liquid: Context-Aware Distributed Queries (TECHNOTE)
Jeffrey Heer, Alan Newberger, Chris Beckmann, and Jason I. Hong
Group for User Interface Research, Computer Science Division, University of California, Berkeley
problem
monitoring context data is complicated because it's distributed and dynamic
context data
- state of people, places, objects, activities
- logical and physical separations
data stored in infospaces
- nested/linked infospaces
liquid shifts processing work to nodes, which pass updates up as things change, and handles transfer between nodes for continuous queries
- entity type paths: location.building.occupant
- query specification: declarative XML-based language
- query execution: needed more functionality to establish and manage subqueries to different nodes
application
- room awareness: gives name, webpage, email for room occupants as they dynamically change
