The MiTAP system was developed as an experimental prototype using human language technologies for the monitoring of infectious disease outbreaks. The system provides timely, multi-lingual, global information access to analysts, medical experts and individuals involved in humanitarian assistance and relief work. Each day, thousands of articles from electronic information sources spanning multiple languages are automatically captured, translated, tagged, summarized, and presented to users in a variety of ways. Over the course of the past year and half, MiTAP has become a useful tool for real users to solve real problems. The success of MiTAP is greatly attributed to its user-focused design that accommodates the imperfect component technologies and that allows users to interact with the system in familiar ways. We will discuss the problem, the design process, and the implementation from the perspective of services provided and how these services support system capabilities that satisfy user requirements.