North to guide [me] in [my] ICT4D work”. Far from it, as you’ll see if you read back over my various blog posts, guest articles and chapter contributions over the past decade or so, and look at the very practical work I did through FrontlineSMS over 8 years. I’ve worked very hard for many years to argue that ICT4D shouldn’t be following the ‘traditional development’ model of outside experts flying in with all the answers, imposing them on others. I’ll happily send you a few links to things I’ve written if you’re interested.

  • BY Joel Selanikio

    ON October 4, 2014 08:53 AM


    I’d work on your sense of irony a little as we watch one privileged elite person (that would be you) criticizing another privileged elite person(that would be me, or perhaps Ken) for being a pawn of other privileged elite people (um, Milton Friedman and Tim Cook, I guess)—while the first privileged elite person (you) claims to be the true representative of the downtrodden.

    Plus, I was kind of hoping that you would respond to the substance of my argument, rather than simply rejecting it because of the color of my skin. I’m still open to hearing your thoughts on that substance.

    To recap: I don’t know or care whether the founders of Skype or Hotmail are rich, or white, or privileged. I don’t care about their opinions, or how they vote, or whether they are pure of heart.  All I care about is that they’ve developed a tech approach that has given a tremendous gift to global health and international development, of infinitely more useful technology than has ever come from ICT4D (a huge understatement).

    To ignore their approach would be unfortunate.  To deliberately reject it in favor of more meetings and declarations seems like madness.



  • BY Matt Berg

    ON October 4, 2014 11:30 AM

    Apologies - I haven’t had the chance to read the full thread [I’ll try and do that when I have a few hours free wink]

    I agree that some of principals outlined by Ken would be helpful in informing the creation of new tools,  technology application and smarter investment by donor’s.

    I think the real problem, however, has been the LACK of funding in this space.  Sure, money has been wasted on some the development of some ill conceived tools and “pilots”. The tolerance of risk, however, afforded to ICT implementors is far less then afforded to silicon valley startups trying to make the next shiny photo sharing app.  Also, let’s not create a double standard in development funding. Let’s face it a lot of development is not much more then throwing stuff against the wall and see what sticks.  Most large scale donor funding is EXTREMELY risk tolerant

    Very few people developing tools in this space can do it without enormous personal skin in the game.  We also have to do it without the luxury of paying the salaries that attracts the top talent.  We work somewhat in a market failure.  Let’s face it, there is just not a lot of money out there if your target market is grass root NGO’s or ministries of health.  If there was, this space would probably be dominated by the big tech firms.

    As a result, we see a paucity of work class tools in this space (it’s starting to get better).  I would argue the LITTLE money that has been invested so far has had enormous impact.  Tools like OpenMRS and DHIS2 are good examples of what’s possible when you build a community and are able to secure long term institutional funding.  ODK and it’s ecosystem of related tools (including Magpi) has introduced the possibility of realtime data to development and that changes everything.  These tools are all starting to have real impact and our serving markets that would go otherwise unserved.  I’m in Liberia right now and I can attest that DHIS2 could potentially to be a Godsend to the Ebola response. This would not have happened without the sustained, continuos investments by the Norwegians (thank goodness for oil!) both in platform development and capacity over the past 5+ years.

    I’d love to know the total amount of money invested in ICT4D.  I’d be shocked if it was in aggregate over 50M USD.  To put that in perspective Snapchat (and app that allows you to safely send nudies) raised 50M in their series C (163M total).  Building world class software, no matter what it’s used for, is expensive.

    In summary, donor’s / implementing agencies - you pay for what you get.  If the way of approaching global problems is through 20K grants and hack-a-thons then you’ll continue to get amateur hour.  Without the big funding, the only approach to build a platform (by hook-or-by-crook)  is a combination of SAAS, consulting and implementation support.  It’s a slow and painful process. I can show you the scars and I’m pretty sure groups like Dimagi can too.  I’m not complaining though.  We are making steady progress. There are real benefits to running lean. And to some extent nothing should be given.  I just know we can do a LOT better.  The tools in this space, including everything I’ve worked on, are not where I know they could be.  A lack of smart and sustained funding in this space is a big part of it.  Let’s not let a few ICT4D taint the progress we as a community are making.  We have actually been a real bargain.

