Finding space for crowd-sourcing in humanitarian response

14. July 2010

Originally published here.

05 Jul 2010 13:23:22 GMT
Source: IRIN

Reuters and AlertNet are not responsible for the content of this article or for any external internet sites. The views expressed are the author’s alone.

NAIROBI, 5 July 2010 (IRIN) – “Crowd-sourcing” is a new buzzword in the world of humanitarian information. The combined power of mobile phones, mapping technology and social networking can enable citizens in crisis to seek help, facilitate aid deliveries, bear witness to abuses and hold governments and aid agencies more accountable, advocates say.

Crowd-sourcing on platforms including Ushahidi, for example, [http://ushahidi.com] [http://irinnews.org/Report.aspx?ReportId=89473] took place on an unprecedented scale after the January 2010 earthquake in Haiti. According to those involved, the impact it had is undeniable: communities were able to report their needs while accurate street maps were created for humanitarians and search and rescue teams trying to save lives.
“Crowd-sourcing had been used in previous emergencies, such as the Wikis [http://en.wikipedia.org/wiki/Wiki] created to map Hurricane Katrina and bird flu, but none seemed to have a life beyond the particular incident,” said Microsoft’s Nigel Snoad, an adviser to the ICT4Peace Foundation [http://www.ict4peace.org]. “But in Haiti, Ushahidi and its partners seemed to have a real impact on the way the humanitarian response worked.”

Beyond Haiti
“There is real excitement in the humanitarian community about crowd-sourcing and what it can do for emergency humanitarian response,” he added.
But, he says, there needs to be a meeting of minds, with the technology experts ready to develop tools that can contribute meaningfully to humanitarian response and traditional organizations such as the UN being prepared to embrace non-standard methods of handling emergencies.

“Technology developers can affect how and by whom their tools are used by the choices they make; they need to look at validation, protection of data, and so on, and they are doing this,” he said. “And traditional actors like the UN have to learn how to bring [in] non-traditional actors and their work, how to channel them, advise them and link them to the humanitarian system while allowing them to remain independent.

“Crowd-sourcing not only brings speed to humanitarian work, it opens it up to allow more effective, non-traditional operators to engage with traditional systems of gathering information,” Snoad added. “It would also be a great way to hold humanitarians accountable – to ensure that help promised is actually received.”

Glenda Cooper, a fellow at the Reuters Institute for the Study of Journalism in Oxford, told IRIN: “The cliché is that the aid system is reluctant to welcome innovations like this – however, I think the reverse is true. Citizen journalism, social networking and crowd-sourcing have been embraced enthusiastically by many NGOs. In some cases, too enthusiastically.

“Just because you have a Twitter account, or have money to put into a new website, doesn’t mean that you use it effectively. The real challenge for NGOs is to learn what they can use effectively – and whether they are duplicating other agencies’ work unnecessarily.”

Hitches and glitches
“It is important to realize that even in Haiti, crowd-sourcing didn’t work perfectly – there are problems with validation and accuracy, and codes of conduct need to be developed… [for example] it is terrible to ask people to report their problems if there is no way to solve them,” Snoad warned.

One of the main problems is the unverified nature of the information. “Anyone can now publish rumours or thoughts online, thereby bypassing an editorial process,” said Guy Collender, senior communications officer at the London International Development Centre, an academic organization. “Proponents of crowd-sourcing recognise the concerns about the credibility of websites relying on information often generated by anonymous sources. However, they believe the risks of false reporting are more than cancelled out by accurate reports.”

Ushahidi, which was created by Kenyans trying to track and mitigate post-election violence in 2008, is trying to address this with a new open source tool, SwiftRiver [http://blog.ushahidi.com/index.php/2009/04/09/explaining-swift-river], which filters large amounts of incoming information to separate the wheat from the chaff using smart algorithms and human operators.

“Crowd-sourcing does not yet have an established standard such as quality control, ground verification or sustainability,” said Akiko Harayama, an information analyst with the UN Office for the Coordination of Humanitarian Affairs (OCHA) New York familiar with the Haiti operation. “Crowd-sourcing is something new, and everything new requires some time to adjust and to improve.”

Crowds of volunteers
The formation of partnerships is also critical, humanitarian workers have found. In Haiti, partners such as the SMS service, Mission 4636 [http://www.mission4636.org], students from the American Tufts University, [http://www.tuftsdaily.com/volunteer-driven-ushahidi-web-platform-contributes-to-relief-efforts-in-haiti-1.2155130]the Thomson Reuters Foundation [http://www.reuters.com/article/idUSTRE60G16420100117], the Google-affiliated disaster technology service, Innovative Support to Emergencies, Diseases and Disasters [http://instedd.org], Sahana [http://www.sahanafoundation.org], an open source disaster management system, local phone companies, conventional aid agencies and countless others pitched in. Agencies also closed the loop by devising tools to feed information back to the general population through SMS broadcasts and local radio programming.

“Without hundreds of individuals monitoring a variety of sources of information and mapping it, this crowd-sourcing operation wouldn’t be possible – it is still primarily more about the human input, commitment, dedication and cooperation than about technology,” said Ushahidi’s Jaroslav Valuch.
“In Haiti, the network of volunteers providing information on the ground, and the number of people volunteering to make sense of this information, was unprecedented,” said Snoad. “Communities on the ground played a huge part, and there is a definite need to find a place for them in the ownership of crowd-sourcing.”

What next?
A crisis mapping conference [http://www.crisismappers.net/page/iccm-2010-haiti-and-beyond] in October at Harvard University will address lessons learned from Haiti and other disasters, developing a code of conduct for the technology community and the future of crisis mapping and humanitarian technology.

According to Valuch, for crowd-sourcing to be successful, there is a need for “lessons learned” processes; consultation with humanitarian actors; establishing links, protocols and partnerships before disasters happen; raising awareness about potential as well as limitations of crowd-sourcing – such as verification of data – and training teams of humanitarian workers in using new crisis mapping tools and collecting their feedback.