User Interface Research for Accessing Scientific Information

Ben Shneiderman
Professor, Department of Computer Science,
Head, Human-Computer Interaction Laboratory 
    at the Center for Automation Research,
Member, Institute for Systems Research
University of Maryland
College Park, Maryland
ben@cs.umd.edu

The next generation of scientific information retrieval systems will
offer novel graphical and direct manipulation user interfaces.
To meet the raised expectations of users of these tools, we have developed
Visual Information Seeking principles (VIS) to support fact-finding and
browsing by novices and experts.  We begin with an overview, and then
allow users to filter out unwanted data, zoom in on relevant items, and
then get details-on-demand.  The interface components are tightly coupled
to give users appropriate feedback, facilitate query refinement, and prevent
errors.

Our contributions include dynamic query methods with continuous visual
presentation of results as the query is changed (using sliders, buttons,
maps), color-coded 2-D space-filling treemaps to show hierarchies in a
single display (hundreds of directories and more than 3000 files can be
seen at once), and the starfield display (a zoom-able 2-D display with
thousands of selectable points of light).  Empirical data from usability
studies and controlled experiments will be presented.