Juliana Rotich of Ushahidi takes to the stage in Abuja to discuss the history of her project and the possibility of crowdsourcing, mobile reporting and rapid responses to elections and violence. Ushahidi was built in the wake of the contentious 2007 elections in Kenya. On December 30, 2007, the Kibaki government issued a ban on live broadcasting. What resulted was a need for people to report news to one another in real time.
Juliana found herself part of a group that was sending SMS and email to track the elections. While it was wonderful to get this information, she and others realized, “if it’s not on the internet, it didn’t happen – there was a need to put it online and let people study it.” So Kenyan geeks, led by David Kobia, built a system that allowed people to report incidents of violence using SMS, email and the web.
The site came to life on January 4, 2008 and had recorded over 200 reports by January 20. She and her team are proud of the fact that the software that runs Ushahidi originated in Kenya, and is now being used around the world. Ushahidi software powers VoteReport India, Cuidemos El Voto in Mexico, and projects in Lebanon and Gaza.
She tells us that mobile phones are the “default device in the developing world”. While Kenya’s got 3% internet penetration – less in rural areas – it’s possible for people to report and access information via mobile phones.
One of the keys to rolling out the platform was Kiwanja’s Frontline SMS, a platform which lets you use a computer and mobile phone to behave as an SMS gateway. Most implementations of Ushahidi involve cooperation with local telephony operators, but Frontline allows you to collect data even if the network operator doesn’t cooperate.
Some of the challenges associated with Ushahidi include:
- Data Poisoning – it’s possible for antagonists to find the tool and use it to report misinformation, or to attempt to incite violence through it
- Verification of data posted is difficult, but can be accomplished if you’ve got highly localized partners
- Because the key data is online, it’s critical to have an online, offline and mobile strategy to call attention to the work
– The project works best with a feedback mechanism – you want people who report to subscribe to SMS alerts or an RSS feed to get updates.
When the system is deployed well, you get rich data that can get dumped from MySQL to do complex mapping. But it’s best not to go directly from reports to the map – you want to hand-check data, even from NGOs you work closely with.
The main message Juliana offers is “don’t start from scratch like we did.” Because Ushahidi is open source and free, you can download it, use PHP to customize it and apply these techniques to anywhere in the world. In the future, custom clients for smartphones will make it even easier to post content to these sites – Ushahidi hopes that you’ll embrace the tools and help her team extend and grow them.
Eric Osiakwan, a man of many talents and almost as many responsibilities, discusses his
African Elections Project. Like Juliana, Eric was in Kenya during the electoral crisis. He found himself wondering, “How do we avoid these situations?” Somewhere between the polling stations where local vote counts were announced and the electoral commission, the vote count changed.
Eric and his friends realized that there’s widespread interest in these elections, and widespread enough use of SMS that SMS could be used to report on and help cover these elections. The African Elections Project is built around a simple idea – let’s take announcements made in local areas and broadcast them widely to help prevent the theft of elections. This involves reporters at polling places, SMS as a transmission mechanism, and the use of broadcast technologies like newspapers and radio to disseminate the information, as well as the internet. Eric is clear that this is a journalistic, reportorial project, not a monitoring one.
Based around this simple idea, AEP has grown to include a much wider set of information around elections. Sites for different country’s elections include pictures from polling places, blog posts, links to news stories and a broad range of information sources about these events.
The main problem with AEP has been a fundamental problem with African elections – the project has been prepared to monitor elections in the Ivory Coast and Guinea, but both elections were put off. So far, the project has monitored Ghana’s elections as well as Malawi’s.
Ghana’s election was close enough that it presented some real challenges for Eric and his colleagues’ team. Their vote count suggested that the opposition was going to win, but the electoral commission hadn’t yet certified the vote. They held off announcing, while one of their media partners went ahead and declared their projections, which angered the sitting government, which lost power.
Eric fields a thorny question about when one reports results – in Nigeria, it’s possible that reporting these results before announcement would be illegal. He’s very clear that the project stays within what’s legal in each country, but tries to collect as much data as possible, for forensic reasons as much as publicity ones.
AEP’s great success may be the replicability of its model – based on AEP training, a team is currently monitoring the Mauritanian elections using the AEP model.
Christiana Charles-Iyoha from the Center for Policy Development would like us to be sensitive to gender issues in mobile phone use. She asks for a show of hands, trying to identify men who carry multiple mobile phones versus women. She’s expecting more men to carry multiple phones – it’s not clear that this is the case, but she pushes on with the point that women use mobiles differently then men.
She references women who aren’t comfortable texting and respond to messages with phone calls. She also mentions that women are doing pioneering work figuring out how to use rechargeable phone cards to transfer money (via a system called “sente” in Uganda) from urban to rural areas. Mobile phones also turn out to be crucial to transmitting agricultural price information, which can be very important to rural women.
Her group is working on a Nigerian human rights hub, which allows people to report on human rights violations via SMS. She’d like to see a toll-free helpline to report gender violence, as well as efforts to build capacity around women and mobiles.