    Note:  Donor’s like Gates, Rockefeller, USAID, PEPFAR, NORAD, DFID, etc. deserve a lot of credit for the progress that has been made.  Some sustained and smart bets have been made.  We just aren’t seeing the major investments you see in the valley required to reap huge rewards.  Where are the moon shots?


  • BY Tony Roberts

    ON October 4, 2014 11:54 AM

    @Joel I am sorry that you have taken personal offence at my arguments. I assure you that none was intended. Please take the time to review my previous comments. I did not write a single thing personally about you at all Joel. Not about the colour of your skin nor about your level of privilege, a subject about which I presume to know nothing. I did not and do not claim to be the representative of the downtrodden. To do so would be ridiculous. It is disingenous of you to claim that I did. I didn’t reject your argument because of the colour of your skin. I didn’t mention you in my argument at any point.

    What I did do was explain why Milton Friedman and the Silicon Valley model that inspires you is, in my opinion, a negative model for ICT4D. Clearly this is a difference of opinion between us but I hope we can discuss that without personal attacks?

    My argument is that the values of neo-liberal self-interest that inform the development of the for-profit technology-as-commodity industry should not be ones that inform the ‘D’ of Development in ICT4D. I don’t subscribe to the charactersiation of ICT4D as being about the production of product (apps and technical artifacts), which is why I have little interest in disussions about who has produced the ‘best’ or ‘most’ apps.

    Rather than the ‘look at my great gadget’ version of ICT4D I am interested in the version of ICT4D which is focused instead around people processes and human development, about people’s ability to take control of their own development - appropriating and modifying what ever technology is appropriate to realise that end. They may find Skype and/or Frontline SMS, Facebook and/or any other tool to be of value. It ain’t about the tool for me. 

    I believe that making a conscious effort to make the process of ICT4D initiatives inclusive, collaborative and ‘community-led’, at each stage, is one way to check my privilege and to avoid the awful record of the corporate technology sector on equity and diversity.

    Apologies again for any offense caused Joel.

  • Hey, @Matt - Thanks for the really useful, considered comment. Suffice to say, I agree with most of it. By way of brief response:

    In a recent post on my own blog - “International development: A problem of image, or a problem of substance?” - - I end with a short list of things donors could (or should) be doing. One included:

    “Seriously get behind, and support, projects that we know are working, or know have the best approach. Stop always looking for the next big thing”.

    I’m TOTALLY with you that funding is in short supply, but only in some areas. Imagine what many of the better tools could do if someone like Gates seriously, seriously got behind them? Or any of the other tools which have shown considerable promise? The fact that the Norwegians have been committed to OpenMRS for so long is admirable, although it’s generally the exception rather than the rule. If they hadn’t gone in for the long haul, would it be around today to help out in the Ebola crisis?

    With the total funding for ICT4D projects, I doubt if anyone really knows, but I’d be surprised if it was UNDER $50 million. If you consider how long it’s been around, the figure it’s likely much higher. Like you, I’m curious.

    I realise the Charter isn’t the whole solution, but in a space where we’re constantly critiquing and hearing (and saying) how bad things generally are, it might be a start. I agree that some good has been done, but ICT4D - as with the wider ‘D’ - has been very poor value for money on the whole. $3 trillion over 60 years, in fact. As ICT4D increasingly becomes a playground for people to try things out, this can only get worse.

  • @Tony - I’d be curious about your take on a related post of mine (given the discussions here) - “How ‘Designing with the end user’ undermines ICT4D best practice” - - but perhaps we’re best saving that for another day.  ;o)

  • BY Joel Selanikio

    ON October 4, 2014 12:10 PM

    Hi Tony,

    What you SAID was that “White men from the global North . . . are the arcitypal [sic] architects of the neo-liberal structural adjustment policies that created such inequality and (dis)advantage”

    That very, very plainly IS disparaging an entire racial group that includes me.

    And ridiculous for too many reasons to list.  However no offense was taken.



