Crisis Crowdsourcing: Harnessing Mass Collaboration to Cope With Emerging Crises

The wave of post-election violence that enveloped Kenya early in 2008 has given rise to a new movement that uses crowdsourcing to provide real-time reporting on unfolding crises., a project spearheaded by a group of African bloggers and software developers, is creating an open-source platform “that allows anyone around the world to set up their own way to gather reports by mobile phone, email and the web — and map them,” according to the organization.

Ushahidi, which means “testimony” in Swahili, began as a web site during the crisis in Kenya. According to Megha Bahree, writing for Forbes (see “Citizen Voices“), Ushahidi began when Ory Okolloh, a Harvard-educated lawyer and blogger living in Johannesburg, South Africa, went home to Kenya in late 2007 to vote in her country’s elections.

When riots and looting erupted in response to alleged corruption in the election process, conventional news media “went black for three days,” writes Bahree. During that time, though, Okolloh continued to blog and to receive reports from around the country from multiple sources, including journalists and government sources.

The stream of information coming to her reached such a volume that she appealed to tech-oriented contacts to assist with converting the stream of reports into a map-based visual form, a mash-up of reports of violence using Google Earth. The following gives an idea of the original mashup presentation on




Speaking in a February 2009 TEDTalk, Erik Hersman, one of the builders of Ushahidi, describes the application as “a system that would allow anyone with a mobile phone to send in information and reports on what was happening around them.” (See “How texting and GoogleMaps helped Kenyans survive crisis“)

Since mobile phones are increasingly available in Africa, Hersman says they make a good “common denominator” as a way for on-the-scene reporters to contribute to an overall view of what is happening during a crisis situation, such as a tumultuous political event, an armed conflict, or a natural disaster.

Hersman says that in the wake of the Kenya crisis, the Ushahidi group decided they needed to do more: “We needed to take what we had built and create a platform out of it so it could be used elsewhere in the world.”

Since then, the platform has been used by Al-Jazeera in Gaza, in the Democratic Republic of Congo, in South Africa to “map xenophobic attacks perpetrated against non-South Africans,” according to Ushahidi, and is now being used in India to monitor that country’s general elections.

However, the “next big thing,” Hersman believes, has to do with coping with information overload and evaluating the accuracy of reporting during crises:

What we’re finding out is that we have this capacity to report eyewitness accounts of what’s going on in real time — and we’re seeing this in events like in Mumbai recently — where it’s so much easier to report now than it is to consume [the reporting].

There’s so much information, what do you do? … How do you decide what is important? What’s the veracity level of what you’re looking at? … we find that there’s this great deal of wasted crisis information. Because there’s just too much information for us to actually do anything with right now.

During the initial hours when a crisis breaks out, says Hersman, a great deal of information can be streaming out of the crisis area — in the form of mobile messages, blogs, web postings, emails,or  Twitter messages. But the world outside the crisis zone, and indeed the citizens inside it, have no way to aggregate and process the extreme volume of reporting, as well as to evaluate the accuracy and truthfulness of it.

Hersman’s group is now working on a filtering process and technology that will “take the crowd and apply them to the information,” using peer ratings to evaluate, refine, and weight the reliability of data coming out of a crisis zone — “so that we have a better understanding of the probability of something being true or not.”

One key application of the Ushahidi crowdsourcing platform is to help relief organizations assess the situation in a crisis zone in real time and direct aid to the area as quickly as possible. “The idea is to get immediate attention and relief to strife zones, and fill the gap left by news organizations that have slashed their foreign bureaus,” writes Bahree in her Forbes article. She relates one experience of how this worked during the Kenya crisis:

A ranger in Bogoria (northeast of Nairobi), William Kimosop, was driving to check on a remote outpost one evening in January when he stumbled across several hundred women, old people and children, lost in a ravine as they fled their villages where the men were still fighting. Four babies were born in that ravine, and supplies were running out.

There were no government officials or police around. He sent a text message to a friend in Nairobi, asking her to get help from aid agencies. The friend forwarded his plea to a few people, after which it got picked up by Ushahidi and within six days a Red Cross truck reached him.

AB — 22 April 2009


Open-Source Crime-Solving

Porting this post over from Socialtext:

I was struck by this article on the BBC:

Amateur sleuths keep cold cases alive

It struck me that this is another application of the open-source model — Internet technology can possibly be used in crime investigations to bring many dedicated minds to bear on problems that can’t be solved by the efforts of a few professional investigators.

AB — originally posted 23 April 2007

Crimefighting by Crowdsourcing

Here is an investigation by Dutch police of a 1995 cold case in which they are inviting outside participation, with a reward offered:

Also, here is the website for Victoria (Australia) Crime Stoppers, inviting assistance with solving crimes and finding fugitives:

Here is an interesting article on CNN about how the Internet was used to solve an old mystery:

Amateur sleuths keep cold cases alive 

AB — originally posted 15 May 2007