Violence 2.0: some lessons from Ushahidi
Because Paul’s claiming to be too confused to write up some lessons learned from Ushahidi.com, I’ll have a go:
What’s cool?
- It’s timely: the number of people who actually get these kind of things off the ground, as opposed to jibber-jabber about them, is very small. Getting preliminary, lead information as close as possible to the time the incident happened is extremely valuable in every possible scenario and not just the “document now, prosecute later” one.
- It challenges the conventional view by providing an opportunity for people to read first hand accounts un-editorialised by the MSM, who seem intent on warping the events into the template of Generic Violence in Africa.
- It’s quite easy to use: the system focussed on getting raw information in the form of sit-rep narratives online without overburdening the person submitting the information.
- It links in with current monitoring processes: although I’ve noticed a few irrelevant submissions up there, it seems they’re making an effort to ask NGOS on the ground to verify the basic thrust of an incident. I am sure that the information from Ushahidi will find its way into other, formal efforts to document what is going on.
What could be cooler?
- Clarify the purpose of the system: Paul raised this in his earlier post - what exactly is this system for? Some of the published material is very general situation narrative unrelated to specific incidence of violence. Some of the incidents are also based on news reports from the international media, but this is a system that aspires to give the raw groundview and not information that’s been twice around the world first.
- Get just a little more structured information from people submitting reports: With all that narrative, I wouldn’t like to be the analyst for the raw info Ushahidi holds at this point. Whilst the whole “Who Did What To Whom” model for documenting violent acts might frighten the general public from actually using the online submission system, separating out the recording of information about the nature of the incident (deaths, theft, destruction of property) from the perpetrators and victims would be a step enabling a useful dimension of analysis. It’s not exactly clear what “names of the involved” really means. They could usefully take a browse through the HURIDOCS Events Standard Formats and Microthesauri for some inspiration.
- More transparent verification could increase its credibility: A clearer indication of how incidents are verified, and who is doing the verification, preferably with some kind of attribution. At present, a cross or a tick next to the incident isn’t going to satisfy anyone that the incident wasn’t fabricated.
- Build follow up into the system: you want people to continue using the site as an information resource. What will prompt them to return is features will give them updates: perhaps a daily email or sms digest. Not sure quite how this could work, but there you go.
Where’s everyone else at with this?
![[del.icio.us]](http://images.del.icio.us/static/img/delicious.small.gif){kind=small}
I notice that the strengths that you identify are mainly to do with the service itself (functions), while the weaknesses mainly address questions of where that service is located (processes). Maybe that sums up my concerns with a lot of the Web2.0 hype - it’s so easy to set things up that people don’t need to think about how that set-up is going to become sustainable. Now there’s a strong argument for “just doing it” and seeing what sticks - but on the other hand, that might have the cost of people getting disillusioned if they see the same thing being repeated every time.
Paul Currion
8 Feb 08 at 15:16
I guess this approach to monitoring is still a novelty, and no truly superior method or technical system, or even consensus about role, has emerged from past efforts. By way of comparison with the outsider status of Ushahidi.com, have a look at the website of the Liberia Truth and Reconciliation Commission, whose hearings are currently ongoing. Anyone with an opinion on the conflict in Liberia can submit a statement to the Liberia TRC through the website. I have no idea how popular or successful this has been; an indicator of this might be that the accompanying discussion forums have had only six posts since coming online in November 2007.
Tom Longley
8 Feb 08 at 18:24
Tom/Paul, we’re listening, thanks for your feedback on Ushahidi.
We didn’t at first realize what we were truly getting into when we cranked out Ushahidi over a weekend. I guess, if there’s one thing we can take away to this point it’s that Ushahidi is a starting point for a different model of online crisis response by citizens. While we continue to try and improve it, both technologically and with the processes that take real people on the ground, we know that there is more that needs to be done.
For instance, two of our biggest hurdles come from closed NGOs who don’t want to share information and from finances to market it in Kenya. So, it’s a prototype that gives some indication of how a web/mobile-based tool can be used in crisis situations, but lacks some structure for growth and increased credibility.
The team is in Istanbul right now discussing what we need to do to upgrade the system and make it better. Your points are being discussed.
Erik Hersman
8 Feb 08 at 20:02
To be fair to the NGO community, I can understand their reticence - partly due from simple misunderstanding of what you’re trying to do, and partly from security concerns of their own. I think the key is to embed this sort of tool into processes - at the moment, it feels to me as if the tool is the process, if you see what I mean. Good luck meeting in Istanbul!
Paul Currion
11 Feb 08 at 17:40
[...] post-election violence in Kenya. I’ve reviewed at Ushahidi a while ago on this blog and so have others, including this story by GlobalVoices Online that details the evolution of her idea and [...]
Interview with Ory Okolloh on post-election violence in Kenya, cyber-activism and Ushahidi « ICT for Peacebuilding (ICT4Peace)
17 Mar 08 at 14:47