  • BY Tony Roberts

    ON October 4, 2014 12:13 PM

    @Ken I’m sorry if you took personal offence, I assure you that I respect your work and integrity. You and I both passionately agree that ICT4D should not (in your words) “follow the ‘traditional development’ model of outside experts flying in with all the answers, imposing them on others”. But then you go and throw Jefferey Sachs at me as a expert authority, when he has the most appalling personal history of ‘flying in with all the answers, imposing the ‘Washington Consensus’ on others’ with quite terrible consequences in terms of economic and social development. It was like waving a red rag to a bull! Sorry for being bullish.

  • BY Joel Selanikio

    ON October 4, 2014 12:13 PM

    Hi Matt,

    Definitely some good points. I agree that in general there is not money to attract really top talent, and ICT4D suffers from that lack of top talent. I would be very worried about investing more money, though, (which otherwise might be spent on things that very definitely produce value, like vaccines) in ICT4D for fear that it would not produce much.

    There’s something wrong, I suspect, with the way ICT4D makes its money, and I think this creates the pilot/fail/pilot/fail cycle.  And though Ken and I agree on many things, I disagree that another set of principles is going to help.  I think it’s worth thinking about changing the way money is made in the field.  Certainly, I’ve found that as Magpi has moved completely away from grants, it frees us to focus solely on the concerns of our employers (i.e. the users).  This is quite different, of course, if the people you work for is a donor who doesn’t use your software.

    Would be interested in discussing this further. 



    P.S. By the way, although Magpi’s Android app was originally made as an ODK implementation (and we’re very grateful to the ODK team for that), our newest app is unrelated to ODK.

  • BY Joel Selanikio

    ON October 4, 2014 12:16 PM

    By the way, does anyone know how to unsubscribe from this @#$% comment list? grin

  • @Tony - Don’t worry - no offence taken. And hands up for using Sachs. Agreed = not a good role model.

    @Joel - It’s a monster you helped create…

  • BY Tony Roberts

    ON October 4, 2014 12:36 PM

    @Joel Not ALL “White men from the global North” were “architects of the neo-liberal structural adjustment policies that created such inequality and (dis)advatage” across much of Latin America and Africa, but Milton Friedman and Jefferey Sach certainly were. It was specifically at them that I directed criticism. You have no reason to be defensive (unless you were also running with the Chicago Boys?)  wink

  • BY Tony Roberts

    ON October 4, 2014 12:41 PM

    @Ken @Joel It’s just become *yet another* example of white blokes gobbing off about development *facepalm*

  • BY Joel Selanikio

    ON October 4, 2014 12:45 PM


    I think that what you meant to say (if clarity was the goal) was that “Milton Friedman and Jeffrey Sachs were architects of the neo-liberal structural adjustment policies.”  You, however, made a generalization about “white men from the global north.”. 

    I wasn’t sure why you brought race into the discussion at all, frankly.  Before you showed up, no one had mentioned it.



  • Matt Berg's avatar

    BY Matt Berg

    ON October 4, 2014 03:19 PM

    It’s not a zero sum game especially when we’re talking about some of the private foundations.
    Money spent on software doesn’t necessarily mean less money spent on vaccines. 

    The fundamental problem is it would be nearly impossible for someone like myself to raise 5M on the market to build tools with a primary focus on tackling development issues.  If I had a clever idea for a consumer app and assembled a good team it would not be a problem.

    5M would allow a startup like mine (or others in the field) to build a team and take some calculated risks that really move the needle from where we are now.  Note: I said startup (for profit) and not an NGO.  I agree with Joel in that you have to ultimately find a biz model to sustain this.  I think such a biz model can potentially sustain a service running with improvements overtime.  I think some risk tolerant investments (subsidy) is needed if we hope to raise the bar on the tools we’re producing as a community today.

    @karl if you and Rockefeller like to help prove my hypothesis right just let me know wink

  • Matt Berg's avatar

    BY Matt Berg

    ON October 4, 2014 03:38 PM

    @Ken - who is complaining?  From the practitioners in the field, I sense a general sense of optimism, growing maturity and real in the direction things are going.

    I don’t think there is really a sense that we have too many different SMS solutions out there either.  While there is some overlap, the ones that have made it all seem to over to some extent some type of unique value proposition or be linked to an individuals that can create their own communities to create value.  If we stopped at FrontlineSMS we never would have had TextIt.  I guess that’s maybe funders are still lining up to fund messaging and photo sharing apps.

