ABSTRACT
This chapter provides an overview of the problem through a series of example application scenarios. It describes a vision for the future, where data at any scale are universally accessible and manipulable in a common environment, and the software assists in focusing user attention. Data acquired from sensors, experiments, or simulations or identified in external databases must be transformed into the right context to make discoveries. The goal of such systems should be to support transformation whenever possible rather than just search and retrieval. The chapter described is especially acute in the long tail of science: the small labs and individual researchers who collectively produce the majority of scientific output. The cost to computer scientists in engaging with long tail scientists on a lab-by-lab basis is prohibitive. Instead, there is a need to understand and attack only the cross-cutting requirements, while also considering deployment models that can reach those with a limited technical background.
