AlertNet provides a source of information for its member agencies but not all of the news will be relevant to all members and some may prefer to have stories relating to certain topics delivered as a matter of urgency. The AlertNet stories could be processed using NLP techniques to identify the stories relevant to each agency. This approach is particularly relevant to aid agencies requiring information for rapid reaction. Other agencies will be more interested in gathering information that is not so time dependent. To serve these alternative requirements, a possible extension to Mark's project would be to apply the technology developed to identify documents of interest to particular aid agencies to web search engines.