    While I agree, ICT4D discussions used to be dominated by the concerns you raised in this post, I generally think amongst those actively working in the field that we’ve moved on.  Good ideas and tools are getting real traction.  Rocket fuel though remains in short supply.

  • Yaw Anokwa's avatar

    BY Yaw Anokwa

    ON October 5, 2014 07:48 AM

    @Joel, you said “Although Magpi’s Android app was originally made as an ODK implementation (and we’re very grateful to the ODK team for that), our newest app is unrelated to ODK.”

    The Magpi app (v4.0.3) on the Google Play Store today uses a lot of code from ODK Collect and JavaRosa. I know because I wrote that code. Glad to decompile the Magpi app and provide you the percentage of code that is original if numbers are needed.

    Visually the apps certainly look the same. Even the “Making data collection a little easier” tag line on the launch screen is unchanged from an earlier build of ODK Collect.

    Is there another Magpi Android app you are referring to? If not, we can at least agree that the app in the Play store is very related to ODK Collect?

  • BY Joel Selanikio

    ON October 5, 2014 07:53 AM

    Hi Yaw.  Quite right: my mistake!  We have another app coming and I thought it was already up. Stay tuned.  Best, Joel

  • BY Joel Selanikio

    ON October 5, 2014 08:23 AM

    Thanks, Yaw. I really appreciate that. (If needed, let’s discuss further offline to this article comments section).  Best, Joel

  • @Matt - Maybe I spend too much time on Twitter, at conferences, and reading blogs. On the whole, most of what I hear are criticisms and concerns about too much focus on the technology and not enough on the ‘D’, a fetish for big data and smartphones (which will save the world), iPad apps for African farmers, and so on. Although there is certainly good stuff going on, the conversation seems to be dominated by problems, challenges and questions we’ve been asking for years, most of which haven’t yet been answered. From -

    - How do we replicate and scale?
    - How do we measure impact?
    - How do we stop the reinventing of wheels?
    - How do we avoid being ‘technology-lead’?
    - How do we break out of our silos?
    - What is our business/sustainability model?
    - Is open source a help or a hindrance?
    - How do we maximise the opportunity mobile brings?

    These are always the conversations I remember. I guess ICT4D is just following the wider development discussion, which is rarely positive, but maybe sometimes should be.

  • Matt Berg's avatar

    BY Matt Berg

    ON October 6, 2014 02:16 AM


    I get the concern.  I just don’t think the conference circuit experts reflect the general sentiment of what’s happening in the field.  I don’t know anyone that thinks giving farmers iPads is a good idea.  I think more the focus is realizing that some farmers might own their own smartphone soon or have access to one through their agri-dealer / coop and how we take advantage of that.

    As tech becomes more ubiquitous (this is really what’s changing) it’s really becoming a question of when not if.  This is allowing us to focus a lot more again on core problems trying to be address.  Not saying the questions you aren’t raising aren’t real.  I just think the debate on the topic above is shifting.  Technology is not a fetish.  It’s just one, albeit increasingly effective, tool development practitioners have at their disposal.

    We are moving forward.


  • @Matt - If only everyone thought like you. smile Thanks for that. And, absolutely - let’s keep moving forward.

  • BY Steve Vosloo

    ON October 6, 2014 11:24 AM

    Congrats, Ken, for taking the initiative with a Donor Code of Conduct. I think it’s a great start and the points made by commentators have been excellent.

    I agree with @Matt on two points in particular: that not enough funds have been invested in ICT4D, and secondly that “donor’s like Gates, Rockefeller, USAID, PEPFAR, NORAD, DFID, etc. deserve a lot of credit for the progress that has been made.”

    Nevertheless, the Donor’s Charter currently puts all the onus on the fundee to pick up their game and none on the funder. In the world of development, as in Buenos Aries, it takes two to tango. The funders also need to align with a code of conduct, and here is a first stab at a few additional points to the charter (wording to be fixed later):

    We, the donors, pledge to:

    —Allow you to be agile and adapt. We like predictability and accountability; it’s what we report on to our boards and governments. You trade in unpredictability. Not only is your context on the ground diverse, severely challenging and sometimes downright dangerous, but technology itself is changing all the time. If you only do what you committed to us to do in three years, you’ll almost certainly have failed to make maximum impact. The iPad didn’t exist five years ago. Technology changes, people change, contexts change, and so should your projects. 

