While there has been great excitement and buzz around various ICT4D projects in recent years, it is also true that many projects started with a bang and later died with a whimper. In short, they have not been scalable or sustainable in the long run.
What makes an ICT4D project fizzle? What are the common mistakes that donors, planners and implementers make when trying to run an ICT4D project?
The problems of a top-down approach to ICT4D
In a 2004 paper, “Using ICT to place Indigenous Knowledge Systems at the heart of Education for Sustainable Development” [pdf]  Arvind Ranganathan, an independent consultant, who was then an Associate Professor at the Asian College of Journalism in India, argued that:
Much as the definition of ICT has grown to be more inclusive of ‘low-tech’ devices, existing implementations of ICT are still largely top-down in their structure of information. They are based on the assumption that flow of information must be predominantly one way: that the developing world needs information of and from the developed world to reduce poverty and to improve its standard of living. There is acknowledgement of the importance of local knowledge and the need to tailor information to be culturally sensitive and context-specific. However, in this scheme, local knowledge is to be incorporated, implying that the mainstream information will be that which flows from the developed world.
According to Ranganathan this top-down approach is a recipe for failure. Instead, he suggests building “circles of knowledge” through a bottom-up approach based on the model of community sharing.
Likewise, in a post on South African website, Tech Leader in August, Kirstin Krauss pointed out  that when an outsider tries to impose ICT based solutions on a community, it is more likely to fail than succeed. The recipe for avoiding disaster – according to him – lies in ‘identifying cultural interpreters and community champions’ and making them an integral part of the process.
Prof. Pradoshnath from the National Institute of Science Technology and Development Studies  (NISTADS) in New Delhi, recently gave a presentation at Centre for Internet and Society  in Bangalore on ICT, transaction cost and development: The flip side. His main observation  is that connectivity matters only if it connects the right way. If connectivity is not used to enable economic and social gains, he says a marginalized population can run the risk of falling prey to new dynamics of exploitation.
A public chat between ICT4D critics and enthusiasts
In an interesting, free-wheeling Twitter Chat [#failday ] that was facilitated by Inveneo's @ICT_works  on December 11, participants associated with the field of ICT4D discussed common mistakes pertaining to ICT4D, their probable underlying causes, and learnings for the future. Some of the learnings shared at this highly participative chat session were as follows:
@ ICT_Works : I see gadget lust taking people in the wrong directions: I want to use shiny, flashy hammer so find me a nail…
@downeym : I think too many ICT4D design a big, complex system, without understanding actual needs – then it gets tossed out the window. :)
@theresac : flip-side: too many people heedless of failure execute poorly thought out ideas, at enormous cost
@HannaHellquist : Lack of ICT competency on donor side might lead to investments in systems that do not meet actual needs
@ ICT_Works : I also see under-funding of technology: costs $1000? Then here's $500 and get volunteers instead of full investment
@partners4cd:  Techno-flash design un-buttressed by social feasibility and sustainability.
@joncamfield:  Don't forget buzzword compliance pushing new tech instead of reliable, tested tech (e.g. cloud computing??)
@giantpandinha : … biggest mistake is not providing space for mistakes –> innovation. (Institutional donors not helping.)
@ithorpe:  Moving onto the next project without taking a little time to reflect on what we learned from the current one
@communicc8:  in general, lack of transparency causes failure: when people can't ask/answer simple Q, “WHY are we using this?”
@mobileactive : end users and techies are not speaking the same language and need time and trust to understand each other
@mobileactive : Open source is critical but also not easy. Tech is messy, social development is messy, together failure is high. Understand that.
@africastrategy : We tend to look at the technology not the objective. Often solution is more simple and closer to home than we think
What is the way forward then? There are several answers hidden in the observations above. Here are some additional suggestions that the #failday  chat participants had for future ICT4D projects.
@sanng : creating a platform for non-dev world AND dev world ICT4D folks to collaborate (is must engage dev world in these discussions)
@joncamfield : Publishing donor-friendly best practice reports with data from a variety of ICT4D projects will help to push back on bad requests
@mobileactive : nobody has talked here yet about agile design practices in ICT4D. Essential to at least reduce likelihood of failure.
@joncamfield : Also, taking risks with new disruptive ideas is important to break out of molds.
@downeym : Open-sourcing a “failed” project turns it into a success – or at least helps others’ chances at being one.
partners4cd : We need core Monitoring & Evaluation standards to heighten donor confidence in tech-assisted solutions to socio-economic development issues
You can read the entire transcript of the [#failday] Twitter Chat here .