Aggregated Information, Decentralized Action: New Roles for ICTs in Disaster Response
Abstract
Disaster response can be highly coordinated without central control. Or can it?
Following last winter"s Tsunami and this fall"s hurricanes, many new Internet-based services emerged. They performed many functions, including locating missing people, matching people with housing, and distributing material aid. A common feature of these new services is that they collected, aggregated, and disseminated information to support coordinated action among people and organizations who had no direct authority relations with each other. We call this the AIDA approach:
Aggregated Information in support of Decentralized Action.
The AIDA services were not always successful. Some were not available in time to be useful. Some just weren"t useful. Some went unnoticed. Some did not provide sufficient information, or information of sufficient quality.
In addition to these internal limitations, however, there was a mismatch in style and approach beteween these new AIDA services and a more centralized command and control approach. For locating missing people, the conventional approach has been for relief agencies to maintain a database that is accessible only to authorized personnel. For housing, large shelters, large-scale rental housing and new FEMAvilles seem to be the preferred approach, presumably because such programs are far easier to administer than relying on networks of churches and individuals who have offered to help. In the wake of Hurricane Katrina, a consistent message went out to those who would offer medical or material aid: don"t self-deploy; don"t act on your own, lest you get in the way of the organized response. Please offer your goods or services, but please wait for an official to contact you about what action to take.
To begin to understand the strengths and limitations of the AIDA approach, we have developed a series of case studies. In addition to profiling particular services, we have attempted to distill the key issues and design options in each service area, and to point to opportunities for future development.
This is a wiki and we invite anyone to contribute to the ongoing development. (If you haven"t registered, try
TWikiRegistration.)
Purpose and Intended Audience
This document is intended for two audiences. First, for people who have been involved in developing particular AIDA-based services, it is provides some language for talking and thinking about these services, a chance to think through the advantages and disadvantages of alternative designs, and some suggestions for future directions of development. Second, for people who are involved in more conventional disaster planning and recovery organizations, and those who make policies about it, this document may help to open a dialogue about the benefits and limitations of decentralized action in disaster response.
Comparative Analyses
Comparative analysis documents:
Cross-cutting Findings
We found several common themes across the various areas in which Aggregated Information, Decentralized Action approaches have been tried.
- The need for users to become aware of a site creates a big advantage for first movers and for established entities that people already look to. Normally, this might lead to a winner-take-all situation where one site becomes the focal point for information sharing. A very compressed timeline, however, changes that dynamic-- people enter and look for information in whatever locations they are aware of, and they are not all aware of the same locations.
- Because of the multiple focal points for information collection, aggregating information from multiple sources is common. This has many implications, including the value of data standards, and the need for mechanisms for de-duplicating data and updating and correcting it.
- The trustworthiness of information and people cannot be assumed if there is no central authority standing behind them. Restricting read or write access to data can create trust and protect people, but also can limit the comprehensiveness and usefulness of an information source. Credentialing and other trust-generating mechanisms may also be useful components of AIDA approaches.
- Services vary in the level of social aggregation at which they gather information, both for supply and demand. Services that provide information about needs or offers at the individual level have very different implications for privacy, dignity, motivation, and efficiency than those that aggregate needs or offers at a larger scale, such as a community, a school, or a church.
Analyses of individual services
If you would like to add a new analysis, please copy our
AnalysisTemplate.
About the Authors
This document has been prepared by students and faculty in a course on Rapid Response Informatics at the University of Michigan School of Information, led by Professor
Paul Resnick. Below is a list of contributors to this document (please add yourself to the bottom of the list if you make significant edits).
- Paul Resnick
- Adrienne Janney
- Brian Kerr
- Andrew McLetchie
- Rachel Pooley
- Matthew Scholl
- Aiko Takazawa
- James Sweeney
- Seth Turner
- Marcia Wallin
- Prashanth Nooguri
- Rohit Laungani
- Narayan Kansal
- ...
Recently updated pages in this wiki
→ See
WebChanges for a more complete listing of changes to this site.
to top