    —Not place onerous reporting duties on you. Of course you need to report and be accountable, but we don’t want you to spend all your time on that. We’re funding you to innovate.

    —Keep thinking about how best to fund you. We recognize the inherent problems of time and budget bound funding, and want to work with you to find creative ways to incentivise both you and us. (The dominant funding approach which is time and money bound—$300,000 over 3 years, for example—is one of the leading causes of pilotitis. The funding is for the pilot, not for the scaling up. Yes, we can’t fund a project forever, but we need to think about more creative ways to fund for scale, which may take longer than the 3 years. (I blogged about this before

    —Value learning as much as impact and scale. We all want scale, but it’s a tough nut to crack. Very few organisations, including governments, have gotten it right successfully. We understand that not everyone will achieve scale, but everyone will learn something along the way.

    —Help you to go to scale by leveraging our networks and voice. We are often well connected and can help you meet the right partners, or just shine a spotlight on you for maximum reach.

  • Matt Berg's avatar

    BY Matt Berg

    ON October 6, 2014 11:28 AM

    Steve - you sound like a VC!

    Last few points in particular are the essence of how VC firm Andreessen Horowitz is supposedly shaking up Silicon Valley.  Donor’s see the article below.  It’s quite insightful.

  • @Steve - Great comments. Thanks.

    In response, I think grantees do need to up their game in many instances. The questions the Charter asks should be the kind of thing ANY project asks itself before it sets off at one hundred miles an hour. So I don’t think this creates much more work - it simply reinforces what they should be doing anyway.

    The other suggestions are great, and would add some transparency to the process. A ‘real’ Charter would likely be a combination of the kind of wording you’ve drafted, along with the set of questions. Imagine, if this existed and it was pushed out as an initial response to anyone approaching a donor for money, wouldn’t that be a better start than what we have now? We’d all know where we stood, and things would be far more open and transparent.

  • BY Jeff Wishnie

    ON October 7, 2014 11:23 AM

    Ken, I read this post and comments with interest.

    I particularly liked the point that Matt Berg raised on the very slow pace of innovation in T4D and how this is tied to funding.

    There is a new generation of far-more professional technology-for-development companies, including Matt’s, Nyaruka, engageSpark, DevResults, HNI and many more—not to mention pioneering organizations like the Frontline CIC, Dimagi and Magpi.

    These companies distinguish themselves from the earlier generation of academic research teams, open source projects, and part-time side-gigs, by:

    1. Offering services, not just tools
    2. Operating as businesses
    2. Staffing real, full-time, competent product development teams (engineers & designers)

    BUT they are all struggling because there is so little money in the space.

    Social Sector clients put many demands on companies and because we manage our budgets so tightly, we rarely pay anything close to what a commercial client would pay for a similar service.

    In order to survive, these companies must focus on consulting services, and they starve their R&D.

    The result: the products evolve, mature, improve, and innovate VERY slowly.

  • Hapee de Groot's avatar

    BY Hapee de Groot

    ON October 7, 2014 12:21 PM

    I do not always agree with Ken but in this case I do, in my 13 year of experience I have seen a lot of projects been funded that on the technical side did not make sense, there are two things I observe:
    most donors at least in my environment do not know enough about technology to assess a project so for certain project tools are build that already exist or in a software that keeps the developers in business but not the sustainability of the project. Even some choices of implementing partners are at least unexpected, they choose to build their products in Ruby and are never up to speed with the latest builds of Ruby, which slows down the development and the innovation.
    The other issue is that whatever has been build and it is a lot throughout the last decades has not been shared as it should, we all should if it is applicable and in most cases it is applicable put the code on Github and build the repository of what has been build and reuse what is already available.

    So lets start with that code of conduct.

  • @Jeff - Agreed on the emergence of ‘ICT4D 2.0’ organisations, if I dare call them that. But again, as per previous comments, if these are emerging as the new model and the new way of doing business, and in a way that helps create sustainable, professional outfits that build carefully thought-out and designed tools, then we should focus more on nurturing and supporting them. The problem again goes back to what life is like on the ground - far too much is getting funded that doesn’t do any of this, and doesn’t have much chance of becoming a ‘2.0’. Hapee’s comment, which follows yours, is testament to that. Yes, there is some good stuff going on, but so much more could be done if we were better at not backing the losers.

