Two collaborating Sieve sessions. (Click on image to view at full size.)
In the above example, data sources, filters and visualizations are dynamically connected to create an analysis tool. Sieve components may also be statically combined into complex analysis applets, as seen in this example applet, Composite Block and Plane Simulation.
Sieve is a Java-based collaborative interactive modular visualization environment (CIMVE). Like other modular visualization environments, it provides an extensible set of modules representing sources of data, data manipulation processes, and image generation tools. These modules can then be linked to specify the flow of data to create a ``program'' for producing a visualization.
Sieve extends the capabilities of traditional modular visualization environments by providing support for real-time collaboration. Multiple users may collaborate in real-time to generate data-flow networks and analyze the resulting visualizations. Changes made within the environment are seen by all collaborators.
Collaboration in Sieve is further supported by a miniature workspace overview showing the viewpoints of all collaborators. Remote collaborators' locations are indicated by named telepointers on both the workspace and the miniature overview.