Juliana Rotich is the Program Director for a website called Ushahidi.com. “Ushahidi” is the Kiswahili word for “testimony.” This site collects “testimony” on violence as people see it. They can send reports using cell phones and computers to the site. The information is logged and registered according to the type of violence (riots, deaths, property loss, rape, looting, etc.) using Google Maps. In places and moments when reporting is dangerous and difficult, this kind of program uses information from ordinary citizens to record violence as it’s happening. Ushahidi was put together by a lawyer/activist, Ory Okollo and a small group of Kenyan, blogger/techies while the post-election violence in Kenya was going on in the early part of 2008. It was then used in South Africa to track xenophobic attacks against foreigners, and then in the Democratic Republic of Congo. All these instances can be viewed on the website. Al Jazeera has also used the platform to track violence in the war on Gaza (Jan, 2009), and many others are just beginning to use it.

In addition to offering a method of tracking violence, Ushahidi also serves as a bloggers’ space to talk about new uses of technologies in Africa. Videos produced on the spot that can be uploaded also add to the testimony. It is cutting edge technology married to a citizen empowerment project. The collected data is useful for all kinds of people and purposes from NGOs, media, human rights workers, aid organizations to ordinary citizens who want to know what is happening in times of crisis.

Ms. Rotich has a background in database management, is a blogger for Ushahidi and other sites, and is an environmental contributor to www.globalvoicesonline.org. The bloggers on the site are generally known by their first names and write about technical and content oriented issues.

Can you tell us a little about how the Ushahidi project got its start?
Ushahidi began as a collaborative project after the post-election violence in Kenya. This was in January of 2008. The elections happened on the 27th of December, 2007. There was a contested election in Kenya. Subsequent violence erupted around the 30th and it continued for most of January until Kofi Annan came and set up a coalition government.

How did the project get started?
Ory Okollo initiated it with a blog post. She said, “how about using Google Maps to indicate where the incidents [of violence] were happening?” At that time the media coverage was spotty and we weren’t sure which incidents were actually happening. I think they were being underreported. So, she wrote about that in her blog, and Erik [Hersman] and David [Kobia] read it and put together a site that did just that. That’s how it started. Daudi [another founding blogger] and other bloggers entered information and also publicized the existence of the site.

Did you already have a group of contacts willing to be witnesses or was it just ordinary people who began sending information?
We use this [the site] as an aggregator. Ory is a lawyer and had a lot of experience working with the Commission on Human Rights, so she had some contacts from whom she was getting information. Daudi also had some contacts with the NGOs, but you have to think of it as a collaborative effort—this wasn’t an organization that existed. This was just a project that we did together, a sort of ad hoc, collaborative effort.

Can you talk about the subsequent projects in South Africa and the Congo?
So the project in South Africa started when there were xenophobic attacks there—Rafik [another blogger] contacted Ushahidi and we set up an instance. You can see it at www.unitedforafrica.co.za. Basically, they used our platform to gather all the information regarding the xenophobic attacks. The Democratic Republic of Congo—this was an instance that was an initiative of Ushahidi late last year. The intent is actually to also continue the idea of crowdsourcing information, particularly to try to get information from areas where the mainstream media does not cover what is happening.

Can you briefly explain crowdsourcing?
Crowdsourcing is getting information from the public by cell phone, e-mail or people coming to the website and filling out a form with a title, time, and description of whatever they saw. The idea is sort of like Wikepedia, a collaboration in which the platform allows for somebody to add to an existing report. So let’s say you’re at a certain location and you see a report about that location, but you see something more. You can click on that incident report that is already there and add more information as a comment.

How easy is this system to set up?
We are currently in Alpha testing. There will be a Beta version that will be available for others to download. It’s a small organization so we can’t deploy every time. We have a finite amount of resources. It does require a technical person at the other end—basically a PHP developer or somebody who knows something about databases needs to be there in order to be able to deploy this. It does require some technical know-how. It’s not simple. But it is simple to use. If you are a person who is submitting a report, then it’s pretty easy.

Do you know who is uploading information—Individuals or NGOs?
It’s individuals, although the individuals may be affiliated with NGOs. Really the crowdsourcing part of it is quite instructive. It’s really the regular people. Although it is an optional field. You can put in your name or your affiliation, or not. From the information we have, it’s the general public.

How do you verify information?
When we first began and it’s still the case right now, we do not purport to verify. If something is unverified, when you click on the report it will clearly say unverified—you’ll see an “X” mark or if it’s verified, you’ll see a “√” mark. Specifically, you’ll see a “√” mark if there’s a link associated with that report and another person’s report confirming the first one. When the project began, with Ory and Daudi trying to confirm what they were getting, it was a very fluid situation. So some information that you got, until you confirmed it from someone else, it remained unverified. When you got a piece of information, you would see if you could get any corroboration. If not, then you would contact somebody, so it was very manual and very tedious. But, we felt it was important to at least have that information available. Then, when you get more information, you can kind of see what is happening around that particular area. But you can clearly see on each incident whether it’s verified or unverified information.

Do you know who uses the information?
The project has now progressed onto an opensourced level, such that our next version, a Beta version, will be available for people who are interested in crowdsourcing information. We realize that the tool can be used by NGOs. It can be used by even media organizations. You may have seen that Al Jazeera is using it in Gaza. Ideally, yes, it could also be used by aid workers. We’re involved in creating the tool. How the tool is used in the field is open. It would be great if it was used for humanitarian reasons, but we’re not limiting it to that. With opensource projects like this, we want to make it available for as wide a range of uses as is possible. Because if you find a use for it, you download it, and you‘re able to run it, you should be able to use it in whatever way it can help. We’d like for NGOs to be able to have information that they can go on when something happens, so that they can actually have a leg up in terms of what’s happening, and where. If they’re the type of NGO that has a mandate to alleviate a problem, they can have some information to begin with when they go out in the field.

Can you talk about the your new BETA version?
It’s the software that runs Ushahidi.com. We’ve added some features and more functionality. You have a dashboard and you have a way to set up to certify that this is the geographical area of mapping you would like to focus on. It also has Twitter integration, meaning that you can gather reports using Twitter. If you look at www.foko.ushahidi.com, you’ll see some visualization or data clustering where you’ll see smaller bubbles showing fewer incidents and bigger bubbles showing more incidents. So that type of data clustering will be new and added into that version. It should be available later this year and we look forward to providing it to the greater human rights community.

Are all of you at Ushahidi technical people?
Three of us are technical. Ory is a lawyer and activist. We’re all bloggers—that’s sort of how we met. Erik is a web strategist, David is a high level programmer and I have experience as a data analyst and in database administration. There is technical bent and a bit of new media background. Also, we are all four Kenyan. Erik is American but he grew up in Kenya.

We’re really proud of the fact that this is a project that came out of Africa, Most people do not expect that. It’s become a global project now because these days, you may be African but you don’t live in Africa. Still, you can contribute. It started out as an African project—the majority of our programmers are still African— but now it is really an opensource global project. It’s grown into that and we’re really glad to be able to work with so many talented and amazing programmers from around the world.

Past Close Up Articles

Christina Iturralde, LatinoJustice PRLDEF
Anene Ejikeme Assistant Professor, Trinity University
Alexander Verkhovsky, Director, SOVA Center for Information and Analysis (Moscow)
Innokenty Grekov, Human Rights First
Paul St. Clair, Executive Director, Roma Community Centre
Rev. Allan Ramirez, Latino Advocate
Dr. Agapito López, M.D.
Bill Habern, defense attorney
Anya Cordell, activist
Rais Bhuiyan, survivor
Allison Moore, Volunteer