    @Hapee - I’m glad that, after 13 years, I’ve written something you agree with! In a related post (not this one) I highlight some of the challenges donors have in identifying good projects to fund. Not having staff with direct experience of tools development, or staff who have never innovated, is an ongoing problem in my opinion, that can be easily fixed.

  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 7, 2014 12:48 PM

    @Hapee, I thought we should always agree with Ken wink

    @Ken, agreed (see ^) , we should be doing more to support these organizations. And to be clear, I don’t want to fool myself that supporting the emergence is easy.

    But I do think that some of the necessary answers go beyond a code of conduct and require fundamental changes to how ICTs are funded and managed.

    For example, project-based funding is a huge contributor to duplication. I have been in the Kafka-esque situation of being unable to pay for a global system, accessible and useable by all projects because funders don’t fund ‘overhead’, but 10 individual programs can re-build the same system, or by 10 different ones, at 10x the cost through their individual budgets.

    Code of Conduct is a good starting point.

    Changing the way technology is purchased and deployed _in particular_ recognizing the incredible value of always-available, globally deployable, platforms, is another necessary condition.

    And to belabor the point ‘enterprise’ service licenses for large clients is a far more manageable business model for ICT4D 2.0 companies than a program-by-program sell.

  • @Jeff - Thanks for that. I’ll be the first to admit, I don’t think a Charter has all the answers, but it’s certainly a starting point. At the very least it’s got people talking, and that can never be a bad thing. Let’s just hope that, if we agree things need to get better (and that there are, indeed, problems), that all the words lead to some kind of meaningful change.

    And good move, by the way, agreeing with me. ;o)

  • BY Herb Caudill

    ON October 8, 2014 06:44 AM

    Hi, all. Thanks @Kiwanja for starting this great conversation. I’m coming in late, so all the smart stuff has already been said. So I’ll just underscore a couple of points others have made.

    Most of what @Jeff Wishnie and @Matt Berg and @Joel Selanakio have said here rings true to me. I’m flattered to see DevResults on the list of “ICT4D 2.0” organizations, and it’s encouraging that this conversation has turned to helping little for-profit social-impact businesses like DevResults succeed. Like Ona and Magpi and others, we’re trying to build something that’s going to last. And building a viable company in this space is really hard to do, especially if you’re bootstrapping without taking outside capital.

    So it would be really nice to see donors using - or at least trying - what we have to offer instead of rolling their own thing.

    It drives me crazy, in my little corner of this space, to see one donor after another paying for bespoke software projects to do what DevResults does - with software development budgets that dwarf DevResults’ annual revenue. It drives me crazy not because I want that business, but because I’ve NEVER seen that approach succeed. Building good software is an iterative and incremental thing and it generally takes a couple of major versions to get it right. Most projects never get it right, and the ones that do even get to a minimally viable product are ready to close down shop by the time they get there, and all that code gets thrown out only to start the cycle over again.

  • Matt Berg's avatar

    BY Matt Berg

    ON October 8, 2014 06:51 AM

    @Herb - nailed it.

  • @Herb - Echoing Matt, who I always agree with. ;o)

    The Donors Charter isn’t perfect, but it might get us to a place where we see less “one donor after another paying for bespoke software projects to do what DevResults does”. As I approach the end of my 12th year in mobile-for-development, I’d love to see us make some genuine progress on some of this. It really isn’t rocket science, but we seem to be content with just tweeting about it, and convening conferences to debate it.

    Anyway, thanks for sharing your thoughts.

  • BY Joel Selanikio

    ON October 8, 2014 08:05 AM


    Appreciate the kind words about Magpi as pioneer. Not where you got the idea that we are “struggling”, however.  We’ve gone from 3000 users worldwide to 33,000 in just the last 4 years—and use of Magpi has expanded well beyond the development sector to lots of commercial and government organizations that just want to collect data.  We haven’t received a dime of grant funding in nearly five years and are entirely funded by customer revenue—and expanding all of our efforts.

