"Crowdsourcing" development data?

A rather unusual report sponsored by the Cabinet Office calls for the UK government to embrace a new strategy for producing and disseminating information. The authors argue that the value of public sector information has – potentially, at least - significantly increased as two new groups of citizens have emerged: those who take part in online communities and those who mix-and-mash data to create new information.

To prove their point, the Office gives among others these two interesting examples: "in medical studies of breast cancer and HIV patients, participants in online communities understand their condition better and generally show a greater ability to cope. In the case of HIV, there are also lower treatment costs," and "sharing restaurants' food safety information in Los Angeles led to a drop in food-borne illness of 13.3%." However, to realize this potential, the government must lower the barriers for re-use of information.

The report contains many lessons applicable to development institutions such as the importance of aggregating data (more on this in an upcoming post), embracing open standards and encouraging self-help fora. Most significantly, though, development institutions should embrace its vision that "citizens can create, re-use and distribute [data] in ways that add maximum value." In other words, the best collective knowledge may reside outside, not within the organizations themselves.

Surprisingly for a sector that has traditionally relied on volunteers to make up for its limited resources, development institutions have been slow to embrace "crowdsourcing" as a model for adding value. Chris Andersen at Long Tail defines crowdsourcing this way: "where users happily do for free what companies would otherwise have to pay employees to do" (think about readers' recommendations in Amazon or Google's self-service model for advertising).

But if data were published in a way that encourages re-use and collaborative analysis such as mash-ups, the impact could dramatically increase. Once information is out there, there's no limit to what creative users can do with it (see this exhilarating GoogleTalk with David Weinberger for the evidence).

With so many highly skilled people with a passion for development, one is only left to wonder at the exciting new insights that "data mashing laboratory" would produce with the UN, IMF and World Bank data (just to provide some examples close to home).

So what do you reckon: how far are we from crowdsourcing to become the standard model for development data publishing?