Ubiquitous computing is having a dual impact on the sciences. New instrumentation is gathering ever increasing volumes of data. The increasing scale and decreasing cost of storage media, coupled with the increasing power of the Internet is making it possible for this data to be archived and shared on a massive scale.
But data curation is far from a simple matter. To create systems that ensure data is preserved and findable requires that complex problems of storage, format, and description are solved. In addition, a service layer needs to be developed to help scientists find appropriate data, and to understand how the provenance, instrumentation, and processing of that data influences how it can be used.
This practice issue of Digital Discouse includes some interesting articles on this topic I've found in the course of my work this semester.