    If this is struggling, I hope it continues. grin



  • BY Joel Selanikio

    ON October 8, 2014 08:10 AM


    I’m with Matt on this: you nailed it.  Rather than using services like Magpi and DevResults, donors are just as likely (or maybe more likely) to pay to build expensive projects that will cost 10x but not work as well. Drives us crazy, too—but luckily there are enough sensible organizations that recognize the value provided by Magpi, and DevResults.

    Here’s to continuing to get the word out about cost-effective, customer-supported (and therefore customer-focused) technology to support development!


  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 8, 2014 08:12 AM

    @joel—I’m using the pace of innovation and functional improvement in the _commercial_ sector as a yardstick.

    I can see you bristling at a negative term like “struggling”, but let’s be honest with ourselves, and compare the progress and growth of the most successful T4D companies (Magpi and Dimagi) to companies like, oh, Salesforce or Amazon, or even far smaller startups like Heroku.

    Our entire field is left sooo far in the dust.

  • Karl Brown's avatar

    BY Karl Brown

    ON October 8, 2014 08:14 AM

    Ken thanks for continuing an important conversation, but I can’t happen to be struck by how many of the points raised have already been raised and are known and many were integrated into the ict4dprinciples discussions. Reinvention of the wheel is endemic in this space and I think to avoid further reinvention of this discussion I’d encourage all of us to join forces with the ict4dprinciples group which has a mailing list and webinars and other venues for engagement. Signing on to the principles or the charter is ultimately only a first step, what really needs to happen is better investments which lead to better products and then joint effort behind those products to replicate and reuse. For product development I see two key pathways (this is a simplification obviously) - private companies which develop proprietary solutions, and non-profit orgs which support and contribute to open source platforms (hybrid models are also becoming more popular, and I think there will be significant growth in these models) We’ve seen successes in both and ultimately it’s strong products that solve real problems that will make the biggest contribution, no matter their licensing model. I also want to problematize ict4d, as I’ve rarely seen ict4d projects which are just about ICT - instead ICT is brought in to help enable more efficiency in service delivery - that’s why we don’t have good data on ict4d successes and failures as it is often embedded in larger projects with larger goals (and even for that broader development agenda we don’t have great data).

  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 8, 2014 08:18 AM

    @Joel (form 8:10am PT wink

    Part of the issue is in fact _not_ the amount of money (which to Ken’s point can be pretty large) but the structure knots we tie ourselves into in grant and procurement structures.

    Specifically, anything spent by an organization as a whole is “overhead”—which is a far smaller pool than the sum of funds available across all individual grants/projects added together.

    I would _love_ to see a donor charter _explicitly_ support the deployment of organization wide infrastructure available to all projects.

    It’s not rocket science—we can be much more efficient with our funds if we have a site-license for Magpi or Ona, rather than having every single grant implementation team signing up individually.

    By it’s very nature, the grant funding model splinters the money available to spend on large scale ICTs.

    If we change that, I think it will invariably drive more concentrated, more resourced, larger scale ICT platforms—primarily commercial ones.

  • BY Joel Selanikio

    ON October 8, 2014 08:19 AM

    Hi Jeff,

    Totally with you regarding impact, and we’re definitely weighing whether we need to think about bringing in investment to boost our capacity.  But “small” and “struggling” aren’t the same thing.



  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 8, 2014 08:33 AM

    @Joel, maybe we can agree on small and not-nearly-as-fast-moving-as-we-need wink

    @Karl—I agree we have examples of both commercial offerings and successful open-source projects.

    My favorite model, for flexibility, is they hybrid model where commercial businesses offer cloud-services based on open-source projects (which they generally drive or support).

    BUT I am going to kick this dead-horse, the open-source and hybrid projects also move very slowly fundamentally because they don’t have the money to spend to hire enough people (yes, open-source projects require paid-staff, whether paid by the project or a supporting company/org) to make fast progress.

    Imagine where OpenMRS and OpenHIE would be today if it they had a fraction of the design and developer resource that Candy Krush does!

  • @Karl - I still don’t think the Principles and the Charter are the same thing, but either way it seems the Charter has started a wider public conversation than the Principles have. I hadn’t even heard of them before I wrote the Charter - had I done I’d likely not have written it. So I’m a bit confused about the strategy of engagement before, during and after their creation.

