This post is based on my notes during a short presentation by the Ushahidi folks. It was one of the projects that caught my attention--it seems to me to have far reaching implications not only in regions that are experiencing unrest, but in any situation where crowdsourcing might have value.

Ushahidi is a system built to group source data having to do with crisis/uprisings where information might be spotty. It isn't intended as a tool to pursue conviction of criminals--data can be poisoned--but to provide a picture of what is going on in crisis situation. The first version of the system was built in 3-4 days. The idea is that you send "short code" data from your mobile phone to the server. That allows the server to aggregate and make the data meaningful. Currently this data is only one way-that is to say there is no feedback loop to the user who generates the content. The intent in version 2 is to provide some kind of feedback.

In a real way, this system could create a community of people to communicate ideas--how each of these groups can use each other's data.

The challenges that the team experience in Kenya and South Africa included the fact that there isn't a simple tool to crowdsource information. This is exacerbated by how you get enough data in a 160 character listing that is meaningful? How do you provide credit to those who have shared information? How can you report the data over a time period? If you are collecting data on a certain day, you will want to be able to show data over a period of time.

Version 2 will be deployable by anybody. There will be a hosted and a downloadable version.

Interestingly, the team has incorporated with the intent of becoming a nonprofit to provide an easier way to fundraise to create version 2 of the software.