Haiti Earthquake Information (HEI): connecting people in need with the information and resources they need.
This project was started on January 19, one week after the earthquake. The original motivation for this Drupal-based web informatics project was a number of reports from doctors and reporters in Haiti during the first week after the January 12 earthquake. This video by Dr. Sanjay Gupta describes the major communication / coordination problems encountered in getting medical supplies already at the Port-au-Prince airport to doctors with critically-injured patients just a few miles away in PAP.
This HEI project was initiated by a group of software developers, and was built on the Drupal Content Management System. Drupal is a very powerful, open source CMS that is widely used for rapidly developing complex, web-based information processing and database storage systems. The HEI web tools that we developed were deliberately designed to provide data integration functionality between the different information "silos" that typically occur during any crisis. In particular, they provide "data mashup" tools for integrating the many requests for aid with the many resources available for meeting those requests.
* HEI Software tools: By the end of the first week (Jan 23), we had a good beta web app that was designed to integrate specific Twitter search feeds with other data sources, including the report system from Ushahidi, another crisis-response web software project. For example, we used the evolving Project EPIC hashtags in Twitter searches using either #haiti #need, or #haiti #have, or #haiti #offer, in order find tweets about large groups of people in need.
Here is screenshot of an early version of our Twitter search feed app, using the Project EPIC hashtags, which was built on the Drupal-based ManagingNews system:
Managing News with Twitter search feeds: http://www.mcaudy.net/haiti/ManagingNews-HEI_haiti-need_feed.png
Here are screenshots of other versions of our tools for rapidly analyzing large numbers of Twitters search results, including a #haiti #need / #haiti #have data mashup, for matching people with specific needs, with people who can address those needs:
(best viewed in Firefox):
#need #have data mashup: http://www.mcaudy.net/haiti/need-have-mashup_screenshot.png
#need only search with casetracker: http://www.mcaudy.net/haiti/need-casetracker_screenshot.png
where the Casetracker module was used to provide added functionality not provided by the Ushahidi reports system.
The CaseTracker feature is specifically designed to allow individual cases to be opened and tracked from start to finish, so that definite resolution of the reported need is confirmed and documented. This also provides accountability in that the resources obtained are known to have been delivered to their intended recipient.
Using this third app, we were able to search and find all of the nearly 3000 tweets containing #haiti #need tags with pleas for aid, which were stored in this Drupal app to create a custom knowledge base that could be later mined for data revealing the most urgent need cases, for camps with very large numbers of people. This app used very effectively for the "PAP Outskirts Food Relief project", below)
* Collaboration with Ushahidi and CrisisCommons: we kept a close watch on developments by other groups, to avoid duplication, and to build collaborations. In late January, we started collaborating with Ushahidi, using our apps to provided separate case-tracking functionality to the Ushahidi report system. This collaboration is still ongoing.
* PAP Outskirts Food Relief project (a formal CrisisCommons project; still ongoing): This project was specifically designed to address a critical problem that emerged in Haiti about 3 - 4 weeks after the earthquake. Due to the aftershocks, most people had fled PAP and were living in large camps on the outskirts of the city. Many or most of these camps were unknown to the UN / WFP, and therefore had received essentially no effective food relief from the WFP. At this time, the first documented cases of death by starvation were being reported.
The project's goals and methods were, and still are:
1) to identify large groups of people on the outskirts of Port-au-Prince (PAP) who had been reported to have critical food needs. Camps typically have been initially identified by several techniques, including by: i) Twitter messages that described the camp in need, or ii) by taking already created Ushahidi reports for that camp, as cases to be "verified", or iii) more recently, several camps have been identified by our local Haitian agent working on the ground, Marilou Roy. Marilou lives in Petion-Ville, which is on the outskirts of PAP, and she knows this area extremely well, as well as other areas of Haiti. During recent months, she has taken aid to groups in Jacmel, Leogane, and several other cities and areas in Haiti. She also has now recruited a group of five local Haitians as coworkers, who help her with the food distributions.
2) to subsequently "verify" that those food needs were still ongoing. Direct verification of ongoing need was either by: i) subsequent, independent cell phone contact with someone in the came, to verify that there is still critical need, or ii) by having our agent, Marilou, directly visit the camp to verify need.
3) once verified, to report the verified needs cases to the WFP, so that they could review our reports, and arrange for emergency food delivery.
Here is the CrisisCommons page for this project (along with a description of the project at time):
http://wiki.crisiscommons.org/wiki/PAP_Outskirts_Food_Relief
This project was carefully designed to meet a major food relief problem, and the Word Food Programme was initially extremely responsive and cooperative. When I submitted our first verified case of urgent food need for a large camp (Villa Manrese, 1500+ people), the WFP responded immediately to my e-mail, and forwarded it to the Irish-based NGO, GOAL, who had been assigned to handle large-scale food delivery in that region. GOAL also responded very quickly, and at least some food was delivered to that camp within two days.
However, subsequently the WFP has refused even to acknowledge receipt of three additional verified cases of camps with large numbers of starving people (totaling 1900), even though several children had already died of malnutrition in at least two of those camps, as described in our reports. Since then, I have learned that there is a "triage" system, where camps must be sufficiently large in order for the WFP to direct the corresponding NGO to deliver food to them.
We have still been able to get some food - specifically nutrients for children - to camps that Marilou has identified and verified as having critical need, but this has depended on private donations, whereas this aid should be provided by the WFP.
We currently are seeking private donations and alternative funding sources to pay for the food we deliver in these urgent need cases?
* Future Plans and Requests for Advice about Possible Funding Support: Here are a few of the ideas and plans that we have for future aid effort in Haiti.
Integrated SMS / laptop system: to facilitate communications between field workers and distant collaborators - such as Marilou Roy, while she is away from PAP, with no internet connection. I have read about your new slingshotSMS system, and I also am aware of the MedicalSMS and RapidSMS systems developed by others. However, I have not been able to determine which of these would be bested suited for us (or others) to use by ther field workers in the Haiti aid effort. I would like to discuss this with you.
Grant Applications for Funds: to allow continuation of the aid work (food, tents, etc., targeted for children). Can you recommend some specific funding agencies to apply for immediate, short term funds to keep our current efforts going?
Salary Support for Marilou Roy's Aid Work: In particular, can you suggest where I could find at least partial salary support for Marilou Roy? Marilou has been extremely generous in giving her time as a volunteer without pay to aid others since the earthquake. However, she cannot continure indefinitely working for free. She is a remarkable woman, and with some salary support she would be willing and able to continue her important work to identify new urgent need cases, and to bring aid to those people.