    I did ask Wayan some questions much earlier in this thread about the Principles after he told us they were the answer, but he’s made a conscious decision to not take part in this discussion, which is odd. I’m still curious on:

    1. Who are the principal audience? Is this to remind solutions developers what they should be doing? Or for donors to sense check proposals?

    2. Are they going to be enforced in any way? If not, what’s different about this than all the other sets of ‘best practice’ we’ve seen over the past decade?

    3. Who’s ‘signed up’ to the Principles, and what does ‘signing up’ actually mean?

    4. I’m curious who else was consulted beyond the giants of the development community listed on the site? There seems to be a lack of any grassroots voice, or any of the smaller organisations who probably have a lot to share from their experiences.

    I’d still love answers to these questions!


  • Karl brown's avatar

    BY Karl brown

    ON October 8, 2014 08:45 AM

    Ken, I’d suggest reaching out to the organizers offline. - as you can see a fair number of groups were consulted. Re: consultation, how open was your own process in developing the charter? Who did you consult with before posting it? Any given process can always be more open but I don’t think I would be speaking out of turn if I said they welcome more inputs - indeed they ask for this on their site. I also think that it’s not just about the donors, ultimately implementors make decisions that are not always under control of the donors (again, depending on how the funding is structured, earmarking rules prevent private foundations from forcing certain solutions on implementors, and it would be arrogant for us to believe that we know better than them the best technology. I’ve tried to influence this in the past but it doesn’t always work, and joint ownership and empowering of local decision making is important as well in these relationships). Ultimately a joint conversation is needed. If you don’t agree with the principles as elaborated or believe that it needs to be stronger, work to build consensus for same or embed new ideas from the charter rather than just going your own way…

  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 8, 2014 08:58 AM

    On the topic of the ICT4DPrinciples, Mercy Corps was one of the first implementing organizations to endorse the principles, and many of the people on this discussion thread have been deeply involved in framing theme (not just Wayan!)

    The principles are important. The principles are about the HOW. How ICT4D projects should be implemented. And a bit on the FORM that the resulting technical artifacts should be formed (e.g. open).

    They do not talk at about how projects should be supported and funded.

    IMHO, a Donor Charter should:
    1. Endorse the ICT4D principles as key to ethical and successful T4D implementations.
    2. Include _specific_ actions for speeding up innovation and product evolution, which must absolutely include innovations in funding models.

  • Jeff Wishnie's avatar

    BY Jeff Wishnie

    ON October 8, 2014 10:51 AM

    I wish we could make edits to these comments. For the record, before I upset anyone, back here < > I meant to say “SOME of the most successful companies (Dimagi and Magpi)” I did not mean to say that I think those are the only or most successful.

  • BY Mikel Maron

    ON October 9, 2014 04:13 AM

    13. Are you willing to share by default (allowing for sensitive data exceptions) any data and analysis produced in your project, under an open knowledge license? What efforts will you make for local access to this information?

  • @Karl - Just to follow up here, I am in contact with the Principles team and have been since very first hearing about them (which was after the Charter was written). All I’ve basically written is a thought-piece (even though I hate that word). I have zero funding and zero resources to implement anything, and was never under any illusion that an 800 word article would change anything. Very little that’s ever written changes anything in ICT4D in my view. If what I’ve written is helpful in any way then I’m happy for it to help steer or guide any ongoing discussion, and am always happy to be a part of that. Anyone else who has commented on this SSIR post is welcome to do the same, obviously, if they think the Principles are the answer.

    Glancing again at them now, though, there are many differences between what they say and the Charter. People can make their own minds up when they read them both, though.

    @Jeff - I’m excited to see how the Principles influence behaviour in the field. We’ll only find out over time, and we’ll only know if people stick to 9c.

    @Mikel - Good point. Certainly worth adding!

  • Given there were a few comments (and some unanswered questions) about the ICT4D Principles, and how they were ‘the same’ as the Donors Charter, I’ve written a separate blog post about the differences here:


Leave a Comment


Please enter the word you see in the image below:


SSIR reserves the right to remove comments it deems offensive